University of Groningen

Today's talented youth field hockey players, the stars of tomorrow?Elferink-Gemser, Marije Titia

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2005

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Elferink-Gemser, M. T. (2005). Today's talented youth field hockey players, the stars of tomorrow? a study.s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 10-10-2020

TODAY’S TALENTED YOUTH FIELD HOCKEY PLAYERS, THE STARS OF TOMORROW? A STUDY ON TALENT DEVELOPMENT IN FIELD HOCKEY

Groningen Studies in Sports Sciences 1 The research described in this thesis was financially supported by NOC*NSF

Paranimfen: Jildou Gemser Pieter-Jorn Gemser Cover: Corina Blom Printed by: Grafisch bedrijf Ponsen & Looijen bv, Wageningen ISBN 90-6464-2923 © Copyright 2005: M.T. Elferink-Gemser, Groningen, the Netherlands. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording or any information storage or retrieval system, without the prior written permission of the copyright owner.

RIJKSUNIVERSITEIT GRONINGEN

Today’s talented youth field hockey players, the stars of tomorrow?

A study on talent development in field hockey

Proefschrift

ter verkrijging van het doctoraat in de Psychologische, Pedagogische en Sociologische Wetenschappen

aan de Rijksuniversiteit Groningen op gezag van de

Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

donderdag 14 april 2005 om 14.45 uur

door

Marije Titia Elferink-Gemser geboren op 7 augustus 1973

te Sneek

Promotor: Prof.dr. Th. Mulder Copromotores: Dr. C. Visscher Dr. K.A.P.M. Lemmink Beoordelingscommissie: Prof.dr. H. Kuipers Prof.dr. R. Bosker Prof.dr. H. Kemper

Voor oma

Contents Chapter 1 General Introduction

1

Chapter 2 Evaluation of the reliability of two field hockey specific sprint and dribble tests in young field hockey players. Lemmink, K.A.P.M., Elferink-Gemser, M.T., and Visscher, C. (2004). British Journal of Sports Medicine, 38, 138-142.

11

Chapter 3 Relation between multidimensional performance characteristics and level of performance in talented youth field hockey players. Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. (2004). Journal of Sports Sciences, 22, 1053-1063.

27

Chapter 4 Multidimensional performance characteristics and performance level in talented youth field hockey players: A longitudinal study. Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. Journal of Sports Sciences (pending minor revisions).

49

Chapter 5 Development of the interval endurance capacity in elite and sub-elite youth field hockey players. Elferink-Gemser, M.T., Visscher, C., Van Duijn, M.A.J., and Lemmink, K.A.P.M.

69

Chapter 6 Psychological characteristics of talented youth athletes in field hockey, basketball, volleyball, speed skating, and swimming. Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. The Sport Psychologist (in revision).

87

Chapter 7 Development of the Tactical Skills Inventory for Sports. Elferink-Gemser, M.T., Visscher, C., Richart, H., and Lemmink, K.A.P.M. (2004). Perceptual and Motor Skills, 99, 883-895.

103

Chapter 8 Discussion and Conclusions 121 Summary 133 Samenvatting 137 List of publications 143 Dankwoord 147

Chapter I

General Introduction

2

Chapter I II III IV V VI VII VIII

3

1.1 Expert performance in sports

The ambition of the Dutch Olympic Committee (NOC*NSF) is a permanent top-10 position in the world of sports. To realize this, structural attention is paid to the identification of talented athletes and their development towards expertise. The present study on talented youth field hockey players has been conducted at the Center for Human Movement Sciences of the University of Groningen and is one of many projects to fulfil the above mentioned objective.

Expert performance in sports can be defined as the consistent superior athletic performance over an extended period (Starkes, 1993a). In the present study, expert performance in field hockey is operationalised by playing in the highest league of the Dutch field hockey competition. Dutch field hockey has been ranked among the best in the world for decades and its competition is recognized world-wide for its high performance level. Although reaching excellence in field hockey is not linearly related to the number of invested hours of practice, current international-level performers have spent many hours of training for at least ten years before reaching the top (Ericsson et al., 1993; Ericsson, 1996; Starkes et al., 1996; Starkes, 2000; Van Rossum, 2000). All of them invested enormous amounts of time preparing for the international sporting arena. In the Netherlands, most experts started playing field hockey when they were seven years old. Obviously, youth players who want to make it to the top have to start training already at an early age.

In a review on talent research, Williams and Reilly (2000a) make clear that from a scientific perspective the pursuit of excellence can be broken down into four key stages: ‘talent detection’, ‘talent identification’, ‘talent development’, and ‘talent selection’ (Russell, 1989; Borms, 1996). Talent detection refers to the discovery of potential performers who are currently not involved in the sport in question (Williams and Reilly, 2000a). Talent identification refers to the process of recognizing youth players with the potential to become elite players whereas talent development implies that these players are provided with a suitable learning environment and resources so that they have the opportunity to realize their potential (Régnier et al., 1993). Finally, talent selection involves the ongoing process of identifying players at various stages who demonstrate prerequisite levels of performance for inclusion in a selection team (Williams and Reilly, 2000a). The present thesis focuses on talented youth field hockey players: players who perform better than their peers during training and competition, and who have the potential to become elite performers in the future (Howe et al., 1998; Helsen et al., 2000). This means that the current performance level of youth players is considered important as well as their potential for the future. They are part of a talent development program of a field hockey club of national prestige, and are playing at the highest level for their age category.

4

1.2 Profile of field hockey

Field hockey is a field invasive sport in which players compete at the same field of action as their opponents (Hughes and Barlett, 2002). To obtain expert status in field hockey, players must excel in no less than four domains: physiological, technical, tactical, and psychological. In addition, the development of their anthropometric characteristics is important. Match analyses at the elite level make clear that field hockey is a high intensity non-continuous game in which the physiological demands are considerable, placing it in the category of ‘heavy exercise’ (e.g., Ghosh et al., 1991; Reilly and Borrie, 1992). Physiological components of expertise include aerobic and anaerobic capacity (Wilmore and Costill, 1999). Specific for field hockey is the intermittent running, e.g. the alternation of accelerating and decelerating, and the many changes of direction while sprinting (Patel et al., 2002; Spencer et al., 2004). The unique requirements of field hockey including dribbling the ball and moving quickly in a semi-crouched posture superimpose the work-load demanded by the game (Reilly and Seaton, 1990). Technical expertise refers to the degree of sensorimotor coordination from which refined, efficient, and effective movement patterns emerge (Janelle and Hillman, 2003). For a technically sound player, dribbling is essentially an automatic process, and the better players distinguish themselves by their running speed while dribbling the ball (Reilly and Bretherton, 1986).

Field hockey is a highly structured analytical game in which players constantly have to deal with a complex and rapidly changing environment (Starkes, 1993b). In order to be successful, they have to perform the right action at the right moment. Therefore, they have to acquire great tactical skills. Tactical expertise is a requisite for expert performance in virtually all achievement domains (Janelle and Hillman, 2003). Sport is unique, however, in that tactical skills involve not only the knowledge to determine what strategy is most appropriate in a given situation, but also whether the strategy can be successfully executed within the constraints of the required movements (e.g., Starkes, 1993a; McPherson, 1994). Thus, the execution of tactical skills in field hockey is always related to the physiological and technical limitations of the individual player, his or her teammates and his or her opponents. To perform at top level, players have to perform under high pressure. It is therefore not surprising that psychological characteristics such as motivation, confidence, anxiety control, mental preparation, team emphasis, and concentration often distinguish elite from non-elite performers (Mahoney et al., 1987; Morris, 2000). Excellent psychological skills can not only play a decisive role in an important match; they are also needed to develop a successful sports career. Commitment from the performers is required since engagement in training is not inherently motivating (Ericsson et al., 1993; Ericsson, 1996).

Chapter I II III IV V VI VII VIII

5

1.3 Research in sport expertise

For several years, researchers have tried to identify key predictors of talent in many sports. A decade ago, Régnier and collegues (1993) published a review on talent detection and development in sports with the purpose of providing a better understanding of the process by which one achieves greatness in sports. Until that time, far most studies were cross-sectional in nature measuring general characteristics. Literature on talent identification and development has largely emerged during the 1990s. Books that contribute substantially to our basic understanding of expertise are ‘The road to excellence: The acquisition of expert performance in the arts, sciences, sports, and games’ by Ericsson (1996) and ‘Expert performance in sports’ edited by Starkes and Ericsson (2003). Some years ago, the Journal of Sports Sciences devoted a special issue to talent identification and development in soccer (Williams and Reilly, 2000b). Research with athletes at the highest level of performance Sport is characterized by a hierarchical organization in which the level of performance of a player is described by the appropriate level of competition (e.g., local, regional, national, and international). The number of players that are allowed to compete at a given level of competition becomes smaller as the level of performance increases. When players of different competition levels are compared on the basis of their performance characteristics it is to be expected that the higher level players outscore the lower level players. However, this does not necessarily apply when players within the same competition level; i.e., within a talent-group are compared with each other. The relation between multidimensional performance characteristics and level of performance might be different. Therefore, to unravel the mechanisms leading to excellence, research should be conducted within a group of talented players, all playing at the highest performance level for their age. This is possible by comparing elite youth players with sub-elite youth players. Both elite and sub-elite players are part of a talent development program of a field hockey club of national prestige, and are playing at the highest level for their age category. However, in contrast to sub-elite players, elite players additionally play in a youth selection team of the Dutch Field Hockey Association (KNHB). Measuring multidimensional performance characteristics in a sports-specific way A group of all talented players is relatively homogeneous with regard to their performance level. As a consequence, measures of general performance characteristics are usually not sensitive enough to detect differences between elite and sub-elite players (Bangsbo and Lindquist, 1992; Lemmink et al., in press 2004). Tests therefore have to measure components

6

that represent the specific demands of the sport in question involving sports-specific variables (Atkinson and Nevill, 2001). Sports scientists often acknowledge that a world-class performance is the result of several factors, advocating a multidimensional approach in studies on talented players (e.g., Régnier et al., 1993; Reilly et al., 2000). Burwitz et al. (1994) also recommend interdisciplinary performance-related sports science research. Therefore, to allow for the characteristics of field hockey, anthropometric, physiological, technical, tactical, and psychological characteristics should be measured in a sports-specific way. Longitudinal research design To improve understanding of the factors that contribute to expert performance, players should be monitored over a prolonged period of time (Williams and Reilly, 2000a). By adopting a longitudinal design it is possible to monitor the development of the performance level of talented youth field hockey players. Although the majority of researchers recommend conducting research within a large group of young talented players, measuring multidimensional performance characteristics in a sports-specific way, following the players from childhood to elite senior standard (e.g., Hoare and Warr, 2000; Morris, 2000; Reilly et al., 2000), thus far no study in talented field hockey players combined all these aspects. 1.4 Objective and outline

The aim of this thesis is to gain a deeper insight into the relation between (the development of) multidimensional performance characteristics and the level of performance in talented youth field hockey players.

In chapter 2, attention is paid to the measurement of the multidimensional performance characteristics important for high-performance in youth field hockey players. The Shuttle Sprint and Dribble Test and the Slalom Sprint and Dribble Test have been developed for the purpose of this study and a paper on the development of these two field hockey specific tests is included.

In chapter 3, a study conducted within a group of all talented youth field hockey players is presented. To determine the relation between multidimensional performance characteristics and performance level, elite youth players were compared with sub-elite youth players on anthropometric, physiological, technical, tactical, and psychological characteristics.

In chapter 4, longitudinal data are presented on the talented youth field hockey players that have been followed across time. A comparison was made between elite and sub-elite youth players in terms of anthropometric, physiological, technical, tactical, and psychological characteristics measured on three occasions, each separated by a time interval of one year.

Chapter I II III IV V VI VII VIII

7

In chapter 5, a model of the development of an important physiological performance characteristic, the interval endurance capacity, is presented. Scores on the Interval Shuttle Run Test for interval endurance capacity were modeled for female and male, elite and sub-elite players in the age-band from 12 to 19 years.

The studies on the relation between multidimensional performance characteristics and performance level in talented youth field hockey players, described in the former chapters, show that psychological characteristics distinguish elite from sub-elite youth field hockey players. To investigate whether this finding is specific for field hockey or can be generalized to other sports, a study to reveal the relationship between psychological skills and level of performance within talented youth athletes in field hockey, basketball, volleyball, speed skating, and swimming is presented in chapter 6.

In chapter 7, the measurement of tactical skills is discussed. In the studies described in chapters 3 and 4, tactical skills were measured by the opinion of the trainers. Although these trainers are experts in the field and their opinion is highly valued, one might argue that their judgment of a player’s tactical skills is influenced by their knowledge of that player’s performance level. Therefore, we conducted a study with the purpose of developing a practical, reliable, and valid self-report instrument to measure tactical skills in sports: the Tactical Skills Inventory for Sports.

In chapter 8, the results of the different studies are combined into a general discussion and conclusions are drawn.

8

References

Atkinson, G. and Nevill, A.M. (2001). Selected issues in the design and analysis of sport performance research. Journal of Sports Sciences, 19, 811-827.

Bangsbo, J. and Lindquist, F. (1992). Comparison of various exercise tests with endurance performance during soccer in professional players. International Journal of Sports Medicine, 13, 125-132.

Borms, J. (1996). Early identification of athletic talent. Keynote Adresssed to the International Pre-Olympic Scientific Congress, Dallas, TX, USA.

Burwitz, L., Moore, P.M., and Wilkinson, D.M. (1994). Future directions for performance-related sports science research: An interdisciplinary approach. Journal of Sports Sciences, 12, 93-109.

Ericsson, K.A. (1996). The acquisition of expert performance: An introduction to some of the issues. In The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games (edited by K.A. Ericsson), pp. 1-50. Mahwah, NJ: Lawrence Erlbaum Associates.

Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406.

Ghosh, A.K., Goswami, A., Mazumdar, P., and Mathur, D.N. (1991). Heart rate & blood lactate response in field hockey players. Indian Journal of Medical Research, 94, 351-356.

Helsen, W.F., Hodges, N.J., Van Winckel, J., and Starkes, J.L. (2000). The roles of talent, physical precocity and practice in the development of soccer expertise. Journal of Sports Sciences, 18, 727-736.

Hoare, D.G. and Warr, C.R. (2000). Talent identification and women’s soccer: An Australian experience. Journal of Sports Sciences, 18, 751-758.

Howe, M.J.A., Davidson, J.W., and Sloboda, J.A. (1998). Innate talents: Reality or myth. Behavioral and Brain Sciences, 21, 399-442.

Hughes, M.D. and Barlett, M. (2002). The use of performance indicators in performance analysis. Journal of Sports Sciences, 20, 739-754.

Janelle, C.M. and Hillman, C.H. (2003). Expert performance in sport. Current perspectives and critical issues. In Expert performance in sports: Advances in research on sport expertise (edited by J.L. Starkes and K.A. Ericsson), pp. 19-47. Champaign, IL: Human Kinetics.

Lemmink, K.A.P.M., Verheijen, R., and Visscher, C. (2004). The discriminative power of the Interval Shuttle Run Test and the Maximal Multistage Shuttle Run Test for playing level of soccer. Journal of Sports Medicine and Physical Fitness, 44, 233-239.

Mahoney, M.J., Gabriel, T.J., and Perkins, T.S. (1987). Psychological skills and exceptional performance. The Sport Psychologist, 1, 181-199.

McPherson, S.L. (1994). The development of sport expertise: Mapping the tactical domain. Quest, 46, 223-240.

Morris, T. (2000). Psychological characteristics and talent identification in soccer. Journal of Sports Sciences, 18, 715-726.

Patel, D.R., Stier, B., and Luckstead, E.F. (2002). Major international sport profiles. Pediatric Clinics of North America, 49, 769-792.

Régnier, G., Salmela, J.H., and Russell, S.J. (1993). Talent detection and development in sport. In A Handbook of Research on Sports Psychology (edited by R. Singer, M. Murphey, and L.K. Tennant), pp. 290-313. New York: Macmillan.

Reilly, T. and Borrie, A. (1992). Physiology applied to field hockey. Sports Medicine, 14, 10-26. Reilly, T. and Bretherton, S. (1986). Multivariate analysis of fitness of female field hockey players. In

Perspectives in kinanthropometry (edited by J.A.P. Day), pp. 135-142. Champaign, IL: Human Kinetics.

Reilly, T. and Seaton, A. (1990). Physiological strain unique to field hockey. The Journal of Sports Medicine and Physical Fitness, 30, 142-146.

Reilly, T., Williams, A.M., Nevill, A., and Franks, A. (2000). A multidisciplinary approach to talent identification in soccer. Journal of Sports Sciences, 18, 695-702.

Chapter I II III IV V VI VII VIII

9

Russell, K. (1989). Athletic talent: From detection to perfection. Science Periodical on Research and Technology in Sport, 9, 1-6.

Spencer, M., Lawrence, S., Rechichi, C., Bishop, D., Dawson, B., and Goodman, C. (2004). Time-motion analysis of elite field hockey, with special reference to repeated-sprint activity. Journal of Sports Sciences, 22, 843-850.

Starkes, J.L. (1993a). Motor experts: Opening thoughts. In Cognitive issues in motor expertise (edited by J.L. Starkes and F. Allard), pp. 3-16. Amsterdam: Elsevier.

Starkes, J.L. (1993b). Skill in field hockey: The nature of the cognitive advantage. Journal of Sport Psychology, 9, 146-160.

Starkes, J.L. (2000). The road to expertise: Is practice the only determinant? International Journal of Sport Psychology, 31, 431-451.

Starkes, J.L. and Ericsson, K.A. (2003). Expert performance in sports: Advances in research on sport expertise. Champaign, IL: Human Kinetics.

Starkes, J.L., Deakin, J.M., Allard, F., Hodges, N.J., and Hayes, A. (1996). Deliberate practice in sports: What is it anyway? In The road to excellence: The acquisition of expert performance in the arts and sciences, sports, and games (edited by K.A. Ericsson), pp. 81-106. Mahwah, NJ: Lawrence Erlbaum.

Van Rossum, J.H.A. (2000). Deliberate practice and Dutch field hockey: An addendum to Starkes. International Journal of Sport Psychology, 31, 452-460.

Williams, A.M. and Reilly, T. (2000a). Talent identification and development in soccer. Journal of Sports Sciences, 18, 657-667.

Williams, A.M. and Reilly, T. (2000b). Searching for the Stars. Special issue of the Journal of Sports Sciences, 18, 655-775.

Wilmore, J.H. and Costill, D.L. (1999). Physiology of sport and exercise (2nd ed.). Champaign, IL: Human Kinetics.

Chapter II

Evaluation of the reliability of two field hockey specific sprint and dribble tests in young field hockey players Lemmink, K.A.P.M., Elferink-Gemser, M.T., and Visscher, C.(2004). British Journal of Sports Medicine, 38, 138-142. Acknowledgements:

The authors wish to express their sincere appreciation to Mieke Richart and Inge Scheek for their assistance in this research project

12

Abstract

The goal was to determine the reliability of two field hockey-specific tests: the Shuttle Sprint and Dribble Test (ShuttleSDT) and the Slalom Sprint and Dribble Test (SlalomSDT). The shuttle sprint and dribble performances of 22 young male and 12 young female field hockey players were assessed on two occasions within 4 weeks. Twenty one young female field hockey players took part in the SlalomSDT twice in a 4 week period. The ShuttleSDT required the players to perform three 30-m shuttle sprints while carrying a hockey stick alternated with short periods of rest and, after a 5-minute rest, three 30-m shuttle sprints alternated with rest while dribbling a hockey ball. The SlalomSDT required the players to run a slalom course and, after a 5-minute rest, to dribble the same slalom with a hockey ball. There were no differences in mean time scores between the two test sessions. The mean differences were small when compared with the means of both test sessions. With the exception of the slalom sprint time, zero lay within the 95% confidence interval of the mean differences indicating that no bias existed between the two measurements. With the exception of delta shuttle time (0.79), all intraclass correlation coefficient values for the ShuttleSDT, met the criterion for reliability of 0.80. Intraclass correlation coefficient values for SlalomSDT were 0.91 for slalom sprint time, 0.78 for slalom dribble time, and 0.80 for delta slalom time. This study shows that the ShuttleSDT and the SlalomSDT are reliable measures of sprint and dribble performances of young field hockey players.

Chapter I II III IV V VI VII VIII

13

2.1 Introduction

Recent developments in field hockey, such as the artificial playing surface, new stick material, and the interchange rule, have increased the number of physiological and technical demands made on field hockey players at all levels, but in particular at the elite level. Competitive field hockey matches place heavy aerobic demands on players and require them to expend energy at relatively high levels (Reilly and Borrie, 1992; Boyle et al., 1994). High-intensity activities such as cruising, sprinting, and activities in which the player is directly involved with the ball (for example, dribbling) have been shown to represent between 17.5 - 30% of the competition time (Lothian and Farrally, 1994), and are considered critical to the outcome of the game. Furthermore, in field hockey, high and low intensity activities alternate by a ratio ranging from about 1:4 to 1:8 (Lothian and Farrally, 1994). Consequently, as well as maximal performance on individual high intensity activities, the ability to produce high intensity efforts is crucial for top level field hockey players.

Field hockey is a multiple high intensity activity sport with a multidirectional nature. The ability to change direction rapidly while maintaining balance without loss of speed – that is, agility - is therefore an important physical component necessary for successful performance in field hockey. Elite field hockey players also need high level technical skills such as being able to dribble without losing running speed. For a technically good player, dribbling is essentially an automatic process, and the better players distinguish themselves by their running speed while dribbling the ball (Reilly and Bretherton, 1986).

Coaches, trainers, and players are continually searching for effective methods of identifying and developing those characteristics in a player that may enhance performance. There are a variety of field tests with which to measure the physiological and technical characteristics of players in team games like soccer, rugby, and handball. However, there was no single test to measure both physiological and technical characteristics in field hockey players and for this reason we developed two tests specifically to measure these characteristics. Based on tests for repeated sprint ability (Baker et al., 1993; Fitzsimons et al., 1993; Bangsbo, 1994; Aziz et al., 2000; Lemmink et al., 2000; Wragg et al., 2000; Bishop et al., 2001; Boddington et al., 2001) and dribbling skills of field hockey players (Reilly and Bretherton, 1986) and soccer players (Reilly and Holmes, 1983; Van Rossum and Wijbenga, 1993) we developed the field hockey specific Shuttle Sprint and Dribble Test (ShuttleSDT) to measure shuttle sprint and dribble performance. Based on tests for agility (Pauole et al., 2000) and dribbling skills of field hockey players (Reilly and Bretherton, 1986) and soccer players (Reilly and Holmes, 1983; Van Rossum and Wijbenga, 1993), the field hockey specific Slalom Sprint and Dribble Test (SlalomSDT) was developed to measure slalom sprint and dribble performance.

14

It is vital that the ShuttleSDT and the SlalomSDT provide reliable information. A reliable test must perform consistently. In other words, if an individual whose ability or skill has not changed is tested twice with a completely reliable measuring device, both scores will be identical (Baumgartner and Jackson, 1999). The aim of this study was therefore to determine the reliability of the ShuttleSDT and the SlalomSDT in young elite field hockey players.

2.2 Methods

Participants A total of 34 young male (n = 22) and female (n = 12) field hockey players participated in the reliability study of the field hockey specific Shuttle Sprint and Dribble Test (ShuttleSDT). The mean age of the boys was 15.5 years (sd = 1.8), and of the girls 13.8 years (sd = 1.0). Twenty one young female field hockey players whose mean age was 13.5 years (sd = 1.3) volunteered to take part in the reliability study of the field hockey-specific Slalom Sprint and Dribble Test (SlalomSDT). After being informed about the study procedure, the subjects gave their verbal consent to participation. The group averaged two training sessions and one match per week. Procedure To examine the reliability of the ShuttleSDT and the SlalomSDT, two trials were conducted within a period of 2 to 4 weeks and during the subjects’ normal training hours, varying between 16:30 and 20:00. On day one and two the average temperature during testing was 4.3 and 3.7 ˚C respectively, humidity was 86-98 % on both days and wind conditions (no high winds) were comparable. The tests were conducted on a synthetic pitch surface laid on a sandy field with subjects wearing their normal playing footwear. The subjects were only given feedback on their performance after completing all the tests.

Shuttle Sprint and Dribble Test (ShuttleSDT) The ShuttleSDT was developed to measure field hockey specific shuttle sprint and dribble performance. This study used a modified version of the Interval Sprint Test protocol first introduced by Lemmink et al. (2000). This was originally performed by soccer players outdoors on a grass surface. Our modifications involved an electronic timing system and slight modifications of the sprint distances, three instead of ten sprints, and the use of a hockey stick and a hockey ball. The protocol consisted of three maximal sprints of 32 m while carrying a hockey stick and three maximal sprints of 32 m while dribbling a hockey ball. Each 32-m sprint included a 6-m and a 10-m shuttle sprint. Timing procedures (timing gates),

Chapter I II III IV V VI VII VIII

15

meant that the initial and final metres of the sprint were not timed, so data are based on 30-m distances (Figure 2.1).

Figure 2.1. Course of the Shuttle Sprint and Dribble Test (ShuttleSDT).

The subject began the test standing with both feet behind line A (marked with two cones 2 m apart). On an auditory signal after a 5 second countdown, the subject sprinted 6 m to line B (marked with two cones,) touched the line with one foot and returned to line A, again touching the line with one foot. The subject then sprinted 10 m to line C (marked with two cones), touched the line with one foot and returned to finish over line A. The subject then tapered down from the sprint, turned and walked back slowly to line A, there waiting for the 5 second countdown and the auditory signal to start the second sprint. The second sprint started exactly 20 seconds after the start of the first sprint. After the third sprint the subject was allowed 5 minutes recovery time, during which he/she walked. The recovery walk was timed so that the subject had returned to line A 10-20 seconds before the start of the dribbling portion of the test. The protocol of the dribbling portion was identical to the sprinting portion, except that the subject was now dribbling a hockey ball.

Timing data were measured by means of photocell gates (Eraton BV, Weert, the Netherlands) placed at 1.05 m above ground (approximately at hip height) and at 1.0 m behind line A. The photocells were linked to an electronic timer with an accuracy of 0.01 seconds.

16

The following variables were noted and recorded: sprint times: individual sprint times dribble times: individual dribble times peak sprint time: fastest sprint time peak dribble time: fastest dribble time total sprint time: total sprint time of the three sprints total dribble time: total dribble time of the three dribbles delta shuttle time: difference between the total dribble time and the total sprint time

Slalom Sprint and Dribble Test (SlalomSDT) Based on tests for agility and dribbling skills, the field hockey specific slalom sprint and dribble test (SlalomSDT) was developed to measure field hockey specific slalom sprint and dribble performance. The protocol consisted of a maximal slalom sprint of 30 m while carrying a hockey stick and a maximal slalom dribble of 30 m while dribbling a hockey ball. Twelve cones were placed in a zigzag pattern (Figure 2.2). Start and finish lines (A and B) were marked by two cones.

Figure 2.2. Course of the Slalom Sprint and Dribble Test (SlalomSDT).

The subject began the test with both feet behind line A; then, upon an auditory signal after a 5 second countdown, the subject ran with a hockey stick around the 12 cones finishing over line B. After the run the subject was allowed 5 minutes for recovery, during which he/she walked slowly. The total distance of the course was 29.07 m. The recovery walk was timed so that the subject had returned to line A 10-20 seconds before the start of the next portion. The protocol of the dribbling portion was identical to the sprinting portion, except that the subject was now dribbling a hockey ball. If the subject lost control of the ball – that is, if the subject was more than approximately 2 m away from the cones, the test was repeated. Timing data were

Chapter I II III IV V VI VII VIII

17

measured using a stopwatch. Slalom sprint time, slalom dribble time, and the difference between the slalom times of the dribble and sprint (delta slalom time) were noted and recorded accurately to within 0.01 seconds.

Data analysis The ShuttleSDT data are expressed as mean (standard deviation, sd). To determine the relation between the times measured, a correlation matrix was calculated for the test scores at t1. A three way (time x sprint/dribble x test session) analysis of variance with repeated measures was used to determine differences in times of each sprint/dribble. A Scheffé post hoc test was used to identify specific differences when the main effects were significant.

To determine the reliability of the ShuttleSDT, the data of both boys and girls were analysed together. ShuttleSDT reliability analysis was carried out on the peak and total times of the sprints and dribbles and on the delta shuttle time. The mean difference between the test results on both days was set as a measure of absolute reliability. If zero lay within the 95% confidence interval (CI) of the mean difference, we concluded that no bias existed between the measurements (Bland and Altman, 1986; Altman and Gardner, 1989; Rankin and Stokes, 1998).

To determine relative reliability, we used a one way analysis of variance (ANOVA) to calculate intraclass correlation coefficients (ICCs) of repeated interval scale measures (Baumgartner, 1989; Rankin and Stokes, 1998; Baumgartner and Jackson, 1999). Intraclass correlation coefficients were determined for the peak and total sprint and dribble times and for the delta shuttle time. Ninety five per cent confidence intervals were determined for all of the ICCs (Rankin and Stokes, 1998). As a general rule, an intraclass correlation coefficient over 0.90 is considered to be high, between 0.80 and 0.90 moderate, and below 0.80 to be insufficient for physiological field tests (Vincent, 1995). Baumgartner and Jackson (1999) stated that ICCs of a minimum of 0.80 are acceptable for physical measures.

The SlalomSDT data are expressed as mean (standard deviation, sd). To determine the relation between the times measured, a correlation matrix was calculated for the test scores at t1. A two way (sprint/dribble x test session) analysis of variance with repeated measures was used to determine the differences in time of the slalom sprint and slalom dribble.

To determine reliability, analysis of variance was performed on the test scores of the SlalomSDT. For absolute reliability the mean difference with a 95% CI between the testing days was calculated (Bland and Altman, 1986; Altman and Gardner, 1989; Rankin and Stokes, 1998). Intraclass correlation coefficients were calculated with a 95% confidence interval for the slalom times of the sprint and dribble and the calculated delta slalom time to determine relative reliability.

18

2.3 Results

The correlation matrix of the ShuttleSDT showed strong correlations between the time scores of the sprinting and dribbling portions individually (Table 2.1). Weak correlations existed between the time scores of the sprints and the dribbles. The delta shuttle time was strongly related to the time scores of the dribbling portion of the ShuttleSDT but not to the time scores of the sprinting portion. SlalomSDT correlations showed a weak relationship between the slalom sprint and dribble times (r = 0.24). As in the ShuttleSDT, the delta slalom time was strongly correlated with the slalom time of dribbling a hockey ball (r = 0.90) but not with the slalom sprint time (r = -0.21). Table 2.1. Intercorrelations (Pearson correlation coefficients) between the sprint (S) and dribble (D)

times, peak sprint and dribble times, total sprint and dribble times, and the calculated delta shuttle time of the ShuttleSDT at t1 (n = 34). All times are expressed in seconds.

S2 S3 Speak Stot D1 D2 D3 Dpeak Dtot Delta

S1 0.79* 0.80* 0.94* 0.93* 0.22 0.48* 0.28 0.35 0.35 -0.15

S2 0.78* 0.87* 0.93* 0.30 0.53* 0.45* 0.43 0.47* -0.02

S3 0.83* 0.92* 0.35 0.60* 0.44* 0.47* 0.50* 0.03

Speak 0.95* 0.24 0.52* 0.29 0.38 0.38 -0.13

Stot 0.31 0.58* 0.43 0.45* 0.48* -0.05

D1 0.75* 0.76* 0.93* 0.91* 0.85*

D2 0.76* 0.84* 0.91* 0.69*

D3 0.85* 0.93* 0.81*

Dpeak 0.96* 0.82*

Dtot 0.86*

Delta

Note: *Significant correlation coefficient (p < 0.01).

Chapter I II III IV V VI VII VIII

19

There were significant differences between the shuttle sprint and dribble times at both test sessions (p < 0.01) (Figure 2.3). There were no differences in time scores between the test sessions (p = 0.98) (Figure 2.3). The mean time of each of the three sprints and dribbles on the ShuttleSDT increased significantly at both tests sessions (p < 0.01) (Table 2.2). Post-hoc analysis showed that each of the three sprint and dribble times was significantly different from the others (p < 0.05). For the SlalomSDT, there were significant differences between the sprint and dribble time at both test sessions but not between test sessions (p = 0.09).

1 2 38

8.5

9

9.5

10

10.5

11

Tim

e (s

)

Sprints t1*

Sprints t2*

Dribbles t1*

Dribbles t2*

Figure 2.3. Group data for the sprint and dribble times of the Shuttle Sprint and Dribble Test (ShuttleSDT) at the first and second test sessions (t1 and t2). Means and standard deviations of each data point are presented in Table 2.2.

* Significant difference between first, second, and third sprints and dribbles (p < 0.05)

20

Table 2.2 shows the means and standard deviations of all sprint and dribble times at the first and second test session (t1 and t2) of the ShuttleSDT and the SlalomSDT, and the mean difference, standard error and 95% CI of the mean difference to determine absolute reliability (Rankin and Stokes, 1998). In general, the values of the mean differences between the first and second test sessions were small when compared with the means of the two test sessions. With the exception of the slalom sprint time, zero lay within the 95% CI, which indicates reasonable agreement between the two testing days. Table 2.2. Results of the Bland and Altman method for absolute reliability of the ShuttleSDT

(n = 34) and the SlalomSDT (n = 21).

ShuttleSDT t1 mean (sd) t2 mean (sd) Mean d (sd) SE of d 95% CI

Sprint 1 8.35 (0.525) 8.33 (0.579) -0.026 (0.446) 0.076 -0.182 – 0.129

Sprint 2 8.55 (0.567) 8.55 (0.632) -0.007 (0.570) 0.098 -0.206 – 0.192

Sprint 3 8.62 (0.522) 8.61 (0.664) -0.010 (0.507) 0.087 -0.187 – 0.167

Peak sprint time 8.28 (0.511) 8.29 (0.563) 0.010 (0.434) 0.075 -0.141 – 0.162

Total sprint time 25.52 (1.495) 25.48 (1.817) -0.043 (1.366) 0.234 -0.520 – 0.434

Dribble 1 10.00 (0.997) 9.83 (1.045) -0.173 (0.737) 0.126 -0.430 – 0.084

Dribble 2 10.21 (0.964) 10.29 (1.115) 0.077 (0.965) 0.165 -0.260 – 0.413

Dribble 3 10.53 (1.186) 10.45 (1.368) -0.076 (1.098) 0.188 -0.459 – 0.307

Peak dribble time 9.79 (0.829) 9.66 (0.887) -0.135 (0.487) 0.083 -0.305 – 0.035

Total dribble time 30.74 (2.882) 30.57 (3.093) -0.173 (1.912) 0.328 -0.840 – 0.494

Delta shuttle time 5.22 (2.539) 5.09 (2.137) -0.130 (1.962) 0.337 -0.814 – 0.555

SlalomSDT

Slalom sprint time 16.36 (0.751) 16.15 (0.736) -0.207 (0.419) 0.091 -0.398 – -0.016

Slalom dribble time 20.93 (1.689) 20.56 (1.513) -0.366 (1.366) 0.298 -0.988 – 0.256

Delta slalom time 4.57 (1.679) 4.41 (1.774) -0.159 (1.403) 0.306 -0.797 – 0.480

Note: t1 = first test session, t2 = second test session, sd = standard deviation; d = difference, SE = standard error, CI = confidence interval. All times are expressed in seconds.

Chapter I II III IV V VI VII VIII

21

Intraclass correlation coefficient values to assess the relative reliability of the ShuttleSDT ranged from 0.71 for the second sprint time to 0.91 for the peak dribble time (Table 2.3). After excluding the separate sprint and dribble times of the ShuttleSDT, all ICC values (with the exception of the delta shuttle time (0.79)) met the criterion of 0.80 for reliability. Intraclass correlation coefficient values to assess the SlalomSDTs relative reliability were 0.91 for the slalom sprint time, 0.78 for the slalom dribble time, and 0.80 for the delta slalom time (Table 2.3). With the exception of the slalom dribble time, the ICC values met the reliability criterion of 0.80. Table 2.3. Intraclass correlation coefficients for relative reliability of the ShuttleSDT (n = 34) and

the SlalomSDT (n = 21).

ICC 95% CI

ShuttleSDT

Sprint 1 (s) 0.81 0.61 – 0.90

Sprint 2 (s) 0.71 0.42 – 0.86

Sprint 3 (s) 0.78 0.56 – 0.89

Peak sprint time (s) 0.81 0.61 – 0.90

Total sprint time (s) 0.80 0.59 – 0.90

Dribble 1 (s) 0.85 0.70 – 0.93

Dribble (s) 0.73 0.45 – 0.86

Dribble 3 (s) 0.77 0.55 – 0.89

Peak dribble time (s) 0.91 0.82 – 0.96

Total dribble time (s) 0.89 0.77 – 0.94

Delta shuttle time (s) 0.79 0.58 – 0.89

SlalomSDT

Slalom sprint time (s) 0.91 0.78 – 0.97

Slalom dribble time (s) 0.78 0.45 – 0.91

Delta slalom time (s) 0.80 0.51 – 0.92

Note: ICC = Intraclass Correlation Coefficient, CI = Confidence Interval.

22

2.4 Discussion

Using repeated sprint ability and agility tests and taking into account the multidirectional and technical nature of field hockey, we developed two field tests to determine shuttle sprint and shuttle dribble performance as well as slalom sprint and dribble performance. Both tests are practical for use on a regular basis as they can be administered easily and are popular with the players. Low correlations and significant differences in mean sprint and dribble times on both tests confirmed our expectation that different physical abilities were being measured. The course, duration, and repetitive nature of the ShuttleSDT mean that the test is probably of more importance for defenders and midfielders. Conversely, the course of the SlalomSDT makes the test more relevant to forwards.

Although several authors use other measures (for example, Pearson’s correlation, 95% limits of agreement, coefficient of repeatability, and coefficient of variation) mean difference, standard errors, 95% CI of the mean differences and ICC values have all recently been reported as being most appropriate and clear in determining reliability (Rankin and Stokes, 1998). In our reliability data, the 95% CIs for the mean differences between the test days can be interpreted as a minimum difference between the results of individuals that indicate a real change in performance level. This indicates the accuracy of the test in monitoring changes over time.

We have no reason to expect that reliability of field testing is influenced by the subjects’ youth or gender. Other characteristics such as heterogeneity, motivation to do well, and learning capabilities are assumed to be factors that affect reliability in a positive way (Baumgartner and Jackson, 1999). Our subjects were a homogeneous group, very motivated, and with above average learning capabilities as they all played field hockey at the highest regional level. Environmental conditions do influence field testing. Therefore, ambient temperature, humidity, and wind conditions were all documented. There were only minor differences in environmental conditions during the test sessions. Furthermore, tests were conducted on the same artificial grass surface with players wearing their normal playing footwear.

The measurements with the most robust relative reliability were the peak dribble time in the ShuttleSDT and the slalom sprint time in the SlalomSDT (ICC = 0.91). As mentioned earlier, an intraclass correlation coefficient of over 0.90 is considered to be high, between 0.80 and 0.90 moderate, and below 0.80 insufficient for physiological field tests (Vincent, 1995). Based on these criteria, it is reasonable to suggest that the peak sprint and dribble times and the total sprint and dribble times of the ShuttleSDT have an acceptable relative reliability. The delta shuttle time has insufficient reliability (ICC = 0.79). The absolute reliability data of the ShuttleSDT showed reasonable agreement between both testing sessions.

Chapter I II III IV V VI VII VIII

23

Reliability ICC values of the SlalomSDT suggest that the slalom sprint and the delta slalom time have acceptable relative reliability. The slalom dribble time has insufficient reliability (ICC = 0.78) and should therefore be used with caution. The reliability of a test depends on many factors, for instance the nature of a test (Baumgartner and Jackson, 1999). Scores may not be stable if subjects have not had experience of or practice at the test before being measured. In tests requiring skill, such as the dribble performance in the ShuttleSDT and the SlalomSDT, the learning effect may have influenced the reliability of the scores. This corresponds with the trend of faster times on the dribble performance on the ShuttleSDT and SlalomSDT seen at t2. However, only the mean difference and 95% CI for the slalom sprint time on the SlalomSDT between the first and second testing sessions indicated a bias and, therefore, a lack of absolute reliability. Subjects needed less time for the slalom sprint course at t2, indicating that there might have been a learning effect. However, only a series of trials can lead to a more definite conclusion about this.

The reliability coefficients of this study are in line with those obtained when evaluating other field tests in adult field hockey players and other subjects. A study of a 5-m multiple shuttle test in 23 female field hockey players to determine players’ match related fitness reported a range of coefficients from ICC = 0.74 to 0.98 (Boddington et al., 2001). Fitzsimons et al. (1993) reported correlation coefficients of 0.75 to 0.94 for a running test of repeated sprint ability in 15 male field hockey players. Pauole et al. (2000) reported an intraclass reliability coefficient of 0.98 for a test of agility (T test) in college aged men and women. Finally, Baker et al. (1993) reported a Pearson correlation coefficient of 0.86 for a repeated maximal shuttle run test in 10 male subjects.

In summary, coaches and trainers can use the field tests examined in this study, as the demands of the test are important ones for field hockey performance (alternation of high- and low-intensity activities, agility, speed, and technical skills). It is also a practical test to use on a regular basis because it can be administered easily. The ShuttleSDT and the SlalomSDT can help coaches and trainers to assess young athletic talent, diagnose specific weaknesses, provide information for the development of individualised training programmes, and assess changes in physical characteristics as result of a training cycle. This study has shown that the absolute and relative reliability of the ShuttleSDT and the SlalomSDT are satisfactory. The ShuttleSDT and the SlalomSDT are sports specific field tests to measure sprint and dribble performances of field hockey players. The tests showed reasonable reliability and can help coaches and trainers assess young athletic talent, differentiate between players and monitor changes over time.

24

References

Altman, D.G. and Gardner, M.J. (1989). Calculating confidence intervals for means and their differences. In Statistics with confidence (edited by M.J. Gardner and D.G. Altman DG), pp. 20-27. London.

Aziz, A.R., Chia, M., and Teh, K.C. (2000). The relationship between maximal oxygen uptake and repeated sprint performance indices in field hockey and soccer players. The Journal of Sports Medicine and Physical Fitness, 40, 195-200.

Baker, J., Ramsbottom, R., and Hazeldine, R. (1993). Maximal shuttle running over 40 m as a measure of anaerobic performance. British Journal of Sports Medicine, 27, 228-232.

Bangsbo, J. (1994). The physiology of soccer - with special reference to intense intermittent exercise. Acta Physiologica Scandinavica, 15 (suppl. 619), 1-156.

Baumgartner, T.A. (1989). Norm-referenced measurement: reliability. In Measurement concepts in physical education and exercise science (edited by M.J. Safrit MJ and T.M. Woods), pp. 45-72. Champaign, IL: Human Kinetics.

Baumgartner, T.A. and Jackson, A.S. (1999). Measurement for evaluation in physical education and exercise science. Boston: WCB-McGraw-Hill, sixth edition.

Bishop, D., Spencer, M., Duffield, R., and Lawrence, S. (2001). The validity of a repeated sprint ability test. Journal of Science and Medicine in Sport, 4, 19-29.

Bland, J.M. and Altman, D.G. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 1, 307-310.

Boddington, M.K., Lambert, M.I., St Clair Gibson, and A., Noakes, T.D. (2001). Reliability of a 5-m multiple shuttle test. Journal of Sports Sciences, 19, 223-228.

Boyle, P.M., Mahoney, C.A., and Wallace, W.F.M. (1994). The competitive demands of elite male field hockey. Journal of Sports Medicine and Physical Fitness, 34, 235-241.

Fitzsimons, M., Dawson, B., Ward, D., and Wilkinson, A. (1993). Cycling and running tests of repeated sprint ability. Australian Journal of Science and Medicine in Sport, 25, 82-87.

Lemmink, K.A.P.M., Dolleman, G., Verheijen, R., and Visscher, C. (2000). Interval Sprint Test en Interval Shuttle Run Test – betrouwbaarheid en discriminerend vermogen van twee nieuwe voetbaltests. [Interval Sprint Test and Interval Shuttle Run Test – reliability and discriminative power of two new tests for soccer players]. Geneeskunde en Sport, 33, 39-48.

Lothian, F. and Farrally, M. (1994). A time-motion analysis of women’s hockey. Journal of Human Movement Studies, 26, 255-265.

Pauole, K., Madole, K., Garhammer, J., Lacourse, M., and Rozenek, R. (2000). Reliability and validity of the T-test as a measure of agility, leg power, and leg speed in college-aged men and women. Journal of Strength and Conditioning Research, 14, 443-450.

Rankin, G. and Stokes, M. (1998). Reliability of assessment tools in rehabilitation: an illustration of appropriate statistical analyses. Clinical Rehabalitation, 12, 187-199.

Reilly, T. and Holmes, M. (1983). A preliminary analysis of selected soccer skills. Physical Education Review, 6, 64-71.

Reilly, T. and Bretherton, S. (1986). Multivariate analysis of fitness of female field hockey players. In Perspectives in kinanthropometry (edited by J.A.P. Day), pp. 135-142. Champaign, IL: Human Kinetics.

Reilly, T. and Borrie, A. (1992). Physiology applied to field hockey. Sports Medicine, 14, 10-26. Van Rossum, J.H.A. and Wijbenga, D. (1993). Soccer skills technique tests for youth players:

construction and implications. In Science and football II (edited by T. Reilly, J.Clarys, and A. Stibbe), pp. 313-318. London: E & FN Spon.

Vincent, W.J. (1995). Statistics in kinesiology. Champain, IL: Human Kinetics. Wragg, C.B., Maxwell, N.S., and Doust, J.H. (2000). Evaluation of the reliability and validity of a

soccer-specific field test of repeated sprint ability. European Journal of Applied Physiology, 83, 77-83.

Chapter I II III IV V VI VII VIII

25

Chapter III

Relation between multidimensional performance characteristics and level of performance in talented youth field hockey players Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th.(2004). Journal of Sports Sciences, 22, 1053-1063. Acknowledgements:

This study has been supported by a grant of the Dutch National Olympic Committee NOC*NSF. The authors thank all players, trainers, and staff of the field hockey clubs HC ’s Hertogenbosch and HC Rotterdam for their cooperation.

28

Abstract

To determine the relationship between multidimensional performance characteristics and level of performance in talented youth field hockey players, elite youth players (n = 38, mean age 13.2 years, sd = 1.3) were compared with sub-elite youth players (n = 88, mean age 14.2 years, sd = 1.3) on anthropometric, physiological, technical, tactical and psychological characteristics. Multivariate analyses with performance level and gender as factors, and age as the covariate, showed that the elite youth players scored better than the sub-elite youth players on technical (dribble performance in a peak and repeated shuttle run), tactical (general tactics; tactics for possession and non-possession of the ball) and psychological variables (motivation) (p < 0.05). The most discriminating variables were tactics for possession of the ball, motivation and performance in a slalom dribble. Age discriminated between the two groups, indicating that the elite youth players were younger than the sub-elite players. In the guidance of young talented players to the top as well as in the detection of talented players, more attention has to be paid to tactical qualities, motivation and specific technical skills.

Chapter I II III IV V VI VII VIII

29

3.1 Introduction

In the Netherlands, elite field hockey is played at a higher standard than in many other countries. The Dutch male field hockey players won the gold medal at the Olympic Games in Sydney 2000 while the female players won the bronze medal. In Athens 2004, both teams won silver. To maintain this level of performance, the Dutch National Olympic Committee has chosen ‘talent identification and development’ as one of its main research programmes. A talented young athlete is considered to be someone who performs better than his or her peers during training and competition, and who has the potential to reach the elite level (Howe et al., 1998; Helsen et al., 2000). Whereas in the 1970s and 1980s scientists focused mainly on the detection of talented athletes and developed sport talent-detection models (for a review, see Régnier et al., 1993), recently there has been a shift in emphasis from talent detection to talent guidance and development (Williams and Reilly, 2000). Talent development is based on the prediction of performance and consequently on the assumption that underlying factors determining excellence in sports really do exist (Kroll, 1970; Régnier et al., 1993). In team games like field hockey, however, the prediction of long-term success in young players is complex because multidimensional qualities are needed.

In the present study, we focus on youth field hockey players who were already designated as talented. Every year, many talented Dutch field hockey players are invited to participate in a selection team for their age category. These teams are provided extra training facilities and highly qualified trainers. Selected players compete in the highest Dutch junior competition for field hockey. Although all of these talented players are given the chance to develop their potential to the full, only a few of them ultimately make it to the top. To develop a successful sporting career, talented players have to perform at a high level at a young age, indicating well-developed anthropometric, physiological, technical, tactical and psychological characteristics (see Figure 3.1).

30

Figure 3.1. Multidimensional performance characteristics and level of performance in field hockey.

The multidimensional performance characteristics shown in Figure 3.1 are based on a limited number of determining factors for elite field hockey. Unique to field hockey is the semi-crouched posture, which causes extra physiological strain on players (Reilly and Seaton, 1990). Competitive match-play is a non-continuous, high-intensity, intermittent activity that places heavy demands on the aerobic energy system. The anaerobic system is also important: brief bursts of high-energy release are separated by periods of lower intensity (Bhanot and Sidhu, 1983; Reilly and Borrie, 1992; Boyle et al., 1994; Lothian and Farrally, 1994). Consequently, a successful player has to be able to perform successive short all-out sprints. The intermittent nature of, and the many changes of direction during, match-play underscores the importance of highly developed sprint capacity and performance in repeated sprints, as well as of an outstanding slalom sprint performance and interval endurance capacity of elite players (Reilly and Seaton, 1990; Lemmink et al., 2000).

Chapter I II III IV V VI VII VIII

31

Control of the ball while sprinting, turning, passing, and scoring goals is only possible if a player possesses excellent technical qualities. Straight dribbling is defined as running or sprinting in a straight line, whereas slalom dribbling is defined as running or sprinting with quick changes of direction while maintaining control of the ball (Smith and Chamberlin, 1992; Reilly et al., 2000). In Figure 3.1, these qualities make up the technical characteristics.

Due to the nature of field hockey, good sprinting ability, good endurance and the performance of highly developed technical skills are not sufficient if the timing of actions is not correct. This tactical knowledge is also referred to as “game intelligence”, and includes anticipation and decision-making skills. Tests to measure these qualities in soccer players show consistent differences between skilled and less skilled players (Williams and Davids, 1995; Williams, 2000). Tactical characteristics are part of the multidimensional performance characteristics in Figure 3.1.

To perform at the top level, elite players must be prepared to invest many hours of intensive training over many years (Ericsson et al., 1993). They also have to achieve under high pressure. It is therefore not surprising that psychological characteristics often distinguish elite from non-elite performers (Mahoney et al., 1987; Morris, 2000).

The multidimensional performance characteristics shown in Figure 3.1 are, to a certain extent, responsive to training interventions (Hoare and Warr, 2000). In addition, the environment of a talented player must not be underestimated in that parents and coaches play an important role in helping talented athletes to improve themselves during their sporting careers (Bloom, 1985; Carlson, 1988; 1993; Côté, 1999; Visscher et al., 2004).

Until now, only a few multivariate approaches focusing on identifying talent in team sports have been completed (Deshaies et al., 1979; Pienaar et al., 1998). In these studies, elite players have been compared to their non-elite counterparts. Reilly et al. (2000) used a multidisciplinary approach to distinguish between elite and sub-elite soccer players on the basis of performance on test items. They recommended a study with a pool of already selected talented athletes who were exposed to systematic training. For this reason, the present study focuses on youth field hockey players who were all considered to be talented.

The main aim of this study is to determine whether a relationship exists between multidimensional performance characteristics and level of performance in talented youth field hockey players. A comparison is made between elite youth players and youth players immediately below this level, in terms of anthropometric, physiological, technical, tactical and psychological characteristics.

32

3.2 Methods

Participants A total of 126 talented field hockey players from 12 selection teams participated in this study. There were 63 female and 63 male players. The mean age of the female players was 13.9 years (sd = 1.3, range 12-16), the mean age of the male players 13.9 years (sd = 1.4, range 11-16). All players were considered to be talented, since they were already playing in a selection team of a field hockey club of national prestige. Thus that all participants were playing at the highest level possible for their age category. All players were tested at the end of the 2000-2001 Dutch competitive field hockey season. Each participant was assessed based on the following five categories: anthropometric, physiological, technical, tactical and psychological. Field tests were organized on modern synthetic field hockey playing surfaces (water-based pitches).

In addition to playing in their club’s selection team, talented Dutch players who are considered to be current elite youth players are invited to train and play in a youth selection team of the Dutch Field Hockey Association. Talented players who are considered to be current sub-elite youth players only play in their club’s selection team. This distinction, based on the performance of the players in the 2000-2001 season, was also adopted in this study, resulting in 38 elite youth players and 88 sub-elite youth players. Table 3.1 shows the general training characteristics of the players.

Table 3.1. Scores of general characteristics related to training of talented youth field hockey players

classified by gender and level of performance (mean; standard deviation).

Female youth players Male youth players

Elite players

n = 17

Sub-elite

players

n = 46

Elite players

n = 21

Sub-elite

players

n = 42

Age (years) 13.18 (1.29) 14.15 (1.25) 13.24 (1.26) 14.19 (1.29)

Field hockey experience (years) 8.45 (1.47) 7.65 (1.62) 7.38 (2.01) 7.61 (2.09)

Training sessions per week 2.75 (0.72) 2.21 (0.41) 3.10 (0.62) 2.20 (0.56)

Matches per week 1.05 (0.22) 1.00 (0.00) 1.24 (0.44) 1.00 (0.32)

Procedure All players were informed about the procedures of the study before providing their verbal consent to participate. The governing body of the clubs and the trainers also gave their permission for the study to proceed. The procedures were in accordance with the ethical

Chapter I II III IV V VI VII VIII

33

standards of the Medical Faculty of the University of Groningen. The players completed all tests at the end of the competitive field hockey season. They were told that the results would be used anonymously and were asked to fill in the questionnaire honestly to ensure maximum accuracy and validity of the results. Anthropometric characteristics Three variables were measured for each player: height, body mass and percentage body fat. The latter was estimated by means of leg-to-leg bioelectrical impedance (BIA) analysis (Valhalla BIA, Valhalla, Inc., San Diego, CA, USA). Both within-day and between-day coefficients of variations of these analyses were comparable to conventional commercially available BIA systems (Nunez et al., 1997). Physiological characteristics All players performed three field tests to determine four physiological characteristics. These characteristics included peak shuttle sprint performance, repeated shuttle sprint performance, slalom sprint performance and interval endurance capacity. Peak shuttle sprint and repeated shuttle sprint performance were measured by means of the Shuttle Sprint and Dribble Test (ShuttleSDT; Lemmink et al., 2004a) (see Figure 3.2). In this field hockey specific test, participants have to sprint 30 m three times while carrying a hockey stick. The players are allowed a short rest between successive 30-m sprints. The length of this rest period depends on how fast the sprint is performed: the next sprint starts exactly 20 s after the start of the previous sprint. Each 30-m sprint has three 180-degree turns, which they have to cross with both feet: after 5 m, participants have to turn and sprint back 6 m. Here they turn for the second time and sprint 10 m. After turning for the third time, they sprint back 9 m to the finish. Electronic timing lights are used to measure the time (TAG Heuer, Eraton BV Digital Timing Equipment, Weert, the Netherlands). Peak shuttle sprint performance is indicated by the time covered in the fastest of three 30-m sprints; repeated shuttle sprint performance is the total time covered by all three 30-m sprints.

34

Figure 3.2. Course for the Shuttle Sprint and Dribble Test (ShuttleSDT).

Relative and absolute test-retest reliability were shown for the sprinting part of the ShuttleSDT (Lemmink et al., 2004a). If the intraclass correlation coefficient (ICC) exceeded 0.80 and if zero lay within the 95% confidence interval (CI) of the mean difference, we concluded that no bias existed between the two measurements (peak shuttle sprint performance: ICC = 0.81 and 95% CI for d = -0.141 to 0.162; repeated shuttle sprint performance: ICC = 0.80 and 95% CI for d = -0.520 to 0.434).

Slalom sprint performance was measured by using the Slalom Sprint and Dribble Test (SlalomSDT; Lemmink et al., 2004a) (see Figure 3.3). In this field hockey specific test, players have to sprint 30 m in a zig-zag fashion with twelve 120-degree turns around cones placed 2 m apart while carrying a hockey stick. Relative test-retest reliability was shown for the sprinting part of the SlalomSDT (ICC = 0.91), whereas in terms of absolute reliability there was some evidence of systematic error (95% CI for d = -0.398 to -0.016).

Chapter I II III IV V VI VII VIII

35

Figure 3.3. Course for the Slalom Sprint and Dribble Test (SlalomSDT).

Interval endurance capacity was measured by using the Interval Shuttle Run Test (ISRT; Lemmink et al., 2000; Lemmink and Visscher, 2003) (see Figure 3.4). The ISRT is another sports specific field test that consists of intervals at a work to rest ratio of 2:1, turns at 20 m and an increasing running velocity. Players are required to run back and forth on a 20-m course with pylons positioned 3 m before lines marked out for the turns. The frequency of the sound signals on a pre-recorded compact disc increases in such a way that running speed is increased by 1 km·h-1 every 90 s from a starting speed of 10 km·h-1 and by 0.5 km·h-1 every 90 s from a starting speed of 13 km·h-1. Each 90-s period is divided into two 45-s periods in which players run for 30 s and walk for 15 s. Periods of running and walking are announced on the pre-recorded compact disc. During the periods of walking, players have simply to walk back and forth to the 8-m line. Players are instructed to complete as many runs as possible. The test stops when the player is unable to maintain the required pace (i.e. more than 3 m before the 20-m lines on two consecutive audio signals) or feels unable to complete the run. The number of fully completed 20-m runs is recorded as the test score. Players have to carry a hockey stick during the test. In previous research, this test has been shown to be reliable and sensitive for differences in level of performance (Lemmink et al., 2000; Lemmink et al., 2004b).

36

Figure 3.4. Course for the Interval Shuttle Run Test (ISRT).

Technical characteristics All players performed two field tests to determine three technical characteristics: peak shuttle dribble performance, dribble performance in a repeated shuttle run and performance in a slalom dribble. Peak shuttle dribble performance and dribble performance in a repeated shuttle run were measured using the ShuttleSDT (see Figure 3.2). In performing the test, players had to keep control of the ball while carrying out the 30-m sprint three times. While turning, players had to cross each turning line with both feet and the ball. Performance in the peak shuttle dribble is the time covered by the fastest of three 30-m dribbles; dribble performance in a repeated shuttle run is the total time covered by all three 30-m dribbles.

Relative as well as absolute test-retest reliability was shown for the dribbling part of the ShuttleSDT (peak shuttle dribble performance: ICC = 0.91 and 95% CI for d = -0.305 to 0.035; dribble performance in a repeated shuttle run: ICC = 0.89 and 95% CI for d = -0.840 to 0.494).

Slalom dribble performance was measured using the SlalomSDT (see Figure 3.3). In performing the test, players had to maintain control of the ball while performing the 30-m sprint with twelve 120-degree turns. Absolute test-retest reliability was shown for the dribbling part of the SlalomSDT (95% CI for d = -0.988 to 0.256), whereas in terms of relative reliability there was some evidence of systematic error (ICC = 0.78).

Chapter I II III IV V VI VII VIII

37

Tactical characteristics The trainers evaluated the tactical characteristics of the players. For this purpose, each of the 12 trainers filled out the ‘Tactics in Sports’ questionnaire to give their opinion about three tactical characteristics of each player: general tactics, tactics for possession of the ball and tactics for non-possession of the ball.

The ‘Tactics in Sports’ questionnaire is based on two pilot studies (Elferink-Gemser et al., internal publication 2001). In the first study, 20 highly qualified Dutch field hockey trainers established the most important tactical qualities for successful field hockey players to determine the categories of tactical qualities in the questionnaire. The qualities mentioned by the trainers can be divided into three categories. Category 1 contains general tactics, shifting in tasks from when the team does not possess the ball to when the team does possess the ball, and vice versa. Category 2 contains tactical qualities for when the team possesses the ball: positioning, overview and anticipation. Category 3 contains tactical qualities for when the team does not possess the ball: man-to-man defence, zone defence and interception.

In the second pilot study, six trainers evaluated 88 elite and sub-elite youth players using the ‘Tactics in Sports’ questionnaire, designed as a 6-point Likert scale ranging from ‘low/ moderate’ to ‘excellent’. The trainers were instructed to compare each player with top players in the relevant age category. Test-retest reliability for this checklist on tactical qualities was shown (Pearson correlation coefficient: Category 1: Z = 0.84, p < 0.01; Category 2: Z = 0.85; p < 0.01; Category 3: Z = 0.88, p < 0.01). The discriminatory power was obtained by comparing elite with sub-elite youth players. Scores on tactical qualities differed significantly for different performance levels (Category 1: Z = -3.954, p < 0.01; Category 2: Z = -5.084, p < 0.01; Category 3: Z = -6.622, p < 0.01), which means that the elite youth players were judged to be better than the sub-elite players.

Psychological characteristics All players filled in a sports specific questionnaire, the Dutch youth version of the Psychological Skills Inventory for Sports (PSIS) (Mahoney et al., 1987; Pennings et al., 1992; Bakker, 1995; Companjen and Bakker, 2003). The PSIS was developed for directly assessing an athlete’s psychological skills relevant to athletic training and exceptional performance. It assesses motivation, confidence, anxiety control, mental preparation, team emphasis and concentration. Internal consistency coefficients of all scales ranged from 0.68 for team emphasis to 0.81 for confidence. Pearson correlation coefficients for test-retest reliability ranged from 0.64 for team emphasis to 0.79 for mental preparation. The questionnaire contains forty-four 5-point Likert-type questions. A high score on the scale corresponds to the

38

psychological skill being present to a large extent. The maximum mean score on each scale is 5, and the minimum is 1. Data analysis Mean scores and standard deviations were calculated for each variable for the different sub-groups according to the five categories of performance characteristics (anthropometric, physiological, technical, tactical and psychological). A multivariate analysis of covariance (MANCOVA) was then carried out (factors of performance level and gender) to determine the effect of performance level and gender on the dependent variables in each category of characteristics. Since the relationship between test items may change with growth and development, age in years was considered a covariate. In this way, each variable was adjusted for age.

Where appropriate, analyses of covariances on each dependent variable were conducted as follow-up tests to the MANCOVA. To correct for multiple tests and thereby keep the overall alpha level below 0.05, the Bonferroni method was used.

Finally, all dependent variables were analysed together to determine which combination of measures best discriminated between the elite and sub-elite youth players. A stepwise discriminant function analysis was used in which level of performance was the dependent variable. Besides performance characteristics, age and gender were considered independent variables. An alpha of 0.05 was adopted for all tests of significance.

3.3 Results

Table 3.2 presents means and standard deviations of the multidimensional performance characteristics.

Tab

le 3

.2.

Mea

n sc

ores

(sd)

of a

nthr

opom

etric

, phy

siol

ogic

al, t

echn

ical

, tac

tical

and

psy

chol

ogic

al c

hara

cter

istic

s for

tale

nted

you

th fi

eld

hock

ey

pl

ayer

s cla

ssifi

ed b

y ge

nder

and

leve

l of p

erfo

rman

ce.

Fem

ale

yout

h pl

ayer

s M

ale

yout

h pl

ayer

s

Elite

pla

yers

n

= 17

Su

b-el

ite p

laye

rs

n =

46

Elite

pla

yers

n

= 21

Su

b-el

ite p

laye

rs

n =

42

Ant

hrop

omet

ric

char

acte

rist

ics

Leng

th (m

) B

ody

mas

s (kg

) %

Bod

y Fa

t

1.65

(0.0

1)

54.8

5 (8

.09)

21

.51

(5.5

7)

1.67

(0.0

1)

56.2

2 (7

.35)

21

.77

(5.1

6)

1.69

(0.0

1)

55.0

9 (8

.74)

9.

61 (2

.82)

1.73

(0.0

1)

59.6

9 (1

2.5)

8.

79 (3

.89)

Ph

ysio

logi

cal c

hara

cter

istic

s

Pe

ak sh

uttle

sprin

t per

form

ance

30m

(s)

Rep

eate

d sh

uttle

sprin

t per

form

ance

3x3

0m (s

) Sl

alom

sprin

t per

form

ance

30m

(s)

Inte

rval

end

uran

ce c

apac

ity (r

uns o

f 20m

)

8.82

(0.3

3)

27.1

5 (1

.01)

15

.05

(0.6

9)

55.7

6 (1

0.97

)

8.95

(0.4

0)

27.6

9 (1

.29)

15

.09

(0.7

9)

51.5

7 (1

2.72

)

8.52

(0.4

5)

26.3

1 (1

.43)

14

.52

(0.7

4)

73.0

0 (2

5.73

)

8.45

(0.3

5)

26.2

3 (1

.09)

14

.61

(0.6

5)

69.4

4 (2

1.50

) T

echn

ical

cha

ract

eris

tics

Peak

shut

tle d

ribbl

e pe

rfor

man

ce 3

0m (s

) D

ribbl

e pe

rf.in

repe

ated

shut

tle ru

n 3x

30m

(s)

Perf

orm

ance

in a

slal

om d

ribbl

e 30

m (s

)

10.1

3 (0

.60)

32

.26

(2.0

6)

19.1

8 (1

.83)

10.4

2 (0

.58)

33

.91

(3.2

8)

20.1

9 (4

.98)

9.72

(0.5

1)

30.2

2 (1

.75)

17

.43

(1.1

4)

9.83

(0.5

9)

30.8

3 (1

.98)

18

.48

(2.2

3)

Tac

tical

cha

ract

erist

ics

Gen

eral

tact

ics (

1-6)

Ta

ctic

s (po

sses

sion

of t

he b

all)

(1-6

) Ta

ctic

s (no

n-po

sses

sion

of t

he b

all)

(1-6

)

4.18

(1.0

7)

4.45

(0.7

8)

4.10

(0.6

6)

3.33

(0.7

9)

3.32

(0.5

4)

3.52

(0.6

3)

4.38

(1.0

7)

4.38

(0.8

5)

4.33

(0.7

6)

3.46

(1.0

3)

3.53

(0.8

4)

3.75

(0.7

9)

Psyc

holo

gica

l cha

ract

erist

ics

Mot

ivat

ion

(1-5

) C

onfid

ence

(1-5

) A

nxie

ty C

ontro

l (1-

5)

Men

tal P

repa

ratio

n (1

-5)

Team

Em

phas

is (1

-5)

Con

cent

ratio

n (1

-5)

4.66

(0.2

8)

3.85

(0.5

3)

3.85

(0.5

1)

2.28

(0.6

4)

3.53

(0.4

9)

3.73

(0.3

5)

4.09

(0.5

7)

3.46

(0.5

5)

3.85

(0.5

4)

2.05

(0.8

3)

3.45

(0.4

6)

3.42

(0.5

6)

4.55

(0.3

2)

3.97

(0.6

9)

4.04

(0.4

8)

2.22

(0.6

4)

3.52

(0.4

8)

3.46

(0.3

9)

4.29

(0.4

5)

3.95

(0.6

1)

3.94

(0.6

4)

2.31

(0.7

5)

3.54

(0.4

3)

3.46

(0.7

1)

40

The results of the MANCOVA showed a significant main effect for performance level (see Table 3.3). The univariate analyses of covariance revealed significant differences between elite and sub-elite youth players for two physiological variables (peak shuttle sprint performance [F (1,113) = 3.937, p < 0.05] and repeated shuttle sprint performance [F (1,113) = 7.498, p < 0.01]), for three technical variables (peak shuttle dribble performance [F (1,113) = 11.578, p < 0.01], dribble performance in a repeated shuttle run [F (1,113) = 11.111, p < 0.01] and performance in a slalom dribble [F (1,113) = 4.822, p < 0.05]), for three tactical variables (general tactics [F (1,113) = 20.592, p < 0.001], tactics for possession of the ball [F (1,113) = 48.051, p < 0.001] and tactics for non-possession of the ball [F (1,113) = 21.822, p < 0.001]) and for one psychological variable (motivation [F (1,113) = 20.916, p < 0.001]). In all comparisons, the elite youth players scored better than the sub-elite youth players. However, after correction for multiple tests, differences between the two groups of players were no longer significant for physiological variables and performance in the slalom dribble. No differences were found between the two groups for any of the anthropometric variables. Table 3.3. Results of MANCOVA for performance level, gender and performance level x gender.

Wilks’

lambda

F-value Hypothesis

df

Error

df

p-value

Performance level 0.550 4.084 19 95 < 0.001

Gender 0.192 21.011 19 95 < 0.001

Performance level x Gender 0.837 0.972 19 95 0.500

Besides a main effect on performance level, a significant main effect was found for gender. The univariate analyses of covariance showed significant differences between female and male players for anthropometric, physiological and technical variables but not for any of the tactical or psychological variables, except for confidence. Overall, males scored better than females. No interaction effects were found between performance level and gender.

Significant differences were also found for age, indicating the relevance of age as a covariate. Scores improved with age (Wilks’ lambda = 0.363, F = 8.618, Hypothesis df = 19, Error df = 95, p < 0.001) on the anthropometric, physiological and technical variables, but not on any tactical or psychological variables.

The results of the stepwise discriminant analysis are presented in Table 3.4. The model predicted that a combination of four variables would successfully discriminate between the elite and sub-elite youth players. These measures were selected in the following order of importance: tactics for possession of the ball (0.716), age (-0.518), motivation (0.463) and

Chapter I II III IV V VI VII VIII

41

performance in a slalom dribble (-0.310). Tactics for possession of the ball and motivation had a positive sign because higher scores denote better performance. With performance in a slalom dribble, the sign was negative since here a lower score (i.e., less time needed for the test) indicates better performance. The negative sign of age can be explained by the mean age of the elite youth players, which was lower than that of the sub-elite youth players.

Table 3.4. Summary of stepwise discriminant analysis: variables entered/removed.

Wilks’ Lambda

Exact F

Step Entered Statistic df1 df2 df3 Statistic df1 df2 p-value

1 Tactics (possession of ball) 0.683 1 1 116 53.760 1 116 < 0.001

2 Age 0.626 2 1 116 34.293 2 115 < 0.001

3 Motivation 0.574 3 1 116 28.205 3 114 < 0.001

4 Performance in a

slalom dribble

0.551 4 1 116 23.056 4 113 < 0.001

Note: At each step, the variable that minimizes the overall Wilks’ lambda is entered. Maximum number of steps is 42. Minimum partial F to enter is 3.84. Maximum partial F to remove is 2.71. F level, tolerance, or VIN insufficient for further computation.

Variables Tolerance F to remove Wilks’

lambda

Step

1 Tactics (possession of ball) 1.000 53.760

2 Tactics (possession of ball) 0.999 48.624 0.891

Age 0.999 10.447 0.683

3 Tactics (possession of ball) 0.990 36.307 0.757

Age 0.989 11.659 0.633

Motivation 0.980 10.414 0.626

4 Tactics (possession of ball) 0.990 33.409 0.713

Age 0.938 14.401 0.621

Motivation 0.966 11.590 0.607

Performance in a slalom dribble 0.941 4.793 0.574

The average squared canonical correlation was 0.670. This means that, knowing the scores on tactics for possession of the ball, age, motivation and performance in a slalom dribble,

42

estimation of the percent variance accounted for is 67%. When group membership is predicted from these four variables, 82.8% of the original grouped players are classified correctly. The other variables provided no additional information when discriminating further between the two groups of players. 3.4 Discussion

In most studies of the relation between multidimensional performance characteristics and level of performance, elite players have been compared with non-elite players. To gain more insight into the characteristics of “tomorrow’s stars”, it seems that the critical focus should be on talented youth players already detected. Importantly, most of today’s top performers played in a youth selection team when they were younger.

Not all young field hockey players who are considered talented will ultimately make it to the top, as only the very best will achieve elite status in adulthood. One cannot predict with certainty which talented youngsters will become top players, but elite players, who play in selection teams of the Dutch Field Hockey Association and of their club, have a better chance of reaching the top than sub-elite players, who play only in their club’s selection team. This is why we compared between elite youth players and sub-elite youth players.

Our results showed that the group of talented players as a whole obtained high scores on all tests. However, the elite players scored better than the sub-elite players on variables from three categories of characteristics (technical, tactical and psychological). This was the case for young talented female as well as male players. No significant differences were found between the performance groups on any of the anthropometric or physiological variables.

Researchers who have compared elite with non-elite players have reported differences in anthropometric and physiological characteristics (Jankovic et al., 1997; Panfil et al, 1997; Janssens et al., 1998; Malina et al., 2000), but comparisons between talented field hockey players seem to suggest similar scores on these characteristics. Evidently, at the elite level, differences between players are less related to physical and physiological characteristics. This finding is in accordance with the results of a study by Franks et al. (1999) on young soccer players, in which it was not possible to discriminate between players at the highest level on the basis of their physical and physiological profiles.

Elite youth field hockey players scored better than the sub-elite players on the tests for both peak shuttle dribble performance and dribble performance in a repeated shuttle run. Sprinting repeatedly over short distances with rapid changes of direction while maintaining control of the ball is an important attribute for these players. In their multidisciplinary approach to talent identification in soccer, Reilly et al. (2000) found that elite soccer players

Chapter I II III IV V VI VII VIII

43

scored better on dribbling tasks than sub-elite soccer players. It does seem that technical qualities remain important at the elite level. Van Rossum and Gagné (1994) also confirmed the importance of technique in field hockey. In their study, Dutch coaches considered technique (motor skill) one of the most important factors affecting the performance of top-level field hockey players.

For all tactical variables, again the elite youth players scored better than the sub-elite players. This was the case for general tactics as well as tactics for possession of the ball and non-possession of the ball. Knowing when to perform the right action, an attribute also referred to as ‘game intelligence’, is crucial for a successful career in field hockey. In a study of perceptual skill in soccer, Williams (2000; p.737) stated that ‘decisions in a match have to be made under pressure with opponents trying to restrict the “time” and “space” available to perform’. A key problem, however, is how to measure this tactical insight. Considering the complex and rapidly changing environment in field hockey matches, it is very difficult to grade players’ tactical insight objectively. The trainers in this study were highly qualified and work with the club selection players throughout the season during training and match-play. These trainers are considered to be experts in the field and their opinion was used to gauge the tactical insight of the players. Reilly et al. (2000) indicated that an interdisciplinary scientific approach has to be combined with the accumulated know-how of experts such as trainers, coaches and scouts.

The only psychological variable for which the elite youth players scored better than the sub-elite players was motivation. According to Ericsson et al. (1993) and Ericsson (1996), expert performance is the end result of individual’s prolonged efforts to improve performance, and since engagement in deliberate practice is not inherently motivating, commitment on the part of the performer is required.

Reilly et al. (2000) indicated that measures of agility, speed, motivation orientation and perceptual skill were the most important indicators of talent in soccer. These findings are in line with Deshaies et al. (1979), who made clear that anaerobic power, speed, perceptual skill and motivation successfully discriminated between elite and sub-elite ice-hockey players. In our study, a stepwise discriminant function analysis was used to determine which combination of measures distinguished most clearly between the two groups of talented field hockey players. The analyses revealed that the groups could be discriminated on the basis of four variables. The most discriminating measures were tactics for possession of the ball (consisting of positioning, overview and anticipation), motivation and performance in a slalom dribble. Age discriminated between both performance groups, indicating that the elite players were younger than the sub-elite players. Although the elite players tended to be shorter and lighter

44

than the sub-elite players, one cannot rule out that the most mature children were performing best at this age, since no maturity measures were taken.

The results from this study suggest that talented players cannot be distinguished from each other on the basis of the same performance characteristics that discriminate between elite and non-elite players. At the elite level, differences between players are less related to physical and physiological characteristics, and more to tactics, motivation and specific technical skills in field hockey. It is also interesting that the elite players were younger than the sub-elite players. One explanation is that the elite players started playing earlier and so had more experience than the sub-elite players (see Table 3.1). It is also possible that the time needed to master the important characteristics differs between elite and sub-elite players. Elite players may need less time to develop better performance characteristics. Not only in the guidance of young talented players to the top, but also in the detection of talented players, more attention has to be paid to tactical qualities, motivation and specific technical skills along with the time needed to master these skills.

Chapter I II III IV V VI VII VIII

45

References

Bakker, F. (1995). Ontwikkeling van tests op het terrein van de sportpsychologie in Nederland en Vlaanderen. [Development of tests for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 6, 3-15.

Bhanot, J.L. and Sidhu, L.S. (1983). Maximal anaerobic power in Indian national hockey players. British Journal of Sports Medicine, 17, 34-39.

Bloom, B.S. (1985). Generalisations about talent development. In Developing Talent in Young People (edited by B.S. Bloom), pp. 507-549. New York: Ballantine Books.

Boyle, P.M., Mahoney, C.A., and Wallace, W.F. (1994). The competitive demands of elite male field hockey players. Journal of Sports Medicine and Physical Fitness, 34, 235-241.

Carlson, R. (1988). The socialisation of elite tennis players in Sweden: An analysis of the players’ backgrounds and development. Society of Sport Journal, 5, 241-256.

Carlson, R. (1993). The path to the national level in sports in Sweden. Scandinavian Journal of Medicine & Science in Sports, 3, 170-177.

Companjen, T. and Bakker, F. (2003). Overzicht van sportpsychologische vragenlijsten in Nederland en Vlaanderen. [Outline of questionnaires for the field of sport psychology in the Netherlands and Belgium). Sportpsychologie Bulletin, 14, 34-41.

Côté, J. (1999). The influence of the family in the development of talent in sport. The Sports Psychologist, 13, 395-417.

Deshaies, P., Pargman, D., and Thiffault, C. (1979). A psychobiological profile of individual performance in junior hockey players. In Psychology of Motor Behaviour and Sport – 1978 (edited by G.C. Roberts and K.M. Newell), pp. 36-50. Champaign, IL: Human Kinetics.

Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. (2001). Ontwikkeling van de ‘Tactiek in Sport’ vragenlijst. [Development of the ‘Tactics in Sports’ questionnaire]. Internal publication, Center for Human Movement Sciences, University of Groningen, The Netherlands.

Ericsson, K.A. (1996). The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games. Mahwah, NJ: Lawrence Erlbaum Associates.

Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406.

Franks, A., Williams, A.M., Reilly, T., and Nevill, A. (1999). Talent identification in elite youth soccer players: Physical and physiological characteristics. Journal of Sports Sciences, 17, 812.

Helsen, W.F., Hodges, N.J., Van Winckel, J., and Starkes, J.L. (2000). The roles of talent, physical precocity and practice in the development of soccer expertise. Journal of Sports Sciences, 18, 727-736.

Hoare, D.G. and Warr, C.R. (2000). Talent identification and women’s soccer: An Australian experience. Journal of Sports Sciences, 18, 751-758.

Howe, M.J.A., Davidson, J.W., and Sloboda, J.A. (1998). Innate talents: Reality or myth. Behavioral and Brain Sciences, 21, 399-442.

Jankovic, S., Matkovic, B.R., and Matkovic, B. (1997). Functional abilities and process of selection in soccer. Communication to the 9th European Congress on Sports Medicine, Porto, Portugal, 23-26 September.

Janssens, M., Van Renterghem, B., Bourgois, J., and Vrijens, J. (1998). Physical fitness and specific motor performance of young soccer players aged 11-12 years. Communication to the 2nd Annual Congress of the European College of Sport Science. Journal of Sports Sciences, 16, 434-435.

Kroll, W. (1970). Current strategies and problems in personality assessment of athletes. In Psychology of Motor Learning (edited by L.E. Smith) , pp. 349-367. Chicago: Athletic Institute.

Lemmink, K.A.P.M. and Visscher, C. (2003). The relationship between the Interval Shuttle Run Test and maximal oxygen uptake in soccer players. Journal of Human Movement Studies, 45, 219-232.

Lemmink, K.A.P.M., Elferink-Gemser, M.T., and Visscher, C. (2004a). Evaluation of the reliability of two field hockey-specific sprint and dribble tests in young field hockey players. British Journal of Sports Medicine, 38, 138-142.

46

Lemmink, K.A.P.M., Verheijen, R., and Visscher, C. (2004b). The discriminative power of the Interval Shuttle Run Test and the Maximal Multistage Shuttle Run Test for playing level of soccer. Journal of Sports Medicine and Physical Fitness, 44, 233-239.

Lemmink, K.A.P.M., Dolleman, G., Verheijen, R., and Visscher, C. (2000). Interval Sprint Test en Interval Shuttle Run Test – betrouwbaarheid en discriminerend vermogen van twee nieuwe voetbaltests. [Interval Sprint Test and Interval Shuttle Run Test – reliability and discriminative power of two new tests for soccer players]. Geneeskunde en Sport, 33, 39-48.

Lothian, F. and Farrally, M. (1994). M. A time-motion analysis of women’s hockey. Journal of Human Movement Studies, 26, 255-265.

Mahoney, M.J., Gabriel, T.J., and Perkins, T.S. (1987). Psychological skills and exceptional performance. The Sport Psychologist, 1, 181-199.

Malina, R.M., Pena Reyes, M.E., Eisenmann, J.C., Horta, L., Rodrigues, J., and Miller, R. (2000). Height, mass and skeletal maturity of elite Portuguese soccer players aged 11-16 years. Journal of Sports Sciences, 18, 685-693.

Morris, T. (2000). Psychological characteristics and talent identification in soccer. Journal of Sports Sciences, 18, 715-726.

Nunez, C., Callagher, D., Visser, M., Pi-Sunyer, F.X., Wang, Z., and Heymsfield, S.B. (1997). Bioimpedance analysis: evaluation of leg-to-leg system based on pressure contact footpad electrodes. Medicine and Science in Sports and Exercise, 29, 524-531.

Panfil, R., Naglak, Z., Bober, T., and Zaton, E.W.M. (1997). Searching and developing talents in soccer: A year of experience. In Proceedings of the 2nd Annual Congress of the European College of Sport Science (edited by J. Bangsbo, B. Saltin, H. Bonde, Y. Hellsten, B. Ibsen, M. Kjaer, and G. Sjogaard), pp. 649-650. Copenhagen: HO + Storm.

Pennings, A., Den Hollander, B.J., Kunst, H., and Schmikli, S. (1992). Een psychometrische evaluatie van de Psychological Skills Inventory for Sports (PSIS) van M.J. Mahoney. (A psychometric evaluation of the psychological skills inventory for Sports (PSIS) of M.J. Mahoney). Master’s thesis, University of Utrecht, The Netherlands.

Pienaar, A.E., Spamer, M.J., and Steyn, H.S. (1998). Identifying and developing rugby talent among 10-year-old boys: A practical model. Journal of Sports Sciences, 16, 691-699.

Régnier, G., Salmela, J.H., and Russell, S.J. (1993). Talent detection and development in sport. In A Handbook of Research on Sports Psychology (edited by R. Singer, M. Murphey, and L.K. Tennant), pp. 290-313. New York: Macmillan.

Reilly, T. and Borrie, A. (1992). Physiology applied to field hockey. Sports Medicine, 14, 10-26. Reilly, T. and Seaton, A. (1990). Physiological strain unique to field hockey. Journal of Sports

Medicine and Physical Fitness, 30, 142-146. Reilly, T., Williams, A.M., Nevill, A., and Franks, A. (2000). A multidisciplinary approach to talent

identification in soccer. Journal of Sports Sciences, 18, 695-702. Smith, M.D. and Chamberlin, C.J. (1992). Effects of adding cognitively demanding tasks on soccer

skill performance. Perceptual and Motor Skills, 75, 955-961. Van Rossum, J.H.A. and Gagné, F. (1994). Rankings of predictors of athletic performance by top level

coaches. European Journal for High Ability, 5, 68-78. Visscher, C., Elferink-Gemser, M. T., and Lemmink, K. A. P. M. (2004). The role of parental support

in sports success of talented young Dutch athletes. In Children and Youth in Organized Sports (edited by M. Coelho e Silva and R. M. Silva), pp. 123-135. Coimbra: Coimbra University Press.

Williams, A.M. (2000). Perceptual skill in soccer: Implications for talent identification and development. Journal of Sports Sciences, 18, 737-750.

Williams, A.M. and Davids, K. (1995). Declarative knowledge in sport: A by-product of experience or a characteristic of expertise? Journal of Sport and Exercise Psychology, 17, 259-275.

Williams, A.M. and Reilly, T. (2000). Talent identification and development in soccer. Journal of Sports Sciences, 18, 657-667.

Chapter I II III IV V VI VII VIII

47

Chapter IV

Multidimensional performance characteristics and performance level in talented youth field hockey players: A longitudinal study Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. Journal of Sports Sciences (pending minor revisions) Acknowledgements:

This study has been supported by a grant of the Dutch National Olympic Committee NOC*NSF. The authors thank all players, trainers, and staff of the field hockey clubs HC ’s Hertogenbosch and HC Rotterdam for their cooperation.

50

Abstract

To reveal performance characteristics, which may have power for predicting future elite field hockey players, we made a comparison between 30 elite and 35 sub-elite youth players in terms of anthropometric, physiological, technical, tactical and psychological characteristics measured on three occasions, each separated by a time interval of one year. Mean age of the players on the first measurement was 14.2 years (sd = 1.1). Repeated measures analyses of covariance with factors of performance level and measurement, and with age as a covariate, showed that the elite players scored better than the sub-elite players on technical and tactical variables. Female elite youth players also scored better on interval endurance capacity, motivation and confidence. Future elite players seem to excel in tactical skills by the age of 14 already. They also stand out in specific technical skills and develop these together with the interval endurance capacity better than sub-elite youth players in the two subsequent years. It will be interesting to follow these players until they reach elite status in adulthood to verify these conclusions.

Chapter I II III IV V VI VII VIII

51

4.1 Introduction

Field hockey is an important sport in the Netherlands. The high level of Dutch field hockey players is recognised world-wide. To reach elite level, players have to start their intensive and time-consuming training at an early age (Alabin et al., 1980; Hahn, 1990). According to Ericsson et al. (1993) and Ericsson (1996), expert performance is the end result of an individual’s prolonged efforts to improve performance, and since engagement in deliberate practice is not inherently motivating, commitment from the performers is required. Notwithstanding the efforts of many players, only a few will become successful in the end. What is it that characterises the ones who succeed? This question forms the background for the present paper in which it is attempted to deliver some preliminary answers to this intriguing question.

A number of studies have focused on performance-related characteristics of elite field hockey players. Unique requirements of the game include dribbling the ball and moving quickly in a semi-crouched posture (Reilly and Seaton, 1990). Analysis of the physiological cost and energy expenditure of playing hockey has placed it in the category of heavy exercise (Ghosh et al., 1991; Reilly and Borrie, 1992; Boyle et al., 1994; Lothian and Farrally, 1994; Aziz et al., 2000). Two decades ago, Hargraeves (1984) already showed high VO2max values for British Olympians, and Withers et al. (1977) did the same for Australian nationals. The intermittent running, accelerating and decelerating increases the overall effort needed in field hockey (Patel et al., 2002).

Researchers who focus on talent development in sports often acknowledge that a world-class performance is the result of several factors (Deshaies et al., 1979; Régnier, 1993; Pienaar et al., 1998; Reilly et al., 2000). According to Williams and Reilly (2000), research should adopt a multidisciplinary approach to identify talent. Burwitz et al. (1994) also recommend interdisciplinary performance-related sports science research. Thereby, Atkinson and Nevill (2001) have argued that more research should involve sports-specific dependent variables. We applied a multidisciplinary design in a recent study on talented Dutch field hockey players. It was shown that a combination of technical, tactical and psychological characteristics distinguished best between elite and sub-elite youth players (Elferink-Gemser et al., 2004).

In most previous research, a cross-sectional rather than longitudinal approach has been applied. However, to improve understanding of the factors that contribute to expert performance, players should be monitored over a prolonged period of time (Williams and Reilly, 2000). The goal of the present study is to reveal performance characteristics, which may have power for predicting future elite field hockey players. Within the pool of talented players, a comparison has been made between elite and sub-elite youth players in terms of

52

anthropometric, physiological, technical, tactical and psychological characteristics measured on three occasions, each separated by a time interval of one year. Questions to be answered are: On which performance characteristics do elite youth players score better than their sub-elite counterparts? How do elite and sub-elite youth players develop their performance characteristics over a period of two years, and is there a difference between elite and sub-elite youth players concerning this development?

4.2 Methods

Participants One hundred and twenty-six talented field hockey players in the 12-16 age-bracket (mean age = 13.9, sd = 1.3) participated in a cross-sectional study on the relation between multidimensional performance characteristics and performance level (Elferink-Gemser et al., 2004). All participants were part of a talent development program of a field hockey club of national prestige, and were playing at the highest level for their age category. Within this group, a distinction was made between 38 elite and 88 sub-elite youth players. In contrast to sub-elite players, elite players train and play in a youth selection team of the Dutch Field Hockey Association (KNHB).

From the initial 126 field hockey players, 85 were tested again one year later (t2), and after two years (t3) 65 players were tested for the third time. Table 4.1 presents the number of participants at t1, t2 and t3 divided by performance level and gender. Thirty players left the study because they were no longer part of the talent development program. They continued playing field hockey but fell back to club performance level. Reasons for players who were still playing on a national level for leaving the study were not being able to attend the measurements because of illness or injuries or because of lacking time or transportation.

Table 4.1. Number of participants at t1, t2 and t3 classified by performance level, gender and number

of players that left the study.

Female players Male players Players that left the study

Elite Sub-elite Elite Sub-elite Elite Sub-elite Club level

Measurement

t1 n = 126 17 46 21 42

t2 n = 85 17 25 20 23 1 18 22

t3 n = 65 15 18 15 17 7 5 8

Chapter I II III IV V VI VII VIII

53

The proportion of female and male participants on the three measurements was about the same. Two female and three male players of the total group (n = 65) were elite players at t1 but sub-elite at t3. One female player was sub-elite at t1 but elite at t3. The other players remained either elite or sub-elite from t1 through t3. Table 4.2 displays the general characteristics of the participants concerning age, field hockey experience, training hours and match play frequency. Table 4.2. Mean (sd) scores of general characteristics at t3 concerning age, field hockey experience,

training hours and matches per week for talented youth field hockey players classified by gender and performance level.

Female youth players Male youth players

Elite players n = 15

Sub-elite players n = 18

Elite players n = 15

Sub-elite players n = 17

15.71 (1.01) 16.40 (1.28) 16.01 (1.00) 16.48 (1.08) Age

(years)

8.43 (2.19) 9.06 (1.69) 8.87 (1.51) 8.80 (2.31) Field hockey experience

(years)

5.13 (1.65) 4.28 (1.70) 5.18 (0.57) 4.70 (0.31) Field hockey training

(hours per week)

8.37 (3.91) 5.64 (2.31) 8.15 (3.64) 7.94 (3.85) Total training

(hours per week)

1.07 (0.26) 1.00 (0.00) 1.17 (0.36) 1.00 (0.00) Matches per week

54

Procedure All players were informed about the procedure of the study before giving their informed consent to participate. The clubs and trainers gave their permission for this study. The procedures were in accordance with the standards of the local medical ethics committee of the University of Groningen. The players completed the tests at the end of the competitive 2000-2001 field hockey season (t1), at the end of the 2001-2002 season (t2) and at the end of the 2002-2003 season (t3). Ambient temperature, humidity and wind conditions were documented. Measurements for each player took place according to five categories of performance characteristics: anthropometric, physiological, technical, tactical and psychological. Field tests were executed on synthetic field hockey playing surfaces (water-based pitches). The employed test procedures are described in detail elsewhere (Elferink-Gemser et al., 2004). Anthropometric characteristics Anthropometric measurements were length (m), body mass (kg) and percentage of body fat. The latter was estimated by means of leg-to-leg bioelectrical impedance (BIA) analysis (Valhalla BIA, Valhalla, Inc., San Diego, CA). This method proved to be reliable for measuring body fat percentage, and results correlated highly with body fat percentage as measured with underwater weighing and dual energy X-ray absorptiometry (Nunez et al., 1997). Physiological characteristics All players performed three field tests to determine four physiological characteristics. These characteristics included peak shuttle sprint performance, repeated shuttle sprint performance, slalom sprint performance and interval endurance capacity. In all tests, players had to carry their hockey stick. Peak shuttle sprint and repeated shuttle sprint performance were measured by means of the Shuttle Sprint and Dribble Test (ShuttleSDT), in which players had to run three 30-m sprints with 180-degree turning points. Each 30-meter sprint consists of 5 m to-and-fro, directly followed by 10 m to-and-fro. Peak shuttle sprint performance is indicated by the time covered in the fastest of three 30-m sprints, whereas repeated shuttle sprint performance is the total time covered by all three 30-m sprints. Reliability proved to be satisfactory in young field hockey players (Lemmink et al., 2004a).

Slalom sprint performance was measured using the Slalom Sprint and Dribble Test (SlalomSDT), in which players have to sprint 30 m in a zigzag fashion with twelve 120-degree turns around cones placed 2 m apart from each other. Reliability for this test was supported (Lemmink et al., 2004a). Interval endurance capacity was measured with the Interval Shuttle Run Test (ISRT) (Lemmink et al., 2000; Lemmink and Visscher, 2003). The

Chapter I II III IV V VI VII VIII

55

ISRT is a field test that contains intervals at a work-rest ratio of 2:1, turning points at 20 m and an increasing running velocity. The number of fully completed 20-m runs is recorded as the test score. In previous research, this test has proven to be reliable and sensitive for differences in performance level (Lemmink, et al., 2000; Lemmink et al., 2004b; Lemmink et al., 2004c).

Technical characteristics All players performed two field tests to determine three technical characteristics. These characteristics included peak shuttle dribble performance, dribble performance in a repeated shuttle run and performance in a slalom dribble. Peak shuttle dribble performance as well as dribble performance in a repeated shuttle run were measured using the ShuttleSDT; performance in a slalom dribble was measured using the SlalomSDT. Players now had to keep control over the ball while performing the tests. Reliability of the dribbling part of both the ShuttleSDT and the SlalomSDT has been supported in young field hockey players (Lemmink et al., 2004a). Tactical characteristics The trainers evaluated the tactical characteristics of their players. For this purpose, each of the 12 trainers filled out the ‘Tactics in Sports’ questionnaire in order to give their opinion about three tactical characteristics of each player: general tactics, tactics for possession of the ball and tactics for non-possession of the ball. The trainers were instructed to compare each player with the top players in the same age category. In a previous study this questionnaire has proven to be reliable and sensitive for differences in performance level (Elferink-Gemser et al., 2001; 2004). Psychological characteristics All players filled in a sports-specific questionnaire: the Dutch Youth Version of the Psychological Skills Inventory for Sports (PSIS) (Mahoney et al., 1987; Elferink-Gemser et al., 2004). The PSIS was developed for directly assessing an athlete’s psychological skills relevant to athletic training and exceptional performance. It assesses the level of motivation, confidence, anxiety control, mental preparation, team emphasis and concentration. This questionnaire has proven to be reliable in previous research (Bakker, 1995; Companjen and Bakker, 2003).

56

Data analysis All data were analysed for male and female players separately using SPSS 10. According to the five categories of performance characteristics (anthropometric, physiological, technical, tactical and psychological), mean scores and standard deviations were calculated on measurements 1, 2 and 3 for the four different subgroups based on performance level (elite and sub-elite youth players) and gender.

Repeated measures analyses of covariance were used to examine group differences based on performance level together with differences in performance characteristics over time. Age was considered as covariate. The statistical techniques provide comparisons of the subgroups over time, taking into account the possible influence of age.

In the between-subjects analysis, a performance-level effect shows differences in average scores on measurements 1, 2 and 3 between elite and sub-elite players. In the within-subjects analysis, a measurement effect shows differences between scores on the three measurements. An interaction effect between performance level and measurement reveals differences between elite and sub-elite players that change as a function of time. An alpha of 0.05 was adopted for all tests of significance.

4.3 Results

Table 4.3 presents mean scores and standard deviations of the multidimensional performance characteristics for talented youth female field hockey players on the three measurements classified by performance level. Table 4.4 provides the same information for talented youth male field hockey players.

Tab

le 4

.3.

Mea

n sc

ores

(sd)

of a

nthr

opom

etric

, phy

siol

ogic

al, t

echn

ical

, tac

tical

and

psy

chol

ogic

al c

hara

cter

istic

s for

tale

nted

you

th fe

mal

e fie

ld

hock

ey p

laye

rs o

n m

easu

rem

ents

1, 2

and

3 c

lass

ified

by

perf

orm

ance

leve

l.

Fe

mal

e el

ite p

laye

rs

n =

15

Fem

ale

sub-

elite

pla

yers

n

= 18

t 1 t 2

t 3 t 1

t 2 t 3

Ant

hrop

omet

ric

char

acte

rist

ics

Leng

th (m

) B

ody

mas

s (kg

) %

Bod

y Fa

t

1.61

(0.0

8)

49.9

6 (7

.75)

18

.77

(6.1

1)

1.64

(0.0

7)

53.7

3 (7

.18)

19

.60

(5.6

6)

1.66

(0.0

5)

57.1

4 (6

.42)

20

.93

(5.9

4)

1.67

(0.0

5)

56.3

3 (7

.09)

22

.89

(7.1

1)

1.68

(0.0

5)

58.7

6 (7

.33)

24

.16

(5.8

5)

1.69

(0.0

4)

60.9

1 (6

.65)

21

.84

(6.1

5)

Phys

iolo

gica

l cha

ract

eris

tics

Peak

shut

tle sp

rint p

erfo

rman

ce 3

0m (s

) R

epea

ted

shut

tle sp

rint p

erf.

3x30

m (s

) Sl

alom

sprin

t per

form

ance

30m

(s)

Inte

rval

end

uran

ce c

apac

ity (r

uns o

f 20m

)

9.00

(0.3

0)

27.5

6 (0

.93)

15

.21

(0.8

5)

55.2

7 (1

2.08

)

8.64

(0.3

3)

26.7

0 (0

.88)

14

.93

(0.5

6)

61.2

0 (1

4.98

)

8.63

(0.2

6)

26.6

4 (0

.81)

14

.75

(0.7

1)

75.3

3 (1

7.16

)

9.08

(0.4

5)

28.0

8 (1

.51)

15

.36

(1.2

0)

49.3

3 (1

7.32

)

8.78

(0.4

0)

27.1

6 (1

.36)

15

.00

(0.9

7)

48.6

7 (1

2.51

)

8.72

(0.4

4)

26.8

7 (1

.35)

15

.03

(0.9

6)

53.8

9 (1

6.46

) T

echn

ical

cha

ract

eris

tics

Peak

shut

tle d

ribbl

e pe

rfor

man

ce 3

0m (s

) D

ribbl

e pe

rf. i

n re

p. sh

uttle

run

3x30

m (s

) Pe

rfor

man

ce in

a sl

alom

drib

ble

30m

(s)

10.3

6 (0

.58)

32

.86

(1.7

5)

19.6

9 (1

.78)

10.0

0 (0

.46)

31

.47

(1.5

8)

19.0

2 (2

.00)

9.72

(0.4

6)

30.8

8 (2

.26)

17

.61

(1.1

8)

10.5

6 (0

.66)

33

.90

(2.6

4)

20.0

8 (2

.72)

10.2

3 (0

.81)

32

.56

(3.4

7)

18.8

4 (1

.55)

9.99

(0.8

7)

31.6

6 (2

.87)

18

.72

(2.0

7)

Tac

tical

cha

ract

erist

ics

Gen

eral

tact

ics (

1-6)

Ta

ctic

s (po

sses

sion

of t

he b

all)

(1-6

) Ta

ctic

s (no

n-po

sses

sion

of t

he b

all)

(1-6

)

4.17

(1.1

0)

4.19

(0.8

0)

3.87

(0.6

5)

3.97

(0.9

3)

3.97

(0.7

2)

3.59

(0.5

2)

4.07

(0.5

9)

4.18

(0.6

0)

3.91

(0.3

7)

3.14

(0.9

0)

3.63

(0.6

6)

3.75

(0.7

1)

3.27

(1.0

3)

3.44

(0.8

6)

3.55

(0.6

6)

3.67

(0.9

1)

3.69

(0.7

5)

3.73

(0.6

4)

Psyc

holo

gica

l cha

ract

erist

ics

Mot

ivat

ion

(1-5

) C

onfid

ence

(1-5

) A

nxie

ty C

ontro

l (1-

5)

Men

tal P

repa

ratio

n (1

-5)

Team

Em

phas

is (1

-5)

Con

cent

ratio

n (1

-5)

4.65

(0.3

1)

3.68

(0.6

3)

3.89

(0.4

8)

2.18

(0.5

3)

3.54

(0.4

1)

3.59

(0.4

1)

4.57

(0.3

7)

3.27

(0.2

4)

3.94

(0.4

8)

2.24

(0.7

4)

3.52

(0.4

4)

3.65

(0.5

3)

4.41

(0.4

8)

3.68

(0.4

8)

3.81

(0.6

7)

2.34

(0.6

8)

3.53

(0.5

5)

3.52

(0.5

3)

4.08

(0.5

5)

3.37

(0.5

5)

3.92

(0.5

1)

1.91

(0.5

8)

3.54

(0.6

5)

3.64

(0.5

9)

4.32

(0.6

0)

3.17

(0.4

1)

3.89

(0.3

9)

2.19

(0.6

4)

3.52

(0.3

8)

3.64

(0.6

0)

4.10

(0.5

6)

3.21

(0.5

6)

3.90

(0.3

8)

2.00

(0.6

6)

3.44

(0.3

5)

3.52

(0.4

7)

Tab

le 4

.4.

Mea

n sc

ores

(sd)

of a

nthr

opom

etric

, phy

siol

ogic

al, t

echn

ical

, tac

tical

and

psy

chol

ogic

al c

hara

cter

istic

s for

tale

nted

you

th m

ale

field

hock

ey p

laye

rs o

n m

easu

rem

ents

1, 2

and

3 c

lass

ified

by

perf

orm

ance

leve

l.

M

ale

elite

pla

yers

n

= 15

M

ale

sub-

elite

pla

yers

n

= 17

t 1

t 2 t 3

t 1 t 2

t 3 A

nthr

opom

etri

c ch

arac

teri

stic

s

Le

ngth

(m)

Bod

y m

ass (

kg)

% B

ody

Fat

1.66

(0.0

6)

52.5

7 (8

.15)

9.

20 (2

.23)

1.72

(0.0

6)

58.1

1 (7

.71)

7.

55 (2

.09)

1.76

(0.0

8)

64.4

2 (8

.03)

7.

83 (1

.56)

1.69

(0.0

8)

54.3

9 (1

0.80

) 9.

51 (4

.64)

1.74

(0.0

8)

59.2

8 (9

.95)

8.

78 (5

.30)

1.77

(0.0

7)

63.9

3 (9

.28)

8.

65 (4

.50)

Ph

ysio

logi

cal c

hara

cter

istic

s

Pe

ak sh

uttle

sprin

t per

form

ance

30m

(s)

Rep

eate

d sh

uttle

sprin

t per

f. 3x

30m

(s)

Slal

om sp

rint p

erfo

rman

ce 3

0m (s

) In

terv

al e

ndur

ance

cap

acity

(run

s of 2

0m)

8.63

(0.4

0)

26.6

1 (1

.20)

14

.56

(0.7

0)

68.8

0 (2

7.76

)

8.61

(0.4

1)

26.2

1 (1

.12)

14

.81

(0.9

1)

79.0

7 (1

8.94

)

8.18

(0.2

9)

25.1

3 (0

.90)

14

.14

(0.5

4)

101.

07 (1

9.14

)

8.58

(0.3

5)

26.6

8 (1

.44)

14

.90

(0.7

1)

70.8

2 (2

2.19

)

8.61

(0.4

1)

26.7

1 (1

.55)

14

.76

(0.6

0)

82.2

9 (2

8.85

)

8.18

(0.3

6)

25.1

2 (1

.22)

14

.39

(1.0

0)

82.9

4 (2

6.07

) T

echn

ical

cha

ract

eris

tics

Peak

shut

tle d

ribbl

e pe

rfor

man

ce 3

0m (s

) D

ribbl

e pe

rf. i

n re

p. sh

uttle

run

3x30

m (s

) Pe

rfor

man

ce in

a sl

alom

drib

ble

30m

(s)

10.0

2 (0

.68)

30

.97

(1.8

9)

17.8

2 (1

.28)

9.38

(0.4

1)

29.4

0 (1

.55)

17

.82

(1.1

3)

9.06

(0.4

0)

28.4

5 (1

.18)

17

.26

(0.9

4)

9.91

(0.6

6)

30.9

9 (2

.41)

18

.95

(2.3

3)

9.49

(0.7

7)

29.9

6 (2

.55)

18

.55

(1.8

3)

9.36

(0.6

4)

29.1

2 (2

.51)

18

.36

(1.7

0)

Tac

tical

cha

ract

erist

ics

Gen

eral

tact

ics (

1-6)

Ta

ctic

s (po

sses

sion

of t

he b

all)

(1-6

) Ta

ctic

s (no

n-po

sses

sion

of t

he b

all)

(1-6

)

4.33

(0.9

0)

4.64

(0.8

3)

4.44

(0.7

2)

4.27

(0.7

0)

4.13

(0.6

2)

4.09

(0.6

8)

4.17

(0.7

9)

4.09

(0.6

5)

4.08

(0.5

0)

3.65

(0.7

9)

3.71

(0.6

1)

3.90

(0.4

8)

2.94

(0.8

3)

3.16

(0.7

5)

3.26

(0.6

0)

3.21

(0.5

9)

3.22

(0.8

2)

3.31

(0.7

0)

Psyc

holo

gica

l cha

ract

erist

ics

Mot

ivat

ion

(1-5

) C

onfid

ence

(1-5

) A

nxie

ty C

ontro

l (1-

5)

Men

tal P

repa

ratio

n (1

-5)

Team

Em

phas

is (1

-5)

Con

cent

ratio

n (1

-5)

4.52

(0.2

4)

3.94

(0.7

0)

4.08

(0.4

4)

2.34

(0.6

8)

3.46

(0.4

3)

3.41

(0.6

4)

4.28

(0.5

0)

3.51

(0.4

3)

3.88

(0.9

1)

2.56

(1.2

4)

3.48

(0.5

6)

3.42

(0.8

5)

4.20

(0.4

7)

3.70

(0.7

9)

3.34

(1.0

7)

3.07

(1.2

8)

3.42

(0.3

8)

3.15

(1.0

1)

4.30

(0.4

6)

3.93

(0.6

1)

4.01

(0.5

1)

2.13

(0.6

0)

3.55

(0.4

3)

3.88

(0.7

0)

4.20

(0.6

1)

3.35

(0.3

5)

3.88

(0.5

7)

2.46

(0.5

8)

3.52

(0.5

0)

3.65

(0.5

6)

3.88

(0.6

4)

3.71

(0.7

2)

3.95

(0.5

5)

2.45

(0.4

2)

3.54

(0.4

6)

3.73

(0.4

8)

Chapter I II III IV V VI VII VIII

59

Talented youth female field hockey players We found a significant main effect on performance level. Elite players performed better than sub-elite players on physiological, technical, tactical and psychological characteristics. Elite players performed more runs on the interval shuttle run test [F (1,30) = 12.538, p < 0.01]. They were also faster in the peak shuttle dribble [F (1,30) = 3.146, p < 0.05], repeated shuttle dribble [F (1,30) = 4.536, p < 0.05] and slalom dribble [F (1,30) = 4.064, p < 0.05], with higher scores on general tactics [F (1,30) = 8.133, p < 0.01] and tactics for possession of the ball [F (1,30) = 4.719, p < 0.05]. Finally, elite players were more motivated [F (1,30) = 6.840, p < 0.01] and had more confidence [F (1,30) = 4.509, p < 0.05] than sub-elite players.

Concerning the development of the performance characteristics in two years, a significant main effect on measurement was found on anthropometric, physiological, technical and tactical characteristics. From measurements 1 through 3, players were taller [F (1,29) = 13.481, p < 0.01) and heavier [F (1,29) = 7.864, p < 0.01]. They improved on repeated shuttle sprint performance [F (1,29) = 4.248, p < 0.05] and interval endurance capacity [F (1,29) = 6.546, p < 0.01], becoming faster in the peak shuttle dribble [F (1,29) = 5.626, p < 0.01] and attaining higher scores on general tactics [F (1,29) = 7.941, p < 0.01].

We found an interaction effect between performance level and measurement on interval endurance capacity [F (1,29) = 2.600, p < 0.05]. In contrast to sub-elite players, elite players showed an increase in the number of runs on the interval shuttle run test, especially from the second to the third measurement (Figure 4.1A). We also found an interaction effect on slalom dribble performance [F (1,29) = 3.178, p < 0.05]. Elite players improved more than sub-elite players (Figure 4.1D). Finally, we found an interaction effect for confidence [F (1,29) = 3.065, p < 0.05]. Scores for confidence on the second measurement were lower than on the first measurement for both elite and sub-elite players. In contrast to the sub-elite players, whose scores remained relatively stable from t2 to t3, the confidence scores of the elite players on t3 were back to the level of measurement 1. No other interaction effects were found, indicating that the development from t1 through t3 in test scores is similar for elite and sub-elite players.

60

t1 t2 t340

50

60

70

80

90

100

110N

umbe

r of r

uns

Interval endurance capacity

t1 t2 t39

9.5

10

10.5

11

Tim

e (s

)

Peak shuttle dribble performance

t1 t2 t328

29

30

31

32

33

34

Tim

e (s

)

Measurement

Dribble performance in a repeated shuttle run

t1 t2 t317

18

19

20

Tim

e (s

)

Measurement

Performance in a slalom dribble

Female eliteFemale sub-eliteMale eliteMale sub-elite

A B

C D

Figure 4.1. Performances of the talented youth field hockey players on the three measurements.

Talented youth male field hockey players As with female players, we found a significant main effect on performance level in male players. Elite players were faster than sub-elite players in the peak shuttle dribble [F (1,29) = 2.914, p < 0.05], repeated shuttle dribble [F (1,29) = 2.988, p < 0.05] and slalom dribble [F (1,29) = 8.306, p < 0.01] (Figures 4.1B, 4.1C, 4.1D). Compared to sub-elite players, elite players scored better on general tactics [F (1,29) = 38.883, p < 0.01], tactics for possession of the ball [F

(1,29) = 23.640, p < 0.01] and tactics for non-possession of the ball [F (1,29) = 25.888, p < 0.01]. Sub-elite players scored better than elite players on concentration [F (1,28) = 6.264, p < 0.01].

A significant main effect on measurement has been found for anthropometric, physiological, technical and psychological characteristics. From measurements 1 through 3, players were taller [F (1,28) = 5.802, p < 0.01], heavier [F (1,28) = 3.752, p < 0.05] and had less body fat [F (1,28) = 3.400, p < 0.05]. They improved on peak shuttle sprint performance [F (1,28) = 3.623, p < 0.05], repeated shuttle sprint performance [F (1,28) = 9.693, p < 0.01], slalom sprint [F (1,28) = 2.875, p < 0.05], interval endurance capacity [F (1,28) = 5.915, p < 0.01] and slalom dribble [F (1,28) = 2.635, p < 0.05]. Scores on anxiety control decreased from measurements 1 through 3 [F (1,27) = 3.678, p < 0.05].

Chapter I II III IV V VI VII VIII

61

We found an interaction effect between performance level and measurement for interval endurance capacity [F (1,28) = 3.699, p < 0.05]. Both elite and sub-elite players improved with about 12 runs from t1 to t2. From t2 to t3, sub-elite players did not improve in contrast to elite players, who ran on average over 20 more runs (Figure 4.1A). We also found an interaction effect for anxiety control [F (1,27) = 6.647, p < 0.01]. From t1 through t3, elite players scored lower in contrast to sub-elite players, whose scores remained relatively stable.

4.4 Discussion

This study deals with talented youth field hockey players in the Netherlands. At the end of the 2000-2001 competitive season we measured 126 players in the 12-16 age-bracket, all part of a talent development program of a field hockey club of national prestige. Most of today’s top performers played in a youth selection team when they were younger. We divided the players into elite and sub-elite youth categories on the basis of membership in an extra selection team of the Dutch Field Hockey Association (KNHB) in the 2000-2001, 2001-2002 and/or 2002-2003 seasons. During this study, there was only one player that shifted from sub-elite to elite. Therefore, it appears that to develop a successful field hockey career, a youth player has to be part of an extra selection team of the KNHB around the age of 14 already. This, however, is no guarantee for success, since there are more shifts from elite to sub-elite (n = 5) and from sub-elite to club level (n = 30), making it clear that in a period of two years more than 25% of the players could not meet the expectations. It is obviously very hard to predict who will ultimately reach elite status in adulthood, especially in a team sport. Unlike individual sports, in which there is a unidimensional performance criterion like time, distance or height, a performance in team sports depends on the combination of numerous mini-performances of the player and his team-mates (Régnier et al., 1993). Due to a lack of objective performance measurements, players therefore have to convince the scout, trainer or coach of their talent.

Over the years, many researchers have attempted to get a grip on the rather vague concept of talent in studies that concentrated on music (Sloboda et al., 1994a,b; Krampe and Ericsson, 1996; Howe et al., 1998) and sports (Starkes and Deakin, 1984; Starkes, 1987; Helsen and Pauwels, 1993; Starkes et al., 1994; Helsen and Starkes, 1999). However, the suggestion that talent provides a basis for predicting excellence is not supported by the current evidence (Helsen et al., 2000). To justify early identification and selection of talented young athletes, it seems crucial to gain more insight into the characteristics of ‘tomorrow’s stars’. One way to do so is by adopting a multidisciplinary, longitudinal approach in which talented youth players are followed over time until some of them reach elite status in adulthood.

62

In our study, 65 players who have been considered as talented for at least three consecutive years were followed and measured on three occasions. Both female and male elite youth players scored better than the sub-elite players on technical and tactical variables. In the female group, elite players also scored better on interval endurance capacity, motivation and confidence than sub-elite players, whereas the sub-elite male players had higher scores on concentration than the elite male players. Hence, the results clearly show that relevant variables to distinguish between elite and sub-elite players do not stem from a single domain of performance characteristics. This is in line with a study of Nieuwenhuis et al. (2002) in which successful and less successful female field hockey players were compared. They concluded that the successful 14-15 year-old player passes the ball more accurately over a distance, is faster in covering a short distance, has a broader humerus and femur, and experiences the competitive situation more positively.

Our results show that maintaining speed while dribbling a hockey ball is important at the elite level. These findings are consistent with other studies. Reilly and Bretherton (1986) reported better dribbling control in elite versus county field hockey players, and Keogh et al. (2003) mentioned better scores for regional representative players in contrast to club-level players while dribbling a hockey ball through an agility course. Top-level coaches also confirmed the importance of technique in field hockey (Van Rossum and Gagné, 1994).

It follows that tactical skills – performing the right action at the right moment – seems crucial for a successful career in field hockey. Our results are in line with other studies showing that skilled players outscore less skilled ones on tactical elements (Williams et al., 1993; Williams and Davids, 1995; Enns and Richards, 1997). However, we are aware that the tactical skill variables in our study do not specify exactly the underlying processes that enable players to perform the right action at the right moment. In the absence of a practical, reliable and valid instrument to measure tactical skills, we used the opinion of the trainer to gain insight into each player’s general tactics, tactics for possession of the ball and tactics for non-possession of the ball. Because we were unable to measure tactical skills directly with the player, results may have been confounded. Trainers work with the players throughout the season during training and match play, and know which players belong to an extra selection team of the KNHB. They might therefore be inclined to grade those players higher than the sub-elite players. However, the trainers in this study were highly qualified and considered as experts in the field, and their opinion was considered as valuable.

Except for the interval endurance capacity in female players, we found no differences in anthropometric or physiological characteristics between both performance groups. In contrast, according to Keogh et al. (2003), measures of body fat percentage and short-duration sprinting speed are useful for distinguishing between field hockey players of different ability.

Chapter I II III IV V VI VII VIII

63

However, in their study they compared regional representative players with local club-level players, and not players who were playing at the national level as was the case in our study. Evidently, differences between players at the elite level are less related to physical and physiological characteristics (Elferink-Gemser et al., 2004). Also Franks et al. (1999) could not discriminate either between young soccer players at the highest level on the basis of their physical and physiological profiles.

In the female group, elite youth players scored higher on motivation and confidence than the sub-elite players; we did not find such results in the male group, where on average all players had high scores on motivation and confidence. It seems that in talent identification and development, psychological characteristics are more important for female than for male youth players. When comparing the scores on confidence of the male and female youth players, it appears that the elite female players had scores similar to those of the male players, whereas the sub-elite female players had lower scores. Other studies show that male players have on average higher confidence scores than females (Cox and Liu, 1993; Sewell and Edmondson, 1996), but it seems that this gender difference on confidence cannot be applied at the elite level.

We found interaction effects showing a different development over time for both performance groups. Compared to sub-elite youth players, elite youth players improved more on interval endurance capacity and slalom dribble performance. The improvement of the interval endurance capacity is consistent with the TOYA study of Baxter-Jones et al. (1993; 1995). They studied the development of aerobic power in young soccer, swimming, gymnastics and tennis athletes in the 8-16 age-bracket. Results showed that VO2max increased significantly with pubertal status in males. In our study, male youth players increased their number of runs on the Interval Shuttle Run Test from the first through the third measurement. Although we did not take any maturity measures, we do not expect significant differences between elite and sub-elite players concerning maturation based on a similar development of their length, body mass and body fat percentage. Nevertheless, one cannot rule out that the most mature children were performing best at this age. Baxter-Jones et al. did not show a significant increase in VO2max in the latter stages of puberty in females. In our study, female sub-elite players increased their score on the ISRT (4 runs) only slightly, whereas the female elite players were able to increase their ISRT score with 20 runs on average from the first through the third measurement.

To sum up, elite youth players seem to either score better than sub-elite youth players on performance characteristics by the age of 14 and subsequently remain better in the following two years, or they have similar scores to the sub-elite youth players on the first measurement but develop these characteristics better in the next two years (ages 14-16). Since at the first

64

measurement both elite and sub-elite youth players were active in field hockey for an average of over 6 years, these findings are not likely to be caused by a difference in active field hockey experience. However, elite players did seem to train more frequently than sub-elite players at the age of 14, even at 16. This concerns both specific field hockey training and general training. These findings are in line with the study of Ericsson et al. (1993), who proposed a model of expertise based on deliberate practice. They argued that practice is the only determinant of expertise. However, another explanation could be that the elite youth players have inherited a more favourable genetic profile for success in field hockey. According to Howe (1998), a talent originates in genetically-transmitted structures and hence is at least partly innate. Probably both nature and nurture are essential. In a study of Starkes et al. (1996), coaches of elite figure skaters acknowledged the role played by talent, but stated that even the most talented athletes must practice hard to succeed. Until the middle of the 20th century, it was possible to become an international athlete without belonging to a nation’s group of most-talented individuals (Bouchard et al., 1997). Today, the level of competition has increased to the point that only those athletes who combine their talent with intensive training are potentially able to reach elite status.

In conclusion, in field hockey future elite players seem to excel in tactical skills by the age of 14 already. They also stand out in specific technical skills and develop these together with their interval endurance capacity favourably in the subsequent two years, and show high levels of motivation and confidence. To verify these conclusions, it will be interesting to follow these players until they reach elite status in adulthood.

Chapter I II III IV V VI VII VIII

65

References

Alabin, V., Nischt, G., and Jefimov, W. (1980). Talent selection. Modern athlete and coach. 36-37. Atkinson, G. and Nevill, A.M. (2001). Selected issues in the design and analysis of sport performance

research. Journal of Sport Sciences, 19, 811-827. Aziz, A.R., Chia, M., and Teh, K.C. (2000). The relationship between maximal oxygen uptake and

repeated sprint performance indices in field hockey and soccer players. The Journal of Sports Medicine and Physical Fitness, 40, 195-200.

Bakker, F. (1995). Ontwikkeling van tests op het terrein van de sportpsychologie in Nederland en Vlaanderen. [Development of tests for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 6, 3-15.

Baxter-Jones, A., Goldstein, H., and Helms, P. (1993). The development of aerobic power in young athletes. Journal of Applied Physiology, 75, 1160-1167.

Baxter-Jones, A.D.G., Helms, P., Muffulli, N., Baines-Preece, J., and Preece, M. (1995). Growth and development of male gymnasts, swimmers, soccer and tennis players: a longitudinal study. Annals of Human Biology, 22, 381-395.

Bouchard, C., Malina, R.M., and Pérusse, L. (1997). Genetics of fitness and physical performance. Champaign, IL: Human Kinetics.

Boyle, P.M., Mahoney, C.A., and Wallace, W.F. (1994). The competitive demands of elite male field hockey players. Journal of Sports Medicine and Physical Fitness, 34, 235-241.

Burwitz, L., Moore, P.M., and Wilkinson, D.M. (1994). Future directions for performance-related sports science research: an interdisciplinary approach. Journal of Sports Sciences, 12, 93-109.

Companjen, T. and Bakker, F. (2003). Overzicht van sportpsychologische vragenlijsten in Nederland en Vlaanderen. [Outline of questionnaires for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 14, 34-41.

Cox, R.H. and Liu, Z. (1993). Psychological skills: a cross-cultural investigation. International Journal of Sport Psychology, 24, 326-340.

Deshaies, P., Pargman, D., and Thiffault, C. (1979). A psychobiological profile of individual performance in junior hockey players. In Psychology of Motor Behaviour and Sport – 1978 (edited by G.C. Roberts and K.M. Newell), pp. 36-50. Champaign, IL: Human Kinetics.

Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. (2001). Ontwikkeling van de ‘Tactiek in Sport’ vragenlijst. [Development of the ‘Tactics in Sports’ questionnaire]. Internal publication, Center for Human Movement Sciences, University of Groningen, The Netherlands.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. (2004). Relation between multidimensional performance characteristics and level of performance in talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

Enns, J. and Richards, J. (1997). Visual attentional orienting in developing hockey players. Journal of Experimental Child Psychology, 64, 255-275.

Ericsson, K.A. (1996). The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games. Mahwah, NJ: Lawrence Erlbaum Associates.

Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406.

Franks, A., Williams, A.M., Reilly, T., and Nevill, A. (1999). Talent identification in elite youth soccer players: Physical and physiological characteristics. Journal of Sports Sciences, 17, 812.

Ghosh, A.K., Goswami, A., Maxumdar, P., and Mathur, D.N. (1991). Heart rate & blood lactate response in field hockey players. The Indian Journal of Medical Research, 94, 351-356.

Hahn, A. (1990). Identification and selection of talent in Australian rowing. Exel, 6, 5-11. Hargraeves, A. (1984). Fitness profile of the British Olympic men’s hockey team, 1978/80. The

Hockey Circle, 68. Helsen, W. F. and Pauwels, J. (1993). The relationship between expertise and visual information

processing in sport. In Cognitive Issues in Motor Expertise (edited by J.L. Starkes and F. Allard), pp. 109-134. Amsterdam: Elsevier.

66

Helsen, W.F. and Starkes, J.L. (1999). A multidimensional approach to skilled perception and performance in sport. Applied Cognitive Psychology, 13, 1-27.

Helsen, W.F., Hodges, N.J., Van Winckel, J., and Starkes, J.L. (2000). The roles of talent, physical precocity and practice in the development of soccer expertise. Journal of Sports Sciences, 18, 727-736.

Howe, M.J.A., Davidson, J.W., and Sloboda, J.A. (1998). Innate talents: Reality or myth. Behavioral and Brain Sciences, 21, 399-442.

Keogh, J.W.L., Weber, C.L., and Dalton, C.T. (2003). Evaluation of anthropometric, physiological, and skill-related tests for talent identification in female field hockey. Canadian Journal of Applied Physiology, 28, 397-409.

Krampe, R. and Ericsson, A. (1996). Maintaining excellence: Deliberate practice and elite performance in young and older pianists. Journal of Experimental Psychology: General, 125, 331-359.

Lemmink, K.A.P.M. and Visscher, C. (2003). The relationship between the Interval Shuttle Run Test and maximal oxygen uptake in soccer players. Journal of Human Movement Studies, 45, 219-232.

Lemmink, K.A.P.M., Elferink-Gemser, M.T., and Visscher, C. (2004a). Evaluation of the reliability of two field hockey-specific sprint and dribble tests in young field hockey players. British Journal of Sports Medicine, 38, 138-142.

Lemmink, K.A.P.M., Verheijen, R., and Visscher, C. (2004b). The discriminative power of the Interval Shuttle Run Test and the Maximal Multistage Shuttle Run Test for playing level of soccer. Journal of Sports Medicine and Physical Fitness, 44, 233-239.

Lemmink, K.A.P.M., Visscher, C., Lambert, M.I., and Lamberts, R. (2004c). The Interval Shuttle Run Test for intermittent sport players: evaluation of reliability. Journal of Strength and Conditioning Research, 18, 821-827.

Lemmink, K.A.P.M., Dolleman, G., Verheijen, R., and Visscher, C. (2000). Interval Sprint Test en Interval Shuttle Run Test – betrouwbaarheid en discriminerend vermogen van twee nieuwe voetbaltests. [Interval Sprint Test and Interval Shuttle Run Test – reliability and discriminative power of two new tests for soccer players]. Geneeskunde en Sport, 33, 39-48.

Lothian, F. and Farrally, M. (1994). A time-motion analysis of women’s hockey. Journal of Human Movement Studies, 26, 255-265.

Mahoney, M.J., Gabriel, T.J., and Perkins, T.S. (1987). Psychological skills and exceptional performance. The Sport Psychologist, 1, 181-199.

Nieuwenhuis, C.F., Spamer, E.J., and Van Rossum, J.H.A. (2002). Prediction function for identifying talent in 14- to 15-year-old female field hockey players. High Ability Studies, 13, 21-33.

Nunez, C., Callagher, D., Visser, M., Pi-Sunyer, F.X., Wang, Z., and Heymsfield, S.B. (1997). Bioimpedance analysis: evaluation of leg-to-leg system based on pressure contact footpad electrodes. Medicine and Science in Sports and Exercise, 29, 524-531.

Patel, D.R., Stier, B., and Luckstead, E.F. (2002). Major international sport profiles. Pediatric Clinics of North America, 49, 769-792.

Pienaar, A.E., Spamer, M.J., and Steyn, H.S. (1998). Identifying and developing rugby talent among 10-year-old boys: A practical model. Journal of Sports Sciences, 16, 691-699.

Régnier, G., Salmela, J.H., and Russell, S.J. (1993). Talent detection and development in sport. In A Handbook of Research on Sports Psychology (edited by R. Singer, M. Murphey and L.K. Tennant), pp. 290-313. New York: Macmillan.

Reilly, T. and Bretherton, S. (1986). Multivariate analysis of fitness of female field hockey players. In Perspectives in Kinanthropometry (edited by J.A.P. Day), pp. 135-142. Champaign, IL: Human Kinetics.

Reilly, T. and Seaton, A. (1990). Physiological strain unique to field hockey. Journal of Sports Medicine and Physical Fitness, 30, 142-146.

Reilly, T. and Borrie, A. (1992). Physiology applied to field hockey. Sports Medicine, 14, 10-26. Reilly, T., Williams, A.M., Nevill, A., and Franks, A. (2000). A multidisciplinary approach to talent

identification in soccer. Journal of Sports Sciences, 18, 695-702.

Chapter I II III IV V VI VII VIII

67

Sewel, D.F. and Edmondson, A.M. (1996). Relationships between field position and pre match competitive state anxiety in soccer and field hockey. International Journal of Sport Psychology, 27, 159-172.

Sloboda, J.A., Davidson, J.W., and Howe, M.J.A. (1994a). Is everyone musical? The Psychologist, 7, 349-354.

Sloboda, J.A., Davidson, J.W., and Howe, M.J.A. (1994b). Musicians: Experts not geniuses. The Psychologist, 7, 363-364.

Starkes, J.L. (1987). Skill in field hockey: The nature of the cognitive advantage. International Journal of Sport Psychology, 2, 146-160.

Starkes, J.L. and Deakin, J. (1984). Perception in sport: A cognitive approach to skilled performance. In Cognitive Sport Psychology (edited by W.F. Straub and J.M. Williams), pp. 115-128. Lansing, NY: Sport Science Associates.

Starkes, J.L., Allard, F., Lindley, S., and O’Reilly, P. (1994). Abilities and skill in basketball. International Journal of Sport Psychology, 25, 249-265.

Starkes, J.L., Deakin, J.M., Allard, F., Hodges, N.J. and Hayes, A. (1996). Deliberate practice in sports: What is it anyway? In The road to excellence: The acquisition of expert performance in the arts, sciences, sports and games (edited by K.A. Ericsson), pp. 81-106. Hillsdale, NJ: Lawrence Erlbaum Associates.

Van Rossum, J.H.A. and Gagné, F. (1994). Rankings of predictors of athletic performance by top level coaches. European Journal for High Ability, 5, 68-78.

Withers, R.T., Roberts, G.D., and Davies, G.J. (1977). The maximum aerobic power, anaerobic power and body composition of South Australian male representatives in athletics, basketball, field hockey and soccer. The Journal of Sports Medicine and Physical Fitness, 17, 391-400.

Williams, M. and Davids, K. (1995). Declarative knowledge in sport: a by-product of experience or a characteristic of expertise. Journal of Sport and Exercise Psychology, 17, 259-275.

Williams, A.M. and Reilly, T. (2000). Talent identification and development in soccer. Journal of Sports Sciences, 18, 657-667.

Williams, A.M., Davids, K., Burwitz, L., and Williams, J.G. (1993). Cognitive knowledge and soccer performance. Perceptual and Motor Skills, 76, 579-593.

Chapter V

Development of the interval endurance capacity in elite and sub- elite youth field hockey players Elferink-Gemser, M.T., Visscher, C., Van Duijn, M.A.J.1, and Lemmink, K.A.P.M.

1Interuniversity Center for Social Science Theory and Methodology, University of Groningen, the Netherlands

70

Abstract

To gain more insight into the mechanisms that underlie the development of the interval endurance capacity in talented youth field hockey players in the 12-19 age band, 377 complete measurements were taken in a period of three years. A longitudinal model for interval endurance capacity was developed using the multilevel modelling program MlwiN. With the model, scores on the Interval Shuttle Run Test can be predicted for elite and sub-elite boys and girls in field hockey in the age-band of 12-19 years. During adolescence both male and female elite youth players have a more promising development pattern of their interval endurance capacity than sub-elite youth players. Besides age, gender, and performance level, the effect of percentage body fat, additional training, and motivation was investigated.

Chapter I II III IV V VI VII VIII

71

5.1 Introduction

Match analyses make clear that field hockey is a high intensity non-continuous game in which the physiological demands are considerable, placing it in the category of ‘heavy exercise’ (Ghosh et al., 1991; Reilly and Borrie, 1992; Aziz et al., 2000). In terms of energy requirements, the aerobic capacity is most important during matches at the elite level. Although great anaerobic capacity is needed during the many brief bursts of high-energy release, it is the aerobic capacity that is needed for efficient recovery during the short rest periods (Bhanot and Sidhu, 1983; Boyle et al., 1994; Lothian and Farrally, 1994). Field hockey players need a well developed interval endurance capacity to carry out all sorts of explosive actions such as intermittent running, accelerating, decelerating, cruising, and dribbling. While performing these actions they repeatedly change their direction to, for example, overtake an opponent, thereby increasing the overall effort needed in field hockey (Patel et al., 2002). The interval endurance capacity is the ability to perform high-intensity activities like running and sprinting, as well as the ability to recover well during low-intensity activities such as walking and jogging (Lemmink and Visscher, 2003).

In our longitudinal study on performance characteristics of talented youth field hockey players in the age-band of 12-18 years, a strong improvement in interval endurance capacity is apparent. As well in boys as in girls, elite youth players improved themselves more than sub-elite youth players on their interval endurance capacity across a period of two years (Elferink-Gemser et al., in revision 2005). When the youth players were on average fourteen years old, however, differences in interval endurance capacity scores were not significant yet between elite and sub-elite players (Elferink-Gemser et al., 2004). One year later, there was a trend that elite players outscored the sub-elite players on their interval endurance capacity, but again differences were not significant. Two years later, at an average age of sixteen years, differences were significant, favoring the elite players. Therefore, to unravel the relationship between the interval endurance capacity and the level of performance in talented field hockey players, it is essential to gain a deeper insight into the development of this performance characteristic.

In ‘normal’ children, aerobic capacity, i.e., VO2max, increases proportionally to body size and mass in both sexes. Most studies show that, when ‘normalizing’ for body size and mass, VO2max remains stable in males throughout childhood and adolescence while it decreases in females (Krahenbuhl et al., 1985). The Amsterdam Growth Study, a 23-year follow-up from teenager to adult about lifestyle and health (Kemper, 2004), shows that in the adolescent period VO2max increases in males whereas it decreases gradually in females (Kemper and Koppes, 2004). Generally, anaerobic performance also increases with age. Girls improve from late childhood to 14-15 years whereas in boys the increase sustains to 19 years. In late

72

childhood and early adolescence gender differences are evident and they are magnified later in adolescence (Martin and Malina, 1998).

The adolescent period is characterized by an acceleration of somatic growth and rapid changes in body composition and hormonal status including growth spurt and increase in fat-free mass (Bitar et al., 2000). Anthropometrics such as body height, lean body mass and percentage body fat influence the physiological aspects of a sports performance, i.e., the interval endurance capacity. Increase in body height is related to an increase in lung volume and therefore with an increase in metabolism and endurance. Gain in lean body mass is related to muscle mass increase and therefore positively influences endurance in contrast to gain in body fat, which negatively influences endurance (e.g., Astrand et al., 2003).

Most world-class field hockey players are in their twenties and as a consequence athletes who want to make it to the top have to start training already at a relatively early age, thereby developing their interval endurance capacity. It is generally known that with training players can improve their performance by increasing the aerobic and anaerobic energy output during a particular movement. This is also the case in youth players (e.g. Powers and Howley, 2001). However, it is not self-evident that all players make use of their full interval endurance capacity during training or competition. A player has to be motivated to do so since intense activity can cause uncomfortable side effects such as fatigue and muscle soreness. Motivation affects the intensity and persistence of a player’s behavior, which in sports can obviously have a strong impact on his or her performance (e.g., Silva and Weinberg, 1984).

Is it possible to adequately model the development of the interval endurance capacity of 12-19 year-old talented youth field hockey players? Can the development of the interval endurance capacity of talented field hockey players be explained by age, performance level, gender, anthropometrics, training, and motivation? These questions led to the goal of this present study which aim it is to gain more insight into the mechanisms that underlie the development of a performance characteristic that is important for a successful career in field hockey: the interval endurance capacity.

5.2 Methods

Participants In the period 2000-2003, 217 talented field hockey players in the 12-19 years age-bracket participated in a semi-longitudinal study on the relation between multidimensional performance characteristics and performance level. This group consisted of 110 male and 107 female players. All participants were part of a talent development program of a field hockey club of national prestige, and were playing at the highest level for their age category. For three

Chapter I II III IV V VI VII VIII

73

consecutive years, measurements were taken at the end of the competitive field hockey season of 2000-2001 (t1), 2001-2002 (t2), and 2002-2003 (t3). In total 404 measurements were taken since 77 players were tested on all three occasions (231 measurements), 33 players were tested on two occasions (66 measurements) and 107 players were tested on one occasion only (107 measurements). Of these measurements, 392 contained scores of the interval endurance capacity and 377 measurements were complete in that there were scores on all variables.

Next to being part of a talent development program, talented Dutch field hockey players who are considered to be current elite youth players are invited to train and play in a youth selection team of the Dutch Field Hockey Association (KNHB). Talented players who are considered to be current sub-elite youth players are part of the talent development program of their field hockey club only. This distinction, based on the performance level of the players was also followed in this study.

Procedure All players were informed about the procedure of the study before giving their verbal consent to participate. The field hockey clubs and trainers gave permission for this study and procedures were in accordance with the standards of the local medical ethics committee of the University of Groningen. The players completed the Interval Shuttle Run Test for interval endurance capacity on a synthetic field hockey playing surface (water-based pitch). Ambient temperature, humidity and wind conditions were documented. In addition, anthropometric measurements were taken and the players filled in questionnaires for training habits and motivation. Anthropometric characteristics Anthropometric measurements were height (m), lean body mass (kg) and percentage of body fat. The latter was estimated by means of leg-to-leg bioelectrical impedance (BIA) analysis (Valhalla BIA, Valhalla, Inc., San Diego, CA). This method proved to be reliable for measuring body fat percentage, and results correlated highly with body fat percentage as measured with underwater weighing and dual energy X-ray absorptiometry (Nunez et al., 1997). Interval Shuttle Run Test Interval endurance capacity was measured with the Interval Shuttle Run Test (ISRT) (Lemmink et al., 2000; Lemmink and Visscher, 2003). The ISRT is a field test that contains intervals at a work-rest ratio of 2:1 and turning points at 20 m. Players alternately run for 30 seconds and walk for 15 seconds. The running speed increases from 10 km/hour every 90

74

seconds until exhaustion. The number of fully completed 20-m runs is recorded as the test score. During the ISRT players were carrying their hockey stick. Although, as a result of the interval character of the test, anaerobic energy production is important, aerobic energy production as indicated by VO2max contributes mainly to the total energy requirement during the test (Lemmink and Visscher, 2003). In previous research, the reliability of the ISRT as a maximal field test for intermittent sport players has been supported (Lemmink et al., 2000; 2004a). The ISRT also shows discriminative power for players at different levels of competition supporting the validity of the ISRT for measuring endurance in a more specific way (Lemmink et al., 2004b). Training In the questionnaire players were asked how many field hockey training sessions they attend per week and what the duration of these training sessions is. Time spend in matches is not included in the field hockey training since all players spend equal time in field hockey matches, on average 1 hour per week. Players also filled in how many times per week they train in other sports or by themselves and what the duration of these training sessions is. Time spend on physical education at school, which is on average 2.5 hours per week, is excluded. Outcome variables are field hockey training (hours/week) and additional training (hours/week). Motivation Motivation was measured using the Dutch youth version of the Psychological Skills Inventory for Sports. The Psychological Skills Inventory for Sports (PSIS-R-5) consists of 5-point Likert type items that are distributed over 6 scales (Mahoney et al., 1987). The PSIS-R-5 has been translated into Dutch and subjected to psychometric testing (Bakker, 1995; Companjen and Bakker, 2003). The Dutch Youth Version of the Psychological Skills Inventory for Sports (PSIS–Youth) is based upon the Dutch version of the PSIS-R-5, but the formulation of questions is simpler. The Motivation scale contains eight 5-point Likert type items. The answer almost never equates to 1, and almost always to 5. Items worded negatively (indicating a problem or concern) are transformed by reversing the aforementioned 1-5 format. In this way, a high score on each scale corresponds to the psychological skill being present to a large extent. The mean scale score was calculated which ranged from (1) low to (5) high. An example of an item is: ‘In my sport, I want to bring out the best in myself’.

Chapter I II III IV V VI VII VIII

75

Study design Measurements were taken annually for three consecutive years from 2001 to 2003. As there were overlaps in ages between the clusters it was possible to estimate a consecutive 7-year development pattern. Although subjects returned each year, they were not measured exactly the same time each year. However, intervals between measurements were never less than 12 or longer than 13 months. The age of the subjects was recorded in months. To create standardized age groups, the players were classified into age groups at the time of measurement. A 14-year old player was defined as a player tested within the age range 13.50-14.49 years. Data analysis Longitudinal changes in interval endurance capacity were investigated using the multilevel modelling program MlwiN (Goldstein et al., 1998). Multilevel modelling is an extension of multiple regression, which is appropriate for analysing hierarchically structured data. In the present longitudinal data set, a simple two-level hierarchy was defined with the repeated measurements (defined as level 1 units) grouped within the individual players who form the level 2 units. An advantage of using a multilevel regression modelling approach is that both the number of measurements and the temporal spacing of the measurements may vary between players (Maas and Snijders, 2003). A multilevel model describes not only underlying population trends in a response (the fixed part of the model), but also models the variation around this mean response due to the time of measurement and due to individual differences (the random part) (Snijders and Bosker, 2000).

Following Snijders and Bosker (2000, chapter 12) the first step in the multilevel modelling of the interval endurance capacity data was to establish a satisfactory variance structure for these longitudinal data, using age (measured as months/12 - 15 years). Then, the difference between elite and sub-elite-groups was modelled, taking into account interactions with age and gender. In a next step, the effects of the anthropometric variables, height (m), lean body mass (kg), and percentage body fat, were investigated. Subsequently, the effect of the total number of training hours per week, as well as the effect of different types of training (distinguishing field hockey training and additional training) were investigated. Finally, the effect of motivation was tested.

5.3 Results

In Table 5.1, the players’ anthropometrics, training, motivation, and interval endurance capacity scores are presented by gender, performance level, and age.

Tab

le 5

.1.

Mea

n sc

ores

(sd)

of t

alen

ted

yout

h fie

ld h

ocke

y pl

ayer

s pre

sent

ed b

y ge

nder

, per

form

ance

leve

l, an

d ag

e.

Not

e: F

ield

hoc

key

train

ing

(hou

rs p

er w

eek)

is e

xclu

sive

of f

ield

hoc

key

mat

ches

. Add

ition

al tr

aini

ng (h

ours

per

wee

k) is

exc

lusi

ve o

f phy

sica

l edu

catio

n at

scho

ol.

a O

ne m

issi

ng v

alue

. b T

wo

mis

sing

val

ues.

Coh

ort

n A

ge (y

ear)

H

eigh

t (m

) L

ean

body

m

ass (

kg)

% B

ody

fat

Fiel

d ho

ckey

tr

aini

ng

Add

ition

al

trai

ning

M

otiv

atio

n

(1-5

) IS

RT

(run

s of 2

0m)

Mal

e el

ite y

outh

pla

yers

12-1

3 ye

ars

11

12.8

8 (0

.55)

1.

62 (0

.06)

43

.3 (5

.1)

10.7

5 (2

.82)

3.

4 (0

.7)

6.2

(2.7

) 4.

51 (0

.35)

58

.82

(18.

64)

14 y

ears

20

13

.95

(0.2

9)

1.67

(0.0

7)

49.6

(7.0

) 9.

23 (2

.48)

3.

9 (1

.2)

5.0

(4.5

) 4.

56 (0

.30)

77

.10

(23.

03)

15 y

ears

21

15

.04

(0.3

0)

1.73

(0.0

7)

52.8

(6.7

) 7.

94 (2

.48)

4.

8 (2

.2)

3.0

(2.4

) 4.

24 (0

.50)

88

.76

(22.

53)

16 y

ears

13

16

.06

(0.2

7)

1.78

(0.0

4)

59.7

(4.3

) 7.

97 (2

.70)

5.

0 (1

.7) a

2.

4 (2

.5) a

4.

27 (0

.51)

a

91.4

6 (2

1.55

) 17

yea

rs

7 16

.98

(0.3

1)

1.79

(0.0

6)

64.1

(2.9

) 7.

46 (3

.03)

5.

5 (1

.4)

2.0

(2.0

) 4.

18 (0

.58)

88

.86

(29.

13)

18-1

9 ye

ars

2 17

.94

(0.2

4)

1.81

(0.0

5)

70.0

(0.3

) 8.

90 (1

.12)

5.

3 (1

.1)

3.0

(2.8

) 3.

56 (0

.80)

10

0.50

(0.7

1)

Mal

e su

b-el

ite y

outh

pla

yers

12-1

3 ye

ars

9 13

.04

(0.4

0)

1.63

(0.0

9)

44.4

(7.3

) 9.

96 (4

.72)

3.

4 (0

.7)

2.8

(2.7

) 4.

22 (0

.44)

54

.67

(23.

77)

14 y

ears

26

13

.96

(0.3

2)

1.67

(0.0

8)

47.4

(6.8

) 10

.54

(5.9

0)

3.5

(0.6

) 3.

6 (4

.4)

4.27

(0.4

4)

56.5

8 (1

8.64

) 15

yea

rs

16

14.8

3 (0

.32)

1.

68 (0

.10)

48

.0 (8

.8)

8.47

(4.4

9)

3.5

(0.9

) a

4.7

(4.0

) a

4.35

(0.3

8) a

81.1

9 (2

6.16

) 16

yea

rs

29

15.9

7 (0

.31)

1.

77 (0

.06)

57

.0 (6

.5)

8.89

(3.9

3)

3.5

(0.7

) b

3.7

(3.5

) b

4.17

(0.5

0) b

84.5

9 (2

0.48

) 17

yea

rs

25

16.8

3 (0

.34)

1.

79 (0

.06)

60

.6 (6

.3)

9.16

(4.4

5)

4.1

(1.4

) a

3.4

(3.0

) a

4.17

(0.6

3) a

74.1

6 (2

7.16

) 18

-19

year

s 21

17

.91

(0.4

5)

1.81

(0.0

5)

67.7

(4.4

) 7.

31 (4

.33)

4.

3 (0

.9)

1.8

(2.5

) 4.

06 (0

.64)

74

.90

(21.

49)

Fem

ale

elite

you

th p

laye

rs

12

-13

year

s 12

12

.95

(0.3

2)

1.56

(0.0

8)

38.3

(5.1

) 15

.64

(5.1

7)

3.5

(1.1

) b

5.3

(1.9

) b

4.54

(0.2

9) b

48.5

0 (1

2.56

) 14

yea

rs

17

14.0

1 (0

.32)

1.

62 (0

.08)

41

.9 (4

.6)

19.0

5 (6

.58)

3.

7 (1

.3)

3.3

(2.5

) 4.

55 (0

.34)

a

58.3

5 (1

2.92

) 15

yea

rs

18

14.9

8 (0

.29)

1.

66 (0

.06)

43

.9 (3

.5)

21.5

7 (4

.76)

4.

4 (1

.1) a

2.

3 (2

.4) a

4.

57 (0

.47)

b

65.6

7 (1

9.42

) 16

yea

rs

15

16.0

7 (0

.30)

1.

67 (0

.06)

44

.5 (3

.3)

22.6

5 (5

.03)

5.

5 (1

.6)

1.5

(1.9

) 4.

36 (0

.52)

59

.80

(18.

16)

17 y

ears

9

17.0

0 (0

.31)

1.

69 (0

.07)

48

.5 (5

.9)

18.8

8 (5

.95)

6.

0 (2

.2)

2.5

(4.5

) 4.

68 (0

.20)

61

.00

(24.

12)

18-1

9 ye

ars

0

Fe

mal

e su

b-el

ite y

outh

pla

yers

12-1

3 ye

ars

14

12.8

9 (0

.45)

1.

61 (0

.05)

39

.8 (3

.5)

17.9

4 (7

.32)

3.

1 (0

.4) b

2.

1 (1

.8) b

3.

91 (0

.47)

b

47.7

9 (1

6.70

) 14

yea

rs

25

13.8

8 (0

.33)

1.

64 (0

.07)

42

.3 (4

.2)

18.2

0 (5

.30)

3.

1 (0

.4)

2.7

(2.2

) 4.

19 (0

.58)

57

.08

(16.

29)

15 y

ears

18

14

.92

(0.3

8)

1.66

(0.0

5)

41.8

(3.2

) 21

.18

(5.4

6)

3.3

(0.9

) 1.

5 (2

.5)

4.06

(0.6

6)

51.4

4 (1

0.00

) 16

yea

rs

27

15.8

8 (0

.27)

1.

69 (0

.04)

44

.5 (2

.8)

24.0

2 (4

.67)

3.

4 (0

.8)

1.4

(1.5

) 3.

93 (0

.57)

48

.88

(15.

92)

17 y

ears

21

16

.79

(0.2

5)

1.69

(0.0

4)

45.7

(3.8

) 25

.23

(4.3

8)

4.5

(1.5

) 2.

2 (2

.5)

4.23

(0.5

6)

51.2

9 (1

3.14

) 18

-19

year

s 16

17

.84

(0.3

8)

1.71

(0.0

5)a

51.0

6 (3

.85)

a

21.9

5 (3

.56)

a

4.7

(1.0

) a

1.8

(2.2

) a

4.31

(0.3

8) a

46.3

5 (1

5.03

)

Chapter I II III IV V VI VII VIII

77

As expected, in both male and female players height and lean body mass increase with age whereas percentage body fat tends to decrease in male and increase in female players. With age, players seem to increase their field hockey training and decrease their additional training. Motivation scores seem to remain relatively stable with age.

In Figure 5.1, predicted mean scores of the ISRT derived from the multilevel model are plotted against age for elite and sub-elite boys and elite and sub-elite girls. The general trend is that the interval endurance capacity increases with age in male youth players. However, elite youth players improve themselves more across time than sub-elite youth players. In females, the interval endurance capacity seems to increase with age in elite youth players only. Sub-elite youth players improve themselves until the age of about fifteen years and decrease their interval endurance capacity afterwards.

12 14 16 18

20

30

40

50

60

70

80

90

100

110

Predicted mean scores

ISR

T

Age (yrs)

Female eliteFemale sub-eliteMale eliteMale sub-elite

Figure 5.1. Predicted development of the interval endurance capacity of talented youth field hockey

players in the age-band of 12-18 years.

It was found that a polynomial model of order 2 adequately represents the variance structure of the data (deviance 3394.0, difference with a fully saturated model of 43.9 on 36 degrees of

78

freedom, p = 0.17). The fixed part of the model contains a different intercept and linear age term for boys and girls, and a common quadratic term; the random part of the model as a common level 2 (between-individual) variance and gender-specific level 1 (measurement) variances. The model was significantly improved by including differential effects of performance level for age and gender (deviance 3367.8, difference with previous model 26.2 on 3 degrees of freedom, p < 0.01). No effect was found for height and lean body mass, but a significant negative effect was found for percentage body fat (t = 4.423, p < 0.01). A positive significant effect was found for additional training (t = 3.374, p < 0.01), whereas no effect was found for field hockey training as such. Finally, a positive significant effect of motivation was found (t = 2.726, p = 0.003). The model parameters are given in Table 5.2. The coefficients of the variables percentage body fat, additional training hours, and motivation are unstandardized. Their effects, however, can be interpreted such that an additional training hour could compensate for 1.23 % body fat (1.093/0.889), or likewise, is equivalent to 0.225 points on the motivation scale (1.093/4.86). Table 5.2. Final multilevel model for interval endurance capacity data (377 measurements).

Fixed effects Coefficient S.E. p

Constant 52.6 9.10 < 0.001

Age (months/12 – 15 years) 6.21 1.20 < 0.001

Age2 -1.83 0.363 < 0.001

Boy 16.5 4.30 < 0.001

Sub-elite 0.786 2.90 0.393

Age x boy 5.27 1.33 < 0.001

Age x sub-elite -5.11 1.39 < 0.001

Boy x sub-elite -13.0 4.55 0.002

Percentage body fat -0.889 0.201 < 0.001

Additional training 1.092 0.324 < 0.001

Motivation 4.86 1.87 0.003

Random effects Variance S.E.

Between-individuals 136.0 25.43

Within-boy 292.8 39.20

Within-girl 105.9 16.31

Deviance 3205.6

Chapter I II III IV V VI VII VIII

79

With the multilevel model for interval endurance capacity, knowing the age of a player, his or her percentage body fat, additional training hours and motivation, scores on the Interval Shuttle Run Test for elite and sub-elite boys and girls can be predicted. Derived from the model in Table 5.2, the equations for the four subgroups are:

Elite boys: ISRT = (52.6) + (16.5) + (6.21 + 5.27) X age – (1.83 X age2) –

(0.889 X percentage body fat) + (1.092 X additional training hours) + (4.86 X motivation)

Sub-elite boys: ISRT = (52.6) + (16.5) + (0.786) – (13.0) + (6.21 + 5.27 – 5.11) X age –

(1.83 X age2) – (0.889 X percentage body fat) + (1.092 X additional training hours) + (4.86 X motivation)

Elite girls: ISRT = (52.6) + (6.21 X age) – (1.83 X age2) – (0.889 X percentage body fat) +

(1.092 X additional training hours) + (4.86 X motivation)

Sub-elite girls: ISRT = (52.6) + (0.786) + (6.21- 5.11) X age – (1.83 X age2) –

(0.889 X percentage body fat) + (1.092 X additional training hours) + (4.86 X motivation)

Thus, the development of the interval endurance capacity in the age-band from 12-19 years can be predicted with the multilevel model. For instance, it is predicted that an elite male player of fifteen years old will increase his performance on the Interval Shuttle Run Test in one year with (6.21 – 1.83 + 5.27) = 9.65 runs. In contrast, in the period from fifteen to sixteen years old, a sub-elite male player will increase ‘only’ with (6.21 – 1.83 + 5.27 – 5.11) = 4.54 runs. An elite girl is predicted to achieve an extra (6.21 – 1.83) = 4.38 runs whereas a sub-elite girl will run (6.21 – 1.83 – 5.11) = 0.73 runs less according to the model.

In Figure 5.2 the data are represented for the four different gender and performance groups. In the figure, the lines connect two or three individual yearly observations; the points are single individual observations. The bold solid lines depict the estimated mean ISRT score

80

for “average” representatives of each group, i.e., with mean scores on percentage body fat, additional training hours and motivation (8.65, 3.82, and 4.35 for elite boys; 9.15, 3.36, and 4.2 for sub-elite boys; 20.0, 2.84, and 4.53 for elite girls, and 21.6, 1.94, and 4.11 for sub-elite girls, respectively).

The dotted lines around the bold line indicate the 95% confidence region taking into account between-individual (level 2) variation. This variation is estimated by the level 2 variance of 136 (see Table 5.2), which is equivalent to a standard deviation of approximately 12 runs, indicating rather large differences between individuals as is also apparent from Figure 5.2.

The curvature of the lines represents the fitted second-order polynomial (quadratic) model. It can be observed that the linear effect of the model is most strong for the elite boys and least strong for sub-elite girls, and approximately equal for sub-elite boys and elite girls (due to the interaction effects with age and sub-elite). Also visible from the figure is the rather large within-individual (level 1) variance, which is much larger for boys than for girls, estimated as 292.8 (equivalent to a standard deviation of about 17 runs) and 105.9 (standard deviation about 10 runs), respectively.

12 14 16 180

40

80

120

Female elite

ISR

T

age (yrs)12 14 16 18

0

40

80

120

Female sub-elite

ISR

T

age (yrs)

12 14 16 180

40

80

120

Male elite

age (yrs)

ISR

T

12 14 16 180

40

80

120

Male sub-elite

ISR

T

age (yrs)

Figure 5.2. Predicted curves of the interval endurance capacity for elite boys, sub-elite boys, elite girls, and sub-elite girls.

Chapter I II III IV V VI VII VIII

81

5.4 Discussion

Talented field hockey players of twelve years old score on average 35 runs on the Interval Shuttle Run Test, regardless whether they are a boy or a girl, an elite or a sub-elite player. Only elite girls score on average 10 runs less, which may indicate that at the age of twelve years it is still possible for talented girls in field hockey to compensate a relatively low interval endurance capacity with other performance characteristics, such as great technique and tactics.

During adolescence, differences between boys and girls become apparent. Boys have a much faster development of their interval endurance capacity than girls but also within the male and female group differences are remarkable. At the age of fifteen, elite boys score on average 15 runs more than sub-elite boys (85 versus 70 runs). At the age of eighteen this difference has grown to 30 runs (100 versus 70 runs) because elite boys still improve in contrast to sub-elite boys who seem to remain relatively stable. Although elite girls start of with a lower score on the ISRT when they are twelve years old, they catch up with sub-elite girls at the age of about fourteen. At fifteen, they are already better and they keep improving themselves. After about seventeen years of age they seem to remain relatively stable. The curve of the elite girls resembles that of the sub-elite boys to a high degree in contrast to the sub-elite girls who increase their number of runs until the age of fifteen and decrease afterwards until they fall back to the levels of twelve-year-olds again.

During the Interval Shuttle Run Test for interval endurance capacity both the aerobic and anaerobic energy production contribute to the total energy requirement (Lemmink and Visscher, 2003). In a ‘normal’ population of adolescents, boys increase their aerobic and anaerobic performance with age whereas girls improve to 14-15 years with a gradually decrease afterwards (e.g., Martin and Malina, 1998; Kemper and Koppes, 2004). The development of the interval endurance capacity of talented youth field hockey players, however, is not quite the same as that of ‘normal’ adolescents. Instead of decreasing their performance after the age of fifteen, elite girls are able to sustain the improvement of their interval endurance capacity. Although boys and sub-elite girls seem to follow the ‘normal’ pattern, elite boys improve themselves more than sub-elite boys on the interval endurance capacity.

Although small (one hour extra training represents only one extra run on the Interval Shuttle Run Test), additional training was found to have a significant effect in improving the model. The explanation that we did not find such effect for field hockey training may be that during field hockey specific training more attention is paid to improving other aspects of a field hockey performance, such as technique and tactics than to endurance capacity. We found a large variation in additional training between players. In some cases the standard deviation

82

was greater than the amount of additional training itself. Apparently there are major differences concerning the amount of additional training between talented players, as well within the elite group as within the sub-elite group.

Motivation was found to have a significant effect in improving the model. The Interval Shuttle Run Test to measure the interval endurance capacity is a maximal test. The intense activity needed in this test causes uncomfortable side effects such as fatigue and muscle soreness, and a player has to be very motivated to continue running until exhaustion. Motivation can be defined as the direction and intensity of one’s effort. The direction of behavior indicates whether an individual approaches or avoids a particular situation and the intensity of behavior relates to the degree of effort put forth to accomplish the behavior (Silva and Weinberg, 1984). Possibly, motivation in sports is one of the greatest differences between ‘normal’ adolescents and talented field hockey players. The latter are more motivated to get the best out of themselves and elite youth players want this even more than sub-elite youth players. Since the road to the top is long, motivation is not only essential for current performance in a match or test, but also in talent development. Talented players have to devote long hours of training for many years in a row in order to improve their performance level (Ericsson et al., 1993; Ericsson, 1996).

Height and lean body mass gave no significant improvement of our model for the development of the interval endurance capacity. Therefore, any difference between a ‘normal’ population of adolescents and this population cannot be explained from these anthropometric variables. However, we did find a significant negative effect for percentage body fat. This is in line with a study on young male gymnasts, swimmers, soccer, and tennis players (Baxter-Jones et al., 1995). Training did not appear to have affected these young athletes’ growth and development. However, training can have an effect on percentage body fat (e.g., Astrand et al., 2003).

There is a rather large variation in interval endurance capacity within and between players. The within-persons variation, i.e. the variance between measurements, is noticeable especially in boys and elite girls. This variation was based upon those players that have been tested repeatedly, which is less than half of the population. Consequently, some bias in this random effect may have occurred, possibly overestimating it. Since previous research underscored the reliability of the ISRT, we do not doubt the reliability of the test (Lemmink et al., 2004a). However, we do not have a clear alternative explanation for this phenomenon. It might have to do with the moment of testing. At the end of the season, the most important matches are played and when players are tested a couple of days before an important match, they might be inclined to take it easy at the test. Other explanations might be differences in weather conditions or previous training sessions for which it was impossible to control for.

Chapter I II III IV V VI VII VIII

83

The between-persons variation is based on the total population and is distinct in the total age-band of 12-19 years. Evidently, a field hockey performance can be broken down into many multidimensional performance characteristics, from which the interval endurance capacity is only one (Nieuwenhuis et al., 2002; Elferink-Gemser et al., 2004). The combination of anthropometric, physiological, technical, tactical, and psychological characteristics results in a player’s performance level (Elferink-Gemser et al., 2004). In their young adolescent years, players still can compensate for less developed performance characteristics such as their interval endurance capacity. However, towards expertise performance demands increase and all players need to meet high values for all performance characteristics, including the interval endurance capacity. Therefore, it is possible for sub-elite players to possess a great interval endurance capacity, for example because they spend a lot of time to additional training, but when lacking a high level of other performance characteristics they will not be able compete at the highest performance level after all.

In sum, the development of the interval endurance capacity of 12-19 year-old talented field hockey players can be modeled with a polynomial model of order 2 with gender- and performance level specific intercepts and linear age terms as well as different level 1 variances for boys and girls. Differential effects of performance level for age and gender significantly improved the model. Results show that during adolescence both male and female elite youth players have, on average, a more promising development pattern of their interval endurance capacity than sub-elite youth players. After taking into account the effect of percentage body fat, additional training hours, and motivation, the remaining differences between individual players are considerable.

84

References

Astrand, P.O., Rodahl, K., Dahl, H.A., and Stromme, S.B.( 2003). Textbook of work physiology: Physiological bases of exercise. 4th edition, New York: McGraw-Hill.

Aziz, A.R., Chia, M., and Teh, K.C. (2000). The relationship between maximal oxygen uptake and repeated sprint performance indices in field hockey and soccer players. The Journal of Sports Medicine and Physical Fitness, 40, 195-200.

Bakker, F. (1995). Ontwikkeling van tests op het terrein van de sportpsychologie in Nederland en Vlaanderen. [Development of tests for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 6, 3-15.

Baxter-Jones, A.D.G., Helms, P., Maffulli, N., Baines-Preece, J.C., and Preece, M. (1995). Growth and development of male gymnasts, swimmers, soccer and tennis players: a longitudinal study. Annals of Human Biology, 22, 381-394.

Bhanot, J.L. and Sidhu, L.S. (1983). Maximal anaerobic power in Indian national hockey players. British Journal of Sports Medicine, 17, 34-39.

Bitar, A., Vernet, J., Coudert, J., and Vermorel, M. (2000). Longitudinal changes in body composition, physical capacities and energy expenditure in boys and girls during the onset of puberty. European Journal of Nutrition, 39, 157-163.

Boyle, P.M., Mahoney, C.A., and Wallace, W.F. (1994). The competitive demands of elite male field hockey players. Journal of Sports Medicine and Physical Fitness, 34, 235-241.

Companjen, T. and Bakker, F. (2003). Overzicht van sportpsychologische vragenlijsten in Nederland en Vlaanderen. [Outline of questionnaires for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 14, 34-41.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. (2004). Relation between multidimensional performance characteristics and level of performance in talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. (pending minor revisions, 2005). Multidimensional performance characteristics and performance level in talented youth field hockey players. A longitudinal study. Journal of Sport Sciences.

Ericsson, K.A. (1996). The acquisition of expert performance: An introduction to some of the issues. In The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games (edited by K.A. Ericsson), pp. 1-50. Mahwah, NJ: Lawrence Erlbaum Associates.

Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406.

Goldstein, H., Rasbash, J., Plewis, I., Draper, D., Browne, W., Yanh, M., Woodhouse, G., and Healy, M. (1998). A user’s guide to MlwiN. London: Multilevel Models Project, Institute of Education, University of London.

Ghosh, A. K., Goswami, A., Mazumdar, P., and Mathur, D. N. (1991). Heart rate & blood lactate response in field hockey players. Indian Journal of Medical Research, 94, 351-356.

Krahenbuhl, G.S., Skinner, J.S., and Kohrt, W.M. (1985). Developmental aspects of maximal aerobic power in children. Exercise and Sport Sciences Reviews, 13, 503-538.

Kemper, H.C.G. (2004). Amsterdam Growth and Health Longitudinal Study. Medicine and Sports Science, 47. Basel, Karger.

Kemper, H.C.G. and Koppes, L.L.J. (2004). Is physical activity important for aerobic power in young males and females? In Amsterdam Growth and Health Longitudinal Study (edited by H.C.G. Kemper), pp. 153-166. Medicine and Sports Science, volume 47. Basel: Karger.

Lemmink, K.A.P.M. and Visscher, C. (2003). The relationship between the Interval Shuttle Run Test and maximal oxygen uptake in soccer players. Journal of Human Movement Studies, 45, 219-232.

Lemmink, K.A.P.M., Verheijen, R., and Visscher, C. (2004b). The discriminative power of the Interval Shuttle Run Test and the Maximal Multistage Shuttle Run Test for playing level of soccer. Journal of Sports Medicine and Physical Fitness, 44, 233-239.

Lemmink, K.A.P.M., Dolleman, G., Verheijen, R., and Visscher, C. (2000). Interval Sprint Test en Interval Shuttle Run Test – betrouwbaarheid en discriminerend vermogen van twee nieuwe

Chapter I II III IV V VI VII VIII

85

voetbaltests. [Interval Sprint Test and Interval Shuttle Run Test – reliability and discriminative power of two new tests for soccer players]. Geneeskunde en Sport, 33, 39-48.

Lemmink, K.A.P.M., Visscher, C., Lambert, M.I., and Lamberts, R. (2004a). The Interval Shuttle Run Test for intermittent sport players: evaluation of reliability. Journal of Strength and Conditioning Research, 18, 821-827.

Lothian, F. and Farrally, M. (1994). A time-motion analysis of women’s hockey. Journal of Human Movement Studies, 26, 255-265.

Maas, C. J. M. and Snijders, T. A. B. (2003). The multilevel approach to repeated measures for complete and incomplete data. Quality & Quantity, 37, 71-89.

Mahoney, M.J., Gabriel, T.J., and Perkins, T.S. (1987). Psychological skills and exceptional athletic performance. The Sport Psychologist, 1, 181-199.

Martin, J.C. and Malina, R.M. (1998). Developmental variations in anaerobic performance associated with age and sex. In Pediatric Anaerobic Performance (edited by E. Van Praagh), pp. 45-64. Champaign, IL: Human Kinetics.

Nieuwenhuis, C.F., Spamer, E.J., and Van Rossum, J.H.A. (2002). Prediction function for identifying talent in 14- to 15-year-old female field hockey players. High Ability Studies, 13, 21-33.

Nunez, C., Callagher, D., Visser, M., Pi-Sunyer, F.X., Wang, Z., and Heymsfield, S.B. (1997). Bioimpedance analysis: evaluation of leg-to-leg system based on pressure contact footpad electrodes.

Medicine and Science in Sports and Exercise, 29, 524-531. Patel, D.R., Stier, B., and Luckstead, E.F. (2002). Major international sport profiles. Pediatric Clinics

of North America, 49, 769-792. Powers, S.K. and Howley, E.T. (2001). Exercise physiology: Theory and application to fitness and

performance. 4th edition, New York: McGraw-Hill. Reilly, T. and Borrie, A. (1992). Physiology applied to field hockey. Sports Medicine, 14, 10-26. Silva, J.M. III and Weinberg, R.S. (1984). Psychological foundations of sport. Human Kinetics:

Champaign, IL. Snijders, T. A. B. and Bosker, R. J. (2000). Multilevel analysis. An introduction to basic and advanced

multilevel modelling. London: Sage Publications.

Chapter VI

Psychological characteristics of talented youth athletes in field hockey, basketball, volleyball, speed skating, and swimming Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. The Sports Psychologist (in revision) Acknowledgements:

This study has been supported by a grant of the Dutch National Olympic Committee NOC*NSF. The authors wish to express their sincere appreciation to Joke Tissingh, Karen Oldenziel, Yvonne van Heijzen, and Alien van der Sluis for their assistance in this research project.

88

Abstract

To reveal the relationship between psychological skills and performance level within atalent group, 458 talented youth athletes (age 14.8 years, sd = 1.5) filled in the Dutch Youth Version of the Psychological Skills Inventory for Sports with scales for motivation, confidence, anxiety control, mental preparation, team emphasis, and concentration. A performance level (elite versus sub-elite) by type of sport (team sports versus individual sports) by gender multivariate analysis of covariance (2 X 2 X 2) with age as a covariate resulted in significant effects. Psychological skills distinguished between more and less successful talented athletes, especially in females. In general, psychological profiles differed between males and females and between team sport athletes and individual sport athletes. However, for discrimination of elite and sub-elite youth athletes, motivation and mental preparation were useful indicators that are independent of gender and type of sport.

Chapter I II III IV V VI VII VIII

89

6.1 Introduction

Elite athletes repeatedly have to perform under high pressure, and it is therefore not surprising that psychological characteristics often distinguish those successful at the highest standard from their less successful counterparts (Morris, 2000). Early research evidence already supported an association between psychological characteristics and sports performance (Morgan and Pollock, 1977; Morgan, 1979; May et al., 1985). Further research evolved with an emphasis in identifying psychological skills relevant to sport (Meyers et al., 1996). Mahoney et al. (1987) identified potential constructs assessing motivation, confidence, anxiety control, mental preparation, team emphasis and concentration. They developed an instrument that assesses a broad range of psychological skills possessed by athletes and moreover is sport-specific: the Psychological Skills Inventory for Sport (PSIS-R-5). Compared to non-elite athletes, elite athletes reported that they were more motivated to do well in their sport, were more self-confident, experienced fewer problems with anxiety, relied more on internally referenced and kinesthetic mental preparations, were more focused on their own performance than that of their team, and were more successful at deploying their concentration (Mahoney et al., 1987; Mahoney, 1989). So far, many other researchers have also distinguished successfully elite from non-elite athletes on the basis of their psychological skills. For example, Grossarth-Maticek et al. (1990) described psychological factors as determinants of success in football and boxing. Meyers and colleagues (1996) reported better scores for elite rodeo athletes than non-elite ones on motivation, confidence, anxiety control and concentration, whereas in a study on Chinese track and field athletes, Cox et al. (1996) found elite athletes outscoring collegiate level athletes on confidence and anxiety control.

It is not self-evident that the relation between psychological skills and performance level is similar for different types of sports or for males and females. Various studies have indicated, for example, that differences exist in psychological skills between individual and team sports (Feltz and Ewing, 1987; Mahoney et al., 1987; Cox and Liu, 1993) and between the genders (White, 1993; Chantal et al., 1996; Sewell and Edmondson, 1996; MacIntyre et al., 1998).

In addition, a relation between psychological skills and performance level has been found within the highest performance level, i.e. when elite and sub-elite athletes are compared to each other. Orlick and Partington (1988) reported that among physical, technical and mental characteristics, mental readiness provided the only statistically significant link with final Olympic ranking of Canadian Olympians. However, it seems that differences are smaller when elite athletes are compared to sub-elite athletes rather than to non-elite ones. In a study on equestrian athletes, elite athletes scored higher than sub-elite athletes on only two of six psychological skills from the PSIS-R-5 (Meyers et al., 1999) whereas Meyers and colleagues

90

(1994) found no differences in psychological skills between top-ranked (1 to 65), middle-ranked (75-180), and bottom-ranked (200+) female world-ranked tennis players.

So far, it is not yet clear whether the same psychological variables that distinguish elite from non-elite or elite from sub-elite athletes in adulthood are important for outstanding performance throughout the process of talent development (Morris, 2000). To assist young athletes in reaching elite level, it is important to gain insight into factors that influence the development of a successful sports career, such as their psychological skills. However, as far as the authors know, no studies have focused primarily on the relation between psychological skills and performance level within a talent group. Therefore, this study concentrates on athletes that have been identified as talent but who have not yet reached the top in adult elite sports. Two different performance-level groups within a group of all-talented athletes were compared on psychological skills. The goal of this study was to reveal the relationship between psychological skills and performance level with possible effects of type of sport and gender in talented youth athletes.

6.2 Methods

Participants A total of 458 talented youth athletes (age 14.8 years, sd = 1.5), all of whom participate in high-level competitive sports in the Netherlands, filled in the Dutch Youth Version of the Psychological Skills Inventory for Sports (Appendix 6.1). Among them were 124 field hockey players (62 male and 62 female), 54 basketball players (30 male and 24 female), 121 volleyball players (59 male and 62 female), 72 speed skaters (41 male and 31 female) and 87 swimmers (52 male and 35 female). The participants were divided into 148 elite athletes and 310 sub-elite athletes based on their performance level. Feltz and Ewing (1987) suggest that an elite-level young athlete can be defined as one who has competed in national-level competitions and has participated in his/her sport for at least 2 years. In our study, all talented youth athletes met these conditions. In team sports, elite youth athletes distinguished themselves from sub-elite athletes by being part of an extra selection team (field hockey: national or district youth selection team; basketball: national youth selection team; volleyball: national youth selection team). In speed skating and swimming, elite youth athletes distinguished themselves from sub-elite athletes by being among the 12 best of their age category in the Netherlands.

Chapter I II III IV V VI VII VIII

91

Procedure All athletes gave their informed consent prior to participation and completed the inventory individually in a group setting. Instructions were standardized, since obtained scores may be influenced by changing test instructions (Nideffer, 1987; Greenspan et al., 1988). To allow mutual comparisons between athletes of different ages, athletes were asked to compare themselves with top athletes in their age category. To avoid socially desirable answers, athletes were told that the results were being used solely for research purposes. Instrument The Psychological Skills Inventory for Sports (PSIS-R-5) consists of 5-point Likert type items that are distributed over 6 scales (Mahoney et al., 1987). The PSIS-R-5 has been translated into Dutch and subjected to psychometric testing (Bakker, 1995; Companjen and Bakker, 2003). The Dutch Youth Version of the Psychological Skills Inventory for Sports (PSIS–Youth) is based upon the Dutch version of the PSIS-R-5, but the formulation of questions is simpler. It contains 44 5-point Likert type items, distributed over the same 6 scales as the PSIS-R-5: Motivation (8 items), Confidence (8 items), Anxiety Control (8 items), Mental Preparation (6 items), Team Emphasis (7 items) and Concentration (7 items) (see Appendix 6.1). The answer almost never equates to 1, and almost always to 5. Items worded negatively (indicating a problem or concern) are transformed by reversing the aforementioned 1-5 format. In this way, a high score on each scale corresponds to the psychological skill being present to a large extent.

Psychometric characteristics In a study at our Center on the psychometric characteristics of the PSIS-Youth, 381 youth field hockey and soccer players (age 14.7 years, sd = 1.7; 32% female, 68% male) filled in the questionnaire (Elferink-Gemser et al., internal publication 2002). Correlations between scales did not exceed 0.42, supporting the PSIS-Youth as a measure of six relatively independent constructs. Internal consistency estimates for each scale were acceptably high, ranging from 0.68 on the Team Emphasis scale to 0.81 on the Confidence scale. Apart from the Team Emphasis scale, Cronbach’s alpha was above 0.70, which is the minimum level recommended for research purposes (Nunnally, 1978). These internal consistency estimates are in line with other studies using the PSIS-R-5. White and Croce (1992) found Cronbach alpha reliability scores ranging from 0.69 to 0.77. White (1993) likewise showed good internal consistency with alpha coefficients ranging from 0.67 to 0.84, while Meyers et al. (1994), using discriminant analysis, successfully classified 84% of selected athletes into rank order using the results of the questionnaire. By contrast, Chartrand et al. (1992) did note internal consistency problems, with the exception of the confidence factor. This however was an

92

isolated result, which appears to stand opposed to most of the evidence (MacIntyre et al., 1998). Data analysis According to the six categories of psychological skills (motivation, confidence, anxiety control, mental preparation, team emphasis and concentration), mean scores and standard deviations were calculated for the eight different subgroups based on performance level (elite youth athletes and sub-elite youth athletes), type of sport (team sports and individual sports) and gender. To make mutual comparisons between scales possible, scores on each of the six scales are also presented as means on the 5-point Likert scale (minimum score = 1; maximum score = 5) ± standard deviation. Because of the nature of competition of speed skating and swimming in the Netherlands, in which an emphasis is placed on individual performance, the questions in the Team Emphasis scale are not valid for the individual sport athletes in this study (e.g., “I get very frustrated when a teammate is performing poorly”). Consequently, only athletes from team sports answered questions in this scale.

Data were analyzed using multivariate analysis of covariance (MANCOVA) general linear models (GLM) procedure. As part of the GLM procedure, least-squares means are calculated. For the MANCOVA, performance level, type of sport and gender served as the independent variables, while the categories of psychological skills served as the multivariate dependent variable. Age was considered as a covariate since the relationship between psychological skills and performance level may change with age. In this way, each variable was adjusted for age.

Univariate analyses of covariance (ANCOVA) with factors of performance level, type of sport and gender and with age as a covariate were carried out separately for each psychological variable, with follow-up analyses to clarify the source and nature of significant relationships. The ANCOVA for the Team Emphasis scale was conducted with scores of the team sport athletes only. An alpha of 0.05 was adopted for all tests of significance. 6.3 Results

A performance level by type of sport by gender multivariate analysis of covariance (2 X 2 X 2) resulted in significant main effects for performance level [F (5,445) = 5.18, p < 0.01]; type of sport [F (5,445) = 23.90, p < 0.01] and gender [F (5,445) = 9.70, p < 0.01]. Table 6.1 displays the means of the psychological skills for categories of performance level, type of sport and gender.

Tab

le 6

.1.

Mea

n sc

ores

(sd)

of t

he p

sych

olog

ical

skill

s as a

func

tion

of p

erfo

rman

ce le

vel,

type

of s

port,

and

gen

der (

N =

458

).

M

ale

athl

etes

Fe

mal

e at

hlet

es

T

eam

spor

ts

Indi

vidu

al sp

orts

T

eam

spor

ts

Indi

vidu

al sp

orts

Elite

n

= 60

Su

b-el

ite

n =

91

Elite

n

= 31

Su

b-el

ite

n =

62

Elite

n

= 38

Su

b-el

ite

n =

110

Elite

n

= 19

Su

b-el

ite

n =

47

Mot

ivat

ion

(MV

)

Sc

ale

scor

e 36

.74

(2.6

8)

34.9

5 (3

.56)

35

.84

(2.9

6)

34.7

4 (3

.50)

37

.08

(2.1

2)

34.3

1 (4

.12)

35

.16

(2.3

2)

33.6

7 (3

.21)

5-

poin

t Lik

ert S

cale

scor

e 4.

59 (0

.34)

4.

37 (0

.45)

4.

48 (0

.37)

4.

34 (0

.44)

4.

64 (0

.27)

4.

29 (0

.52)

4.

40 (0

.29)

4.

21 (0

.40)

C

onfid

ence

(CF)

Sc

ale

scor

e 31

.99

(4.8

0)

31.7

7 (4

.80)

31

.84

(4.4

3)

32.5

7 (4

.76)

30

.00

(4.7

0)

27.6

7 (4

.66)

30

.00

(4.9

3)

31.0

4 (5

.20)

5-

poin

t Lik

ert S

cale

scor

e 4.

00 (0

.60)

3.

97 (0

.60)

3.

98 (0

.55)

4.

07 (0

.60)

3.

75 (0

.59)

3.

46 (0

.58)

3.

75 (0

.62)

3.

88 (0

.65)

A

nxie

ty C

ontr

ol (A

X)

Scal

e sc

ore

32.3

8 (4

.42)

32

.10

(4.5

3)

32.3

2 (5

.22)

30

.74

(4.5

9)

31.2

7 (4

.21)

31

.13

(4.4

7)

29.6

3 (4

.94)

29

.31

(5.4

6)

5-po

int L

iker

t Sca

le sc

ore

4.06

(0.5

4)

4.01

(0.5

7)

4.04

(0.6

5)

3.84

(0.5

7)

3.91

(0.5

3)

3.89

(0.5

6)

3.70

(0.6

2)

3.66

(0.6

8)

Men

tal P

repa

ratio

n (M

P)

Scal

e sc

ore

14.7

0 (4

.15)

14

.40

(4.6

4)

19.5

5 (4

.67)

17

.66

(4.8

1)

14.5

6 (4

.17)

12

.75

(4.9

7)

17.1

1 (4

.63)

16

.97

(3.8

4)

5-po

int L

iker

t Sca

le sc

ore

2.44

(0.6

9)

2.40

(0.7

7)

3.26

(0.7

8)

2.94

(0.8

0)

2.43

(0.7

0)

2.12

(0.8

3)

2.85

(0.7

7)

2.83

(0.6

4)

Tea

m E

mph

asis

(TM

)

Sc

ale

scor

e 24

.45

(3.0

2)

25.1

1 (3

.17)

25

.37

(3.0

9)

23.8

5 (2

.94)

5-

poin

t Lik

ert S

cale

scor

e 3.

50 (0

.43)

3.

59 (0

.45)

3.

62 (0

.44)

3.

41 (0

.42)

C

once

ntra

tion

(CC

)

Sc

ale

scor

e 24

.68

(3.4

7)

24.5

7 (3

.99)

25

.26

(3.9

5)

26.6

0 (3

.54)

25

.57

(2.9

6)

24.0

0 (3

.40)

27

.00

(3.7

7)

26.0

5 (3

.60)

5-

poin

t Lik

ert S

cale

scor

e 3.

53 (0

.50)

3.

51 (0

.57)

3.

61 (0

.56)

3.

80 (0

.50)

3.

65 (0

.42)

3.

43 (0

.49)

3.

86 (0

.54)

3.

72 (0

.51)

94

Performance level In the relation of psychological skills and performance level we found significant main effects for motivation and mental preparation (Table 6.2). We also found significant interaction effects for confidence (performance level by type of sport), team emphasis (performance level by gender) and concentration (performance level by gender). Table 6.2. Summary of univariate F-Ratios calculated using Type III sums of squares with

Hypothesis df = 1 and Error df = 449 for MV, CF, AX, MP, CC and Error df = 294 for TM (General Linear Model).

Psychological Skills

MV CF AX MP TM CC

Performance level (P) 23.06** 0.62 1.55 3.56* 1.24 0.95

Type of sport (T) 6.36** 4.15* 6.22** 56.95** 15.65**

Gender (G) 1.73 26.28** 10.04** 4.98* 0.17 0.63

P X T 1.87 3.92* 0.47 0.01 1.66

P X G 0.89 0.73 0.57 0.01 7.33** 5.81**

T X G 0.93 1.37 0.94 0.37 0.26

P X T X G 0.15 1.83 0.26 2.53 0.23

Note: * p < 0.05. ** p < 0.01.

Elite athletes scored higher than sub-elite athletes on motivation and mental preparation, irrespective of gender or type of sport. Regardless of gender, elite athletes also scored higher than sub-elite athletes on confidence, but this can only be applied to team sports (p < 0.01). No significant differences between elite and sub-elite athletes were found in individual sports (p > 0.05). On team emphasis, female elite athletes had better scores than female sub-elite athletes (p < 0.01), but male elite and sub-elite athletes had similar scores (p > 0.05). On concentration, again female elite athletes scored better than female sub-elite athletes (p < 0.01), whereas male scores of elite and sub-elite athletes did not differ significantly (p > 0.05). These results can be applied to both team and individual sport athletes. Type of sport In the relation of psychological skills and type of sport, we found significant main effects for motivation, confidence, anxiety control, mental preparation and concentration (Table 6.2). Irrespective of performance level or gender, team sport athletes had higher scores than individual sport athletes on motivation and anxiety control, whereas on confidence, mental preparation and concentration individual athletes outscored team sport athletes. We also found

Chapter I II III IV V VI VII VIII

95

a significant interaction effect for confidence (type of sport by performance level). Among elite athletes, team sport athletes had scores similar to individual sport athletes (p > 0.05), whereas among sub-elite athletes individual sport athletes outscored team sport athletes (p < 0.01). This can be applied to males as well as females. Gender In the relation of psychological skills and gender, we found significant main effects for confidence, anxiety control and mental preparation (Table 6.2). Regardless of performance level or type of sport, male athletes outscored female athletes in all these psychological skills. We also found significant interaction effects for team emphasis and concentration (both gender by performance level). Among elite athletes, females and males had similar scores on team emphasis (p > 0.05), but in the sub-elite group male athletes scored better than female athletes (p < 0.01). Elite female athletes outscored their male counterparts on concentration (p < 0.05), while no significant differences based on gender were found among sub-elite athletes (p > 0.05). These results can be applied to both team and individual sport athletes. 6.4 Discussion

The goal of this study was to reveal the relationship between psychological skills and performance level with possible effects of type of sport and gender in talented youth athletes. To accomplish this purpose, two different performance level groups within a group of all-talented athletes were compared on psychological skills.

The relation between motivation and performance level, i.e. that elite youth athletes outscore sub-elite ones is in general congruence with studies examining differences between elite and sub-elite players in adulthood (e.g., Smith and Christensen, 1995; Chantal et al., 1996). It is however unclear whether athletes perform better because of their high motivation or whether they are more motivated because of their high performance level. The relation between mental preparation and performance level favoring elite athletes in contrast to sub-elite ones was also found in a study on golf players in which skilled golfers reported greater mental preparation than less skilled ones (Thomas and Over, 1993).

The relation between psychological skills and performance level is different in male than in female athletes. Male elite athletes outscore sub-elite athletes on motivation and mental preparation only, female elite youth athletes distinguish themselves from their sub-elite counterparts by their higher scores on four of six psychological skills as measured with the PSIS-Youth (motivation, mental preparation, team emphasis and concentration). In team sports, confidence can be added to this list.

96

The relation between psychological skills and performance level is also different in team sport and individual sport athletes. In team sports, elite youth athletes outscored the sub-elite athletes on confidence, whereas all individual sport athletes scored relatively high on this scale, taking the highest possible scores into account. Weinberg and Gould (1999) stated that less confident athletes doubt whether they are good enough or whether they have what it takes to be successful. Positive feedback about their performance is thought to build confidence. Duda and Nicholls (1992) also stated that confidence plays an important role in beliefs regarding success in sports. In individual sports, the athlete gets feedback individually most of the time as opposed to team sports, in which feedback is mostly presented to the team as a whole. Only the best performers get positive feedback individually, which may be an explanation for the difference in confidence scores between elite and sub-elite team sport athletes found in this study. In addition, team sports are characterized by a lack of objective performance measurements, making it hard to give feedback. Unlike individual sports, in which there is a unidimensional performance criterion like time or distance, a performance in team sports depends on the combination of numerous mini-performances of the player and his teammates (Régnier et al., 1993).

Comparably to our study, Cox and Liu (1993) found that those athletes exhibiting the highest levels of mental preparation were the individual sport athletes. They concluded that this might be due to the fact that individual sport athletes do not have the luxury of being able to rely on their teammates. Another explanation could relate to the character of the individual sports in this study. Speed skating and swimming are cyclic sports in which the same movement pattern is repeated frequently. During mental preparation, this movement pattern can be practiced in one’s head. In team sports like field hockey, basketball and volleyball, environmental characteristics change constantly, which makes mental preparation more difficult.

Noise and sounds during training and competition are distracters that can complicate an athlete’s concentration. These distracters are part of most team sports, whereas more quiet environments are expected for most individual sports (Weinberg and Gould, 1999). Individual sport athletes may therefore have better environmental circumstances to concentrate. This is in congruence with the results obtained in our study, in which team sport athletes tended to report significantly lower concentration than individual sport athletes. Thus it seems logical that concentration and mental preparation are related to each other (Weinberg and Gould, 1999), supporting our findings that individual sport athletes outscore team sport athletes in both concentration and mental preparation.

Average motivation scores of all athletes were very high in that they surpassed 4.0 on a 5-point Likert scale. Hence all talented athletes have a relatively high motivation related to

Chapter I II III IV V VI VII VIII

97

sports. Before reaching the top, an athlete has to invest many years in training. According to Ericsson et al. (1993), only those who are committed to their sport can persist in deliberate practice. Therefore athletes have to be highly motivated if they are to have a chance of becoming elite athletes. This supports what has been reported in other research with talented children in disciplines other than sports (e.g., Bloom, 1985).

In contrast to their high motivation scores, all athletes had relatively low mental preparation scores in comparison to the other scales. This confirms a statement of Reilly (1996), who notes that “an increasing minority of soccer players are now paying attention to psychological preparation”. In their study on professional sport psychology in Ireland, MacIntyre et al. (1998) also reported low levels of mental preparation in top athletes. Since mental preparation distinguishes elite from sub-elite athletes, it seems valuable to give more attention during training to developing this psychological skill in talented youth athletes.

To provide them with an external reference point, all participants were told to compare themselves with top athletes in their age category. Although this may be difficult, we think it is easier than comparing themselves with adult elite athletes. Because of this age-bound external reference point, we did not draw conclusions for different age groups. It is also interesting to gain insight into the development of psychological skills though. This is possible only when all athletes are provided with the same external reference point, e.g. the absolute top in adult sports.

Researchers who focus on talent development in sports often acknowledge that a world-class performance is the result of several factors (e.g., Deshaies et al., 1979; Régnier et al., 1993; Reilly et al., 2000; Elferink-Gemser et al., 2004). Accordingly, it is recommended to relate the described relationships between psychological skills and performance level to other performance characteristics, such as an athlete’s physiological, technical and tactical characteristics. Only by adopting a multidisciplinary design can the relative contribution of psychological skills to performance level be made clear. Nonetheless, from this study it becomes clear that psychological skills can distinguish between more and less successful talented athletes, especially among females. Psychological profiles differ between males and females, and between team sport athletes and individual sport athletes. However, for differentiation purposes between elite and sub-elite athletes within a talent group, motivation and mental preparation are useful indicators that are independent of type of sport and gender.

98

References

Bakker, F. (1995). Ontwikkeling van tests op het terrein van de sportpsychologie in Nederland en Vlaanderen. [Development of tests for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 6, 3-15.

Bloom, B.S. (1985). Generalizations about talent development. In Developing talent in young people (edited by B.S. Bloom), pp. 507-549. New York: Ballantine Books.

Chantal, Y., Guay, F., Dobreva-Martinova, T., and Vallerand, R.J. (1996). Motivation and elite performance: An exploratory investigation with Bulgarian athletes. International Journal of Sport Psychology, 27, 173-182.

Chartrand, J.M., Jowdy, D.P., and Danish, S.J. (1992). The Psychological Skills Inventory for Sports: Psychometric Characteristics and Applied Implications. Journal of Sport & Exercise Psychology, 14, 405-413.

Companjen, T. and Bakker, F. (2003). Overzicht van sportpsychologische vragenlijsten in Nederland en Vlaanderen. [Outline of questionnaires for the field of sport psychology in the Netherlands and Belgium]. Sportpsychologie Bulletin, 14, 34-41.

Cox, R.H. and Liu, Z. (1993). Psychological skills: A cross-cultural investigation International Journal of Sport Psychology, 24, 326-340.

Cox, R.H., Liu, Z., and Qiu, Y. (1996). Psychological skills of elite athletes. International Journal of Sport Psychology, 27, 123-132.

Deshaies, P., Pargman, D., and Thiffault, C. (1979). A psychobiological profile of individual performance in junior hockey players. In Psychology of Motor Behaviour and Sport – 1978 (edited by G.C. Roberts and K.M. Newell), pp. 36-50. Champaign, IL: Human Kinetics.

Duda, J.L. and Nicholls, J.C. (1992). Dimensions of achievement motivation in schoolwork and sport. Journal of Educational Psychology, 84, 290-299.

Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. (2002). ‘De Psychological Skills Inventory for Sports’ aangepast voor Nederlandse jeugd (PSIS-jeugd): psychometrische eigenschappen. [The Psychological Skills Inventory for Sports adapted for Dutch youth (PSIS-Youth): psychometric characteristics]. Internal publication, Center for Human Movement Sciences, University of Groningen, The Netherlands.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. (2004). Relation between multidimensional performance characteristics and level of performance level in talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406.

Feltz, D.L. and Ewing, M.E. (1987). Psychological characteristics of elite young athletes. Medicine and Science in Sports and Exercise, 19, 98-105.

Greenspan, M.J., Murphey, S.M., Tammen, V., and Jowdy, D. (1988). Effects of Athlete Achievement Level and Test Administration Instructions on the Psychological Skills Inventory for Sports (PSIS). Annual Conference of the Association for the Advancement of Applied Sport Psychology, Nashua, NH.

Grossarth-Maticek, R., Eysenck, H.J., Rieder, H., and Rakic, L. (1990). Psychological factors as determinants of success in football and boxing: the effects of behaviour therapy. International Journal of Sport Psychology, 21, 237-255.

MacIntyre, T., Mahoney, C., and Moran, A. (1998). Professional sport psychology in Ireland. Irish Journal of Psychology, 19, 504-515.

Mahoney, M.J. (1989). Psychological predictors of elite and non-elite performance in Olympic weightlifting. International Journal of Sport Psychology, 20, 1-12.

Mahoney, M.J., Gabriel, T.J., & Perkins, T.S. (1987). Psychological skills and exceptional athletic performance. The Sport Psychologist, 1, 181-199.

May, J.R., Veach, T.L., Reed, M.W., and Griffey, M.S. (1985). A psychological study of health, injury and performance in athletes on the US alpine ski team. Physician and Sportsmedicine, 13, 111-115.

Chapter I II III IV V VI VII VIII

99

Meyers, M.C., Bourgeois, A.E., LeUnes, A., and Murray, N.G. (1999). Mood and psychological skills of elite and sub-elite equestrian athletes. Journal of Sport Behavior, 22, 399-409.

Meyers, M.C., LeUnes, A., and Bourgeois, A.E. (1996). Psychological skills assessment and athletic performance in collegiate rodeo athletes. Journal of Sport Behavior, 19, 132-145.

Meyers, M.C., Sterling, J.C., Treadwell, S., Bourgeois, A.E., and LeUnes, A. (1994). Mood and psychological skills of world-ranked female tennis players. Journal of Sport Behavior, 17, 156-165.

Morgan, W.P. (1979). Prediction of performance in athletics. In Coach, Athlete and the Sport Psychologist (edited by P. Klavora and J.V. Daniel), pp. 172-186. Champaign, IL: Human Kinetics.

Morgan, W.P. and Pollock, M.L. (1977). Psychologic characterization of the elite distance runner. Annals of the New York Academy Sciences, 301, 382-403.

Morris, T. (2000). Psychological characteristics and talent identification in soccer. Journal of Sports Sciences, 18, 715-726.

Nideffer, R.M. (1987). Issues in the use of psychological tests in applied settings. The Sport Psychologist, 1, 18-28.

Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. Orlick, T. and Partington, J. (1988). Mental links to excellence. The Sport Psychologist, 2, 105-130. Régnier, G., Salmela, J.H., and Russell, S.J. (1993). Talent detection and development in sport. In A

Handbook of Research on Sports Psychology (edited by R. Singer, M. Murphey, and L.K. Tennant), pp. 290-313. New York: Macmillan.

Reilly, T. (1996). Introduction to science and soccer. In Science and Soccer (edited by T. Reilly), pp. 1-7. London: E & FN Spon.

Reilly, T., Williams, A.M., Nevill, A., and Franks, A. (2000). A multidisciplinary approach to talent identification in soccer. Journal of Sports Sciences, 18, 695-702.

Sewell, D.F. and Edmondson, A.M. (1996). Relation between field position and pre-match competitive state anxiety in soccer and field hockey. International Journal of Sport Psychology, 27, 159-172.

Smith, R.E. and Christensen, D.S. (1995). Psychological skills as predictors of performance and survival in professional baseball. Journal of Sport & Exercise Psychology, 17, 399-415.

Thomas, P.R. and Over, R. (1993). Psychological and psychomotor skills associated with performance in golf. The Sport Psychologist, 8, 73-86.

Weinberg, R.S. and Gould, D. (1999). Foundations of sport and exercise psychology. Champaign: Human Kinetics.

White, S.A. (1993). The relationship between psychological skills, experience, and practice commitment among collegiate male and female skiers. The Sport Psychologist, 7, 49-57.

White, S.A. and Croce, R.V. (1992). Nordic disabled skiers and able-bodied skiers: An exploratory analysis of the psychological skills inventory for sports (PSIS-R5). Clinical Kinesiology, 45, 7-9.

100

Appendix 6.1. Items distributed over the six scales of the PSIS-Youth.

Motivation I am very motivated to do well in my sport.

I sometimes lack the motivation to train.

Winning is very important to me.

Right now, the most important thing in my life is to do well in my sport.

My sport is my whole life.

I want to train hard to belong to the top in my sport.

In my sport, I want to bring out the best in myself.

I want to succeed in my sport

Confidence In most competitions, I go in confident that I will do well.

It doesn’t take much to shake my self-confidence.

A minor injury or a bad practice can really shake my self-confidence.

I have frequent doubts about my athletic ability.

When I begin to perform poorly, my confidence drops very quickly.

I can usually remain confident even through one of my poorer performances.

My self-confidence jumps all over the place.

I have faith in myself.

Anxiety

Control

I am more tense before I perform than I am during the performance.

I am often panic-struck during those last few moments before I begin my

performance.

I spend a lot of energy trying to stay calm before a meet.

I get nervous, because I want to start performing.

I am anxious to perform in strange places.

Before a meet, I worry if I will do well.

Before important meets, I feel intense anxiety.

The period right before a performance feels unpleasant.

Chapter I II III IV V VI VII VIII

101

Mental

Preparation

I often dream about competition.

I often “rehearse” my performance in my head before I perform.

When I mentally practice my performance, I “see” myself performing- just like

I was watching a videotape.

When I am preparing to perform, I try to imagine what it will feel like in my

muscles.

When I close my eyes, I can imagine what my muscles feel like.

I prepare for a meet by making mental representations of my performance.

Team

Emphasis

I get very frustrated when a teammate is performing poorly.

I concentrate more on my own performance than on the performance of the team.

I think team spirit is very important.

When my team loses, I feel badly – no matter how well I did as an individual.

I think the performance of the team is more important than my individual

performance.

If my teammates don’t exert themselves to the utmost, I get angry.

If I decline the performance level of the team, I have to be replaced.

Concentration I often have trouble concentrating during my performance.

I experience frequent “hot streaks” in which my performance is unusually good.

When I am performing poorly, I tend to lose my concentration.

During my performance, I am incommoded by comments of people surrounding 66

me.

At the beginning of my performance, I have trouble forgetting things I was

doing before.

During my performance, others distract me.

I can concentrate better on a difficult meet than on an easy one.

Note: Items were rated on a 5-point scale, using anchors of 1 = almost never and 5 = almost always, while comparing oneself with top players in the same age category.

Chapter VII

Development of the Tactical Skills Inventory for Sports Elferink-Gemser, M.T., Visscher, C., Richart, H.,and Lemmink, K.A.P.M. Perceptual and Motor Skills, 99, 883-895 Acknowledgements:

This study has been supported by a grant of the Dutch National Olympic Committee, NOC*NSF. The authors wish to thank Nynke Schippers for her assistance in collecting the data.

104

Abstract

Purpose of this study, in which 19 trainers and 415 competitive youth field hockey and soccer players (age = 15.9 years, sd = 1.6; 283 boys and 132 girls) selected by their age, sex, and performance status participated, was to develop a practical, reliable, and valid measure of tactical skills in sports. With trainers, 34 questions were formulated involving tactical skills. Factor analysis yielded the Tactical Skills Inventory for Sports. Scales were labeled Positioning and Deciding, Knowing about Ball Actions, Knowing about Others, and Acting in Changing Situations, covering all aspects of tactical skills regarding declarative versus procedural knowledge, and attack and defense. Internal consistency and test-retest measures for reliability (except Knowing about Ball Actions) were within acceptable limits. Elite players scored better than non-elite players, supporting construct validity. The inventory is suitable for measuring tactical skills in youth field hockey and soccer players in sports practice.

Chapter I II III IV V VI VII VIII

105

7.1 Introduction

Elite athletes not only need well-developed physiological and technical characteristics, but certain cognitive characteristics too (French and Thomas, 1987; Starkes, 1987; Williams et al., 1993; Helsen and Starkes, 1999; Nougier and Rossi, 1999). This certainly applies to players of invasive games, in which players compete at the same field of action as their opponents. Invasive games are time dependent and can be subcategorized into goal-throwing (e.g., basketball), try scoring (e.g., rugby), and goal striking games (e.g., soccer). A characteristic of invasive game players is that they constantly need to adapt to opposition by punctual adaptation to new play configurations and to the circulation of the ball (Gréhaigne and Godbout, 1995). In this type of games, players have to deal with a complex and rapidly changing environment while invading the opposing team’s area of the field to score (Almond, 1986; Williams, 2000; Hughes and Bartlett, 2002).

A common way to categorize the cognitive skills needed in sports is the distinction in declarative and procedural knowledge (Anderson, 1982; Thomas and Thomas, 1994; Turner and Martinek, 1999). Both motor skills and tactical skills have elements of declarative knowledge and procedural knowledge (McPherson and Kernodle, 2003). Declarative knowledge includes knowledge of the rules and goals of the game (French and Thomas, 1987; Williams and Davids, 1995), whereas procedural knowledge involves the selection of an appropriate action within the context of the game. In other words, ‘knowing what to do’ refers to declarative knowledge and ‘doing it’ refers to procedural knowledge (McPherson, 1994). Bjurwill (1993) stated that, only if a player has a proper understanding of the game, that is, only when he is very good at ‘reading the game’, can the player be a top player.

So far, many different terms have been used to describe the concept of performing the right action at the right moment. The action and the moment are right when the performance or outcome is successful. For example, Bjurwill (1993) used the terms ‘game intelligence’ and ‘reading the game’. Many other descriptors have been applied, including ‘implicit knowledge’, ‘practical intelligence’, ‘tricks of the trade’, ‘tactical knowledge’, and ‘tactics’ (Davids and Myers, 1990; McPherson, 1994; Gréhaigne and Godbout, 1995; Gréhaigne et al., 1999). At present the term ‘tactical skills’ is utilized (McPherson and Kernodle, 2003). Tactical Skills refer to the quality of an individual player to perform the right action at the right moment; it should therefore be distinguished from strategy, which refers to choices discussed in advance with the trainer in order for the team to organize itself (Gréhaigne and Godbout, 1995).

Most studies of tactical skills applied experimental test situations in which, for example, subjects viewed action sequences on a video projection screen (e.g., Starkes and Deakin, 1984; Williams et al., 1993; Bard et al., 1994; McMorris and Graydon, 1997; Helsen and

106

Starkes, 1999). Others, especially cognitive psychologists, have used propositional-type analyses of subjects’ think-aloud protocols to examine the representation of conceptual knowledge, e.g., declarative, procedural, and to examine how this knowledge guides the solution process during problem-solving or task performance (McPherson, 1994).

Although these settings are useful for fundamental research, they are less suitable for applied purposes. In the field, there is a clear need for information about the tactical skills of individual players, for example, to help trainers guide players toward a higher performance. Information on tactical skills could also prove to be very valuable in leading talented players to the top or in evaluating training effects. Therefore, the goal of this study is to construct an inventory that can be used in sports practice; that is a practical, reliable, and valid measure of tactical skills in sports.

7.2 Methods

To construct the self-reporting inventory, the theoretical elements on tactical skills according to the framework created by McPherson (1994) with one continuum that moves from response selection to response execution and the other continuum that moves from knowledge (knowing what to do) to action (doing it), were discussed with 19 highly qualified trainers of youth national and district selection sports teams in the Netherlands. They were asked to put forward those elements they considered most important for high performance. Elements frequently named as important were overview, anticipation, fast switching from ball possession to no ball possession and vice versa, positioning, man-to-man defense, zone defense, and interception (Elferink-Gemser et al., 2004). These elements are specific to match play in invasive games and concern mostly the combination of picking up relevant information from the environment and reacting to that. Questions were formulated and reformulated until consensus was reached on the content of the inventory within the team of experts. Thirty-four items were put in questionnaire form; these were answered on a 6-point scale regarding sports performance with anchors of 1 = very poor and 6 = excellent or of 1 = almost never and 6 = always, while comparing oneself with top players in the same age category (Table 7.1). Factor analysis was applied in Study 1 to examine the structure of relations among the items in the original sample with the purpose of bringing them together into a smaller set of variables or constructs (Nunnally and Bernstein, 1994). After that, the internal consistency of the inventory was examined in Study 2A and test-retest reliability in Study 2B. Starkes (1987) pointed out the importance of cognitive abilities in the development of skill in field hockey, whereas Williams et al. (1993) concluded that experienced soccer players’ cognitive knowledge permitted more meaningful associations between players’

Chapter I II III IV V VI VII VIII

107

positions resulting in more efficient retrieval. Based on these studies showing that elite players in field hockey and soccer have better cognitive features than lower-performance players, construct validity was examined by comparing scores of players at different playing levels. 7.3 Study 1: Factor Analysis

Method

Participants A total of 209 youth players (age = 15.8 years, sd = 1.6 years, range = 12.6 - 18.9 years), all participating in competitive field hockey (n = 123) or soccer (n = 86), gave their informed consent prior to participation. This population consisted of 148 boys and 61 girls. All players were given the same instructions and were taught in the same way. They filled out the original sample of 34 questions individually.

Data analysis Principal component analysis of the 34 item sample, with four factors fixed, followed by varimax rotation, yielded a structure which accounted for 50% of the response variance. The number of four fixed factors was based on the transition point in the scree plot where successive eigenvalues are plotted against component number (Nunnally and Bernstein, 1994). Items that met the criterion of loading at greater than or equal to 0.55 with a factor were selected to make interpretation of the inventory possible (Kline, 1994; Smith et al., 1995).

Results Twenty-three items met the criterion and are indicated in Table 7.1. Factor 1 consists of Items 1, 2, 4, 5, 6, 7, 8, 9 and 10, and, based on their content, is labeled Positioning and Deciding. Factor 2 consists of Items 16, 17, 18, 19 and 20, and is labeled Knowing about Ball Actions. Factor 3 consists of Items 11, 15, 21, 22 and 23, and is labeled Knowing about Others. Factor 4 has Items 3, 12, 13 and 14, and is labeled Acting in Changing Situations. These four factors make up the four scales in the 23 item Tactical Skills Inventory for Sports.

Tab

le 7

.1.

Orig

inal

34

item

s and

thei

r fac

tor l

oadi

ngs (n

= 20

9).

It

ems

Fa

ctor

#

1 2

3 4

I k

now

whi

ch p

ositi

on I

shou

ld ta

ke d

urin

g m

atch

esx

0.50

0.

10

0.23

0.

26

1 D

ecis

ions

I m

ake

durin

g m

atch

es a

bout

pro

ceed

ing

actio

ns a

re g

ener

ally

* 0.

68

0.09

0.

14

0.05

2 I k

now

how

to g

et o

pen

durin

g a

mat

ch*

0.69

0.

23

-0.0

4 0.

09

3 M

y in

terc

eptio

n of

the

oppo

nent

’s b

all i

s*

0.26

-0

.04

0.25

0.

72

4 M

y po

sitio

ning

dur

ing

a m

atch

is g

ener

ally

* 0.

76

0.10

0.

10

0.12

5 M

y ov

ervi

ew (i

n ba

ll po

sses

sion

or i

n te

am’s

bal

l pos

sess

ion)

is*

0.66

0.

34

0.24

-0

.08

6 M

y an

ticip

atio

n (th

inki

ng a

bout

pro

ceed

ing

actio

ns) i

s*

0.71

0.

14

0.24

0.

10

I k

now

my

stro

ng a

nd w

eak

poin

ts e

xact

lyx

0.22

0.

33

0.19

0.

05

7 I a

m g

ood

at m

akin

g th

e rig

ht d

ecis

ions

at t

he ri

ght m

omen

ts*

0.65

0.

21

0.25

0.

12

8 In

the

opin

ion

of m

y tra

iner

, my

unde

rsta

ndin

g of

the

gam

e is

* 0.

73

0.08

0.

16

0.09

9 M

y ge

tting

ope

n an

d ch

oosi

ng p

ositi

on is

* 0.

64

0.29

0.

07

0.15

10

In th

e op

inio

n of

my

train

er, m

y po

sitio

ning

is*

0.67

0.

20

0.13

0.

04

11

My

judg

men

t of t

he o

ppon

ent’s

pla

y is

*

0.32

0.

10

0.62

0.

15

12

My

inte

rcep

tion

of th

e ba

ll is

* 0.

21

0.05

0.

41

0.68

I a

pply

rule

s of t

he g

ame

smar

tly to

mat

ches

x 0.

29

0.45

0.

14

0.20

D

urin

g m

atch

es I

quic

kly

mak

e de

cisi

onsx

0.36

0.

49

0.14

0.

22

13

If o

ur te

am lo

ses t

he b

all d

urin

g a

mat

ch, I

qui

ckly

switc

h to

my

task

as d

efen

der*

0.

04

0.20

-0

.04

0.80

14

I qui

ckly

reac

t to

chan

ges,

as fr

om n

ot p

osse

ssin

g th

e ba

ll to

bal

l pos

sess

ion*

0.

29

0.46

-0

.12

0.63

D

urin

g m

atch

es, I

look

not

onl

y at

the

ball

but a

lso

look

ove

r the

fiel

dx 0.

33

0.51

0.

07

0.13

15

I kno

w q

uick

ly h

ow th

e op

pone

nt is

pla

ying

* 0.

06

0.18

0.

57

0.16

16

I kno

w e

xact

ly w

hen

to p

ass t

he b

all t

o a

team

mat

e or

whe

n no

t to*

0.

27

0.60

0.

09

0.01

I k

now

qui

ckly

wha

t to

do to

win

a m

atch

x 0.

06

0.46

0.

16

0.18

17

I qui

ckly

ada

pt m

y pl

ay to

circ

umst

ance

s, su

ch a

s rai

ny o

r win

dy w

eath

er*†

0.

01

0.60

0.

11

0.13

I s

ee th

e w

eak

poin

ts o

f the

opp

onen

t qui

ckly

x 0.

06

0.45

0.

43

0.03

I q

uick

ly re

act t

o co

rrec

t mis

take

s of m

y te

amm

ates

x 0.

03

0.43

0.

12

0.44

18

If w

e re

ceiv

e th

e ba

ll (g

ettin

g ba

ll po

sses

sion

), I k

now

exa

ctly

wha

t to

do*

0.21

0.

60

0.12

0.

15

W

hile

rece

ivin

g th

e ba

ll, I

do n

ot h

ave

to lo

ok w

here

my

team

mat

es a

re; I

alre

ady

know

x 0.

33

0.50

0.

47

0.03

19

Whi

le e

xecu

ting

an a

ctio

n in

a m

atch

, I k

now

exa

ctly

wha

t to

do su

bseq

uent

ly*

0.21

0.

63

0.43

0.

07

20

If I

poss

ess t

he b

all,

I kno

w e

xact

ly w

hom

I ha

ve to

pas

s to*

0.

17

0.56

0.

43

-0.0

2

21

Alth

ough

I do

not

see

my

oppo

nent

s, I k

now

whe

re th

ey a

re g

oing

*

0.25

0.

23

0.66

0.

12

If

our

team

lose

s bal

l pos

sess

ion,

I kn

ow e

xact

ly w

hat t

o do

x 0.

07

0.35

0.

37

0.50

If

I re

ceiv

e th

e ba

ll fr

om a

team

mat

e, I

know

in a

dvan

ce w

here

to p

ass t

he b

allx

0.29

0.

47

0.48

0.

06

22

With

out s

eein

g m

y te

amm

ates

, I k

now

whe

re th

ey a

re g

oing

*

0.28

0.

28

0.60

-0

.10

23

If a

n op

pone

nt re

ceiv

es th

e ba

ll, I

know

exa

ctly

wha

t he

is g

oing

to d

o*

0.05

0.

12

0.63

0.

32

Note:

Item

s w

ere

rate

d on

a 6

-poi

nt s

cale

, usi

ng a

ncho

rs o

f 1 =

ver

y po

or a

nd 6

= e

xcel

lent

or o

f 1 =

alm

ost n

ever

and

6 =

alw

ays,

whi

le c

ompa

ring

ones

elf

with

top

play

ers

in th

e sa

me

age

cate

gory

. The

num

bers

indi

cate

the

item

num

ber

in th

e Ta

ctic

al S

kills

Inv

ento

ry f

or S

ports

; unn

umbe

red

item

s w

ere

not

incl

uded

. Fa

ctor

1 =

item

s 1,

2, 4

, 5, 6

, 7, 8

, 9, 1

0: P

ositi

onin

g an

d D

ecid

ing;

Fac

tor 2

= it

ems

16, 1

7#, 1

8, 1

9, 2

0: K

now

ing

abou

t Bal

l Act

ions

; Fac

tor 3

=

item

s 11,

15,

21,

22,

23:

Kno

win

g ab

out O

ther

s; F

acto

r 4 =

item

s 3, 1

2, 1

3, 1

4: A

ctin

g in

Cha

ngin

g Si

tuat

ions

.

x Item

s not

mee

ting

the

crite

ria o

f a >

0.5

5 fa

ctor

load

ing.

*Ite

ms m

eetin

g th

e cr

iteria

of a

> 0

.55

fact

or lo

adin

g.

†Ite

m o

mitt

ed a

fter r

elia

bilit

y st

udie

s.

110

7.4 Study 2: Reliability A – Internal Consistency

Method

Participants A different sample of 206 competitive youth field hockey players (n = 139) and soccer players (n = 67) filled out the Tactical Skills Inventory for Sports (age = 15.9 years, sd = 1.7 years, range = 12.2 - 19.3 years; 135 boys and 71 girls). Again, all players gave their informed consent prior to participation, and procedures were equivalent to those in Study 1.

Data analysis Raw data were screened for missing values. In case of 20% or more missing values within a scale, a participant was excluded from the analysis. Otherwise, a missing value was replaced by the participant’s mean score on the scale involved. Item-total correlations, interitem correlations, Cronbach coefficients alpha for internal consistency, and interscale correlations were used to assess reliability. Concerning item-total correlations, items should correlate more with the scale to which they are assigned than with a different scale. With regard to the interitem correlations, items should correlate positively within their assigned scale. Scales should have a Cronbach coefficient alpha of at least 0.70 (Nunnally, 1978), and interscale correlations should not exceed 0.80 (Carron et al., 1985).

Results None of the participants had 20% missing values or more. Means, standard deviations, and Cronbach coefficients alpha for the inventory are presented in Table 7.2. Table 7.2. Descriptive statistics and internal consistencies (α) of the four subscales of the Tactical

Skills Inventory for Sports (n = 206).

Scale Mean sd α

1 Positioning and Deciding 3.79 0.61 0.89

2 Knowing about Ball Actions 4.11 0.62 0.75

3 Knowing about Others 3.74 0.67 0.74

4 Acting in Changing Situations 4.15 0.69 0.72

Sum of scales 3.95 0.51 0.91

Chapter I II III IV V VI VII VIII

111

Internal consistency estimates for the scales ranged from 0.72 to 0.89. Item-total correlations showed that items had higher correlations with their assigned scale than with any other scale, with the exception of Item 11 (which correlated 0.50 with the assigned Scale 3 and 0.51 with Scale 1), Item 12 (which correlated 0.50 with the assigned Scale 4 and 0.54 with Scale 3) and Item 17 (which correlated 0.31 with the assigned Scale 2 and 0.33 with Scale 3). Interitem correlations within each scale were all positive, ranging from 0.17 to 0.75. The interscale correlations varied from 0.37 between Scales 1 and 4 and 0.59 between Scales 1 and 3 (Table 7.3). Table 7.3. Tactical Skills Inventory for Sports interscale correlations (n = 206).

Scale 2 Scale 3 Scale 4

Scale 1 0.52 0.59 0.37

Scale 2 0.56 0.48

Scale 3 0.54

Scale 4

Note: Scale 1 = Positioning and Deciding; Scale 2 = Knowing about Ball Actions

Scale 3 = Knowing about Others; Scale 4 = Acting in Changing Situations 7.5 Study 2: Reliability B - Test-retest

Method

Participants From the participants of Study 2A, a sample of 47 competitive youth field hockey players filled out the inventory twice (age = 15.6 years, sd = 1.58 years, range = 12.3 - 18.7 years; 18 boys and 29 girls). The second session took place two to four weeks after the first questionnaire completing session, to minimize test-retest effects.

Data analysis Mean scores and standard deviations for the four scales and the sum of scale scores for the first measurement (t1) and second measurement (t2) were calculated. Baumgarter (1989) identified two types of reliability, relative and absolute. Relative reliability is the extent to which individuals maintain their position in a sample with repeated measurements. Absolute reliability is how much repeated measurements vary for individuals. It provides an indication

112

of the variability in repeated tests for specific individuals, irrespective of the individual’s rank in a particular sample (Atkinson and Nevill, 1998; 2001).

The mean difference between the test scores on both days was set as a measure of absolute reliability. If zero lay within the 95% confidence interval of the mean difference, it was concluded that no bias existed between the two measurements. To estimate relative reliability, a one-way analysis of variance was conducted to calculate Intraclass Correlation Coefficients (ICCs) of repeated measures. Ninety-five percent confidence intervals were calculated for all ICC’s (Rankin and Stokes, 1998). An ICC above 0.75 was considered to indicate good stability (Lee et al., 1989; Streiner and Norman, 1995).

Results Zero lay within the 95% confidence interval of the mean difference for Scales 1, 3, and 4 and the sum of scales. Scales 1, 3 and 4 and the sum of scales had an ICC varying between 0.76 and 0.89. Only Scale 2 did not meet the criterion, with an ICC of 0.53 (Table 7.4). Table 7.4. Measures for absolute and relative reliability of the Tactical Skills Inventory for Sports

(n = 47).

t1

(sd)

t2

(sd)

t1 – t

(sd)

SE of

t1 – t2

95% CI

for t1 – t2

ICC 95% CI

for ICC

Scale 1 3.3 (0.6) 3.4 (0.5) -0.06 (0.35) 0.05 -0.17 – 0.04 0.88 0.78 – 0.93

Scale 2 3.7 (0.6) 3.4 (0.4) 0.30 (0.60) 0.09 0.13 – 0.48 0.53 0.16 – 0.74

Scale 3 3.3 (0.7) 3.3 (0.6) 0.00 (0.59) 0.09 -0.17 – 0.17 0.76 0.57 – 0.87

Scale 4 3.8 (0.7) 3.7 (0.7) 0.09 (0.54) 0.08 -0.07 – 0.25 0.82 0.67 – 0.90

Sum of scales 3.5 (0.5) 3.5 (0.4) 0.08 (0.31) 0.05 -0.00 – 0.17 0.89 0.80 – 0.94

Note: t1 – t2 = mean difference between scores from testing times 1 and 2; SE of t1 – t2 = Standard Error of the mean difference; 95% CI for t1 – t2 = 95% Confidence Interval for the mean difference; ICC = Intraclass Correlation Coefficient; 95% CI for ICC = 95% Confidence Interval for each Intraclass Correlation Coefficient.

7.6 Study 3: Construct validity

Elite and non-elite youth players were compared on the basis of their scores on the Tactical Skills Inventory for Sports. It was hypothesized that the elite youth group would have higher mean tactical skills scores than the non-elite youth group. Youth players participating in the highest national leagues for their age were considered elite youth players, whereas youth

Chapter I II III IV V VI VII VIII

113

players at a moderate performance status, i.e., played in a regional competition, were considered non-elite youth players. Method

Participants A total of 148 youth field hockey players filled out the inventory. Among them were 76 elite youth field hockey players (age = 15.7 years, sd = 1.7 years, range = 12.8 - 18.4 years; 34 boys and 42 girls) from Study 1 and 72 non-elite youth field hockey players (age = 15.3 years, sd = 1.7 years, range = 12.3 – 18.7 years; 28 boys and 44 girls) from Study 2. Again, all players gave their informed consent prior to participation, and procedures were equivalent to those in Study 1 and Study 2.

Data analysis Mean scores and standard deviations were calculated for each scale and the sum of scales. The scores of the elite players were then compared with those of the non-elite players using an analysis of variance.

Results The lowest mean scores were obtained for Scale 3, Knowing about Others; the highest mean scores for Scale 4, Acting in Changing Situations. The mean Scale 3 score of the elite youth field hockey players was 3.8, and their mean scale score was 4.3 for Scale 4, whereas non-elite youth players showed means of 3.4 for Scale 3 and 3.8 for Scale 4 (Table 7.5). On all scales, elite youth players scored higher than non-elite youth players (p < 0.01). Table 7.5. Scale score statistics for groups playing at different skill levels (n = 148).

Scale

Elite players

(n = 76)

Non-elite players

(n = 72)

1 Positioning and Deciding 3.97 (0.56) 3.43 (0.61)

2 Knowing about Ball Actions 4.22 (0.57) 3.77 (0.68)

3 Knowing about Others 3.77 (0.60) 3.41 (0.72)

4 Acting in Changing Situations 4.25 (0.65) 3.82 (0.69)

Sum of scales 4.05 (0.44) 3.61 (0.55)

Note: Elite and non-elite player groups’ mean scores differed on all scales and the sum of scales (p < 0.01).

114

7.7 Discussion

The goal of this study was to construct a practical, reliable, and valid measure of tactical skills in invasive game players. The content of the inventory was selected with the help of a team of expert trainers. Factor analysis yielded four scales which were labeled Positioning and Deciding, Knowing about Ball Actions, Knowing about Others, and Acting in Changing Situations.

Two factors (2 and 3) contain questions more related to declarative knowledge. In these factors, Knowing about Ball Actions and Knowing about Others, knowledge of the game is the central element. The other two factors (1 and 4) contain questions more related to procedural knowledge. In these factors, Positioning and Deciding, and Acting in Changing Situations, selection of the appropriate action is the central element. A way to categorize elements of tactical skills related to the nature of match play in invasive games is by making a distinction between on-the-ball and off-the-ball situations (Oslin et al., 1998). Tactics related to scoring or attack can be distinguished from tactics related to preventing scoring or defense (Bjurwill, 1993). According to Mitchell (1996), tactical skills such as maintaining possession of the ball, attacking the goal, and creating space in the attack are similar across invasive games, as are defending space or defending against an attack. Among the four factors, Factors 1 and 2 are more related to the attack, whereas the other two factors (3 and 4) are more related to defense. Questions for Positioning and Deciding and for Knowing about Ball Actions mostly concern situations in which the team possesses the ball. Questions in Knowing about Others and Acting in Changing Situations, on the other hand, mostly concern situations in which the opposing team possesses the ball. By combining both ways of categorizing elements of tactical skills, i.e., declarative versus procedural knowledge and attack versus defense, the four factors in the inventory cover all four of these aspects of tactical skills.

Cronbach coefficients alpha of all four scales were above the criterion value of 0.70, indicating good internal consistency (Nunnally, 1978). In addition, item-total correlations supported the categorization, although three items correlated better with a scale different than their assigned one. However, the small difference between the correlations and the other satisfying psychometric results were the basis for not altering the inventory derived from Study 1. Interscale correlations were moderate, varying from 0.37 to 0.59. This is in line with the assumption that the scales are all part of the same construct. The correlations did not have such high values (< 0.80) that one scale should replace two of them (Carron et al., 1985).

Except for Scale 2, Knowing about Ball Actions, values of test-retest reliability led to the conclusion that the scales, as well as the sum of scales, met the criteria for absolute and relative reliability. It was remarkable that the average scores on Scale 2 were lower on t2 than on t1, whereas no such decrease was found on the other three scales. When examining the

Chapter I II III IV V VI VII VIII

115

items of Scale 2, we detected that Item 17 (‘I quickly adapt my play to circumstances, such as rainy or windy weather’) had a very low ICC compared to the other items (ICC = 0.03). An explanation could be that, between measurements, some players actually had to play a match in rainy or windy weather and found that they were better or worse in adapting to those circumstances than they formerly thought. Reliability coefficients of Scale 2 increase when Item 17 is omitted (ICC = 0.60 instead of 0.53). Besides, the content of this item does not fit well in the scale. Based on these findings, in combination with the results from Study 2A that this item correlated higher with Scale 3 than with its assigned Scale 2, Item 17 should be omitted from the Tactical Skills Inventory for Sports.

Study 3 showed that elite field hockey players scored significantly better on all scales and on the sum of scales than non-elite field hockey players. The above-mentioned findings support the construct validity of the questionnaire. The results are in line with those of other studies showing that skilled players outscore less skilled ones on tactical skills elements (Williams et al., 1993; Williams and Davids, 1995; Enns and Richards, 1997).

Whether the inventory is measuring the whole concept of tactical skills cannot completely be ascertained without an accepted reference criterion (inventory). However, this inventory was constructed with help of expert trainers and embedded in theory. This method of gathering items can be considered logical validity, also referred to as face validity, and supports the notion that the inventory is really measuring tactical skills (Thomas and Nelson, 1996). Nevertheless, the results may be influenced by the limitations of the inventory, requiring self-report. Self-reported measures are susceptible to the individual’s self-confidence, and, since confidence is associated with elite performance in various sports, this might have affected the results (Mahoney et al., 1987; Woodman and Hardy, 2003). Therefore, one could argue that the results of Study 3 for construct validity may have been influenced by enhanced self-confidence of elite players. However, an alternative hypothesis might also be true. The elite players have on average over eight years of active field hockey experience, and they are all part of a talent development program of a field hockey club of national prestige. This means that they have been confronted frequently with all aspects of their performance on the field. Trainers, coaches, peers, and parents give feedback on how fast they are, how well they dribble the ball, and also whether they perform the right action at the right moment. When players are confronted by (significant) others with their tactical skills for many years in a row, they ultimately know how good (or bad) they really are. In other words, regardless of their enhanced confidence, elite players are thought to have a realistic perspective on their tactical skills. It will be interesting to test this hypothesis.

Caution should be taken in generalizing the results to other populations. This sample consisted of competitive youth field hockey and soccer players from the Netherlands.

116

Therefore Dutch is the original language in which the Tactical Skills Inventory for Sports was constructed. So far, the English version of the inventory has not yet been applied and it can not be assumed straightforwardly that the same results will be obtained. Based on performance indicators, formal games can be classified in three categories: net and wall games, invasive games, and striking and fielding games (Read and Edwards, 1992). Field hockey and soccer are invasive games which fall into the subcategory goal striking games (Hughes and Barlett, 2002). Research could be directed to populations of competitive sports athletes in other categories of formal games and in other countries. Moreover it would be valuable to study the tactical skills from the inventory with other scales than the self-reported inventory.

In conclusion, the internal consistency, test-retest reliability, and construct validity of the Tactical Skills Inventory for Sports were acceptable. With the Tactical Skills Inventory for Sports, which can be used in sports practice, information can be gathered on ‘positioning and deciding’, ‘knowing about ball actions’, ‘knowing about others’, and ‘acting in changing situations’.

Chapter I II III IV V VI VII VIII

117

References

Almond, L. (1986). Reflecting on theses: a games classification. In Rethinking games teaching (edited by R. Thorpe, D. Bunker, and L. Almond), pp. 71-72. Loughborough, England: Loughborough University of Technology.

Anderson, J.R. (1982). Acquisition of cognitive skill. Psychological Review, 89, 369-406. Atkinson, G. and Nevill, A.M. (1998). Statistical methods for assessing measurement error (reliability)

in variables relevant to sports medicine. Sports Medicine, 26, 217-238. Atkinson, G. and Nevill, A.M. (2001). Selected issues in the design and analysis of sport performance

research. Journal of Sports Sciences, 19, 811-827. Bard, C., Fleury, M., and Goulet, C. (1994). Relationship between perceptual strategies and response

adequacy in sport situations. International Journal of Sport Psychology, 25, 266-281. Baumgarter, T.A. (1989). Norm-referenced measurement reliability. In Measurement concepts in

physical education and exercise science (edited by M.J. Safrit and T.M. Wood), pp. 45-72. Champaign, IL: Human Kinetics.

Bjurwill, C. (1993). Read and react: the football formula. Perceptual and Motor Skills, 76, 1383 -1386. Carron, A.V., Widmeyer, W.N., and Brawley, L.R. (1985). The development of an instrument to

assess cohesion in sport teams: The group environment questionnaire. Journal of Sport Psychology, 7, 244-266.

Davids, K. and Myers, C. (1990). The role of tacit knowledge in human motor performance. Journal of Human Movement Studies, 19, 273-288.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., and Mulder, Th. (2004). The relation between multidimensional performance characteristics and level of performance in talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

Enns, J. and Richards, J. (1997). Visual attentional orienting in developing hockey players. Journal of Experimental Child Psychology, 64, 255-275.

French, K.E. and Thomas, J.R. (1987). The relation of knowledge development to children’s basketball performance. Journal of Sport Psychology, 9, 15-32.

Gréhaigne, J.F. and Godbout, P. (1995). Tactical knowledge in team sports from a constructivist and cognitivist perspective. Quest, 47, 490-505.

Gréhaigne, J.F., Godbout, P., and Bouthier, D. (1999). The foundations of tactics and strategy in team sports. Journal of Teaching in Physical Education, 18, 159-174.

Helsen, W.F. and Starkes, J.L. (1999). A multidimensional approach to skilled perception and performance in sport. Applied Cognitive Psychology, 13, 1-27.

Hughes, M.D. and Bartlett, M. (2002). The use of performance indicators in performance analysis. Journal of Sports Sciences, 20, 739-754.

Kline, P. (1994). An easy guide to factor analysis. London: Routledge. Lee, J., Koh, D., and Ong, C.N. (1989). Statistical evaluation of agreement between two methods for

measuring quantitative variables. Computers in Biology and Medicine, 19, 61-70. Mahoney, M.J., Gabriel, T.J., and Perkins, T.S. (1987). Psychological skills and exceptional athletic

performance. The Sport Psychologist, 1, 181-199. McMorris, T. and Graydon, J. (1997). The effect of exercise on cognitive performance in soccer

specific tests. Journal of Sports Sciences, 15, 459-468. McPherson, S.L. (1994). The development of sport expertise: mapping the tactical domain. Quest, 46,

223-240. McPherson, S.L. and Kernodle, M.W. (2003). Tactics, the neglected attribute of expertise. Problem

representations and performance skills in tennis. In Expert performance in sports. Advances in Research on Sport Expertise (edited by J.L. Starkes and K.A. Ericsson), pp. 137-167. Champaign, IL: Human Kinetics.

Mitchell, S.A. (1996). Tactical approaches to teaching games: improving invasion game performance. Journal of Physical Education, Recreation and Dance, 67, 30-33.

Nougier, V. and Rossi, B. (1999). The development of expertise in the orienting of attention. International Journal of Sport Psychology, 30, 246-260.

118

Nunnally, J.C. (1978). Psychometric theory. New York: McGraw-Hill. Nunnally, J.C. and Bernstein, I.H. (1994). Psychometric theory. New York: McGraw-Hill. Oslin, J.L., Mitchell, S.A., and Griffin, L.L. (1998). The game performance assessment instrument

(GPAI): development and preliminary validation. Journal of Teaching in Physical Education, 17, 231-243.

Rankin, G. and Stokes, M. (1998). Reliability of assessment tools in rehabilitation: an illustration of appropriate statistical analysis. Clinical Rehabilitation, 12, 87-199.

Read, B. and Edwards, P. (1992). Teaching children to play games. Leeds: White Line Publishing. Smith, R.E., Schutz, R.W., Smoll, F.L., and Ptacek, J.T. (1995). Development and validation of a

multidimensional measure of sport-specific psychological skills: the athletic coping skills inventory-28. Journal of Sport & Exercise Psychology, 17, 370-398.

Starkes, J.L. (1987). Skill in field hockey: the nature of the cognitive advantage. Journal of Sport Psychology, 9, 146-160.

Starkes, J.L. and Deakin, J. (1984). Perception in sport: a cognitive approach to skilled performance. In Cognitive sport psychology (edited by W.F. Straub and J.M. Williams), pp. 115-128. Lansing, NY: Sport Science Associates.

Streiner, D.L. and Norman, G.R. (1995). Health measurement scales: a practical guide to their development and use. Oxford: Oxford University Press.

Thomas, J.R. and Nelson, J.K. (1996). Research methods in physical activity. Champaign, IL: Human Kinetics.

Thomas, K.T. and Thomas, J.R. (1994). Developing expertise in sport: the relation of knowledge and performance. International Journal of Sport Psychology, 25, 295-312.

Turner, P. and Martinek, J. (1999). An investigation into teaching games for understanding: effects on skill, knowledge, and game play. Research Quarterly for Exercise and Sport, 70, 286-296.

Williams, A. M. (2000). Perceptual skill in soccer: implications for talent identification and development. Journal of Sports Sciences, 18, 737-750.

Williams, M. and Davids, K. (1995). Declarative knowledge in sport: a by-product of experience or a characteristic of expertise. Journal of Sport and Exercise Psychology, 17, 259-275.

Williams, A.M., Davids, K., Burwitz, L., and Williams, J.G. (1993). Cognitive knowledge and soccer performance. Perceptual and Motor Skills, 76, 579-593.

Woodman, T. and Hardy, L. (2003). The relative impact of cognitive anxiety and self-confidence upon sport performance: a meta-analysis. Journal of Sports Sciences, 21, 443-457.

Chapter I II III IV V VI VII VIII

119

Chapter VIII

Discussion and Conclusions

122

Chapter I II III IV V VI VII VIII

123

8.1 Development of a field hockey specific test battery

In chapters 2 and 3, attention is paid to how to measure the multidimensional performance characteristics important for high-performance in youth field hockey players. With a field hockey test battery the multidimensional performance characteristics can be measured practically, reliably, and with validity in talented youth field hockey players in a sports-specific way. This test battery consists of measurement of height, body mass, and percentage body fat, the Shuttle Sprint and Dribble Test (ShuttleSDT), the Slalom Sprint and Dribble Test (SlalomSDT), the Interval Shuttle Run Test (ISRT), the ‘Tactics in Sports’ questionnaire, and the Dutch Youth Version of the ‘Psychological Skills Inventory for Sports’ (PSIS-Youth). To avoid depending exclusively on the opinion of the trainer to measure tactical skills, it is recommended to apply the ‘Tactical Skills Inventory for Sports’ in stead of the ‘Tactics in Sports’ questionnaire in the future (see chapter 7). 8.2 Studies on the relation between multidimensional performance characteristics

and performance level

In chapter 3, a study conducted within a group of all talented youth field hockey players is presented. The research question to be addressed was: which of the multidimensional performance characteristics; anthropometric, physiological, technical, tactical and / or psychological, makes it possible to discriminate between elite and sub-elite youth field hockey players? Results show that at the age of about fourteen years, an elite player as well as a sub-elite player has a high level of physiological characteristics, i.e. sprints fast over short distances, can perform these sprints repeatedly, is agile while sprinting and has a great interval endurance capacity. An elite player, however, distinguishes him/herself from a sub-elite player not on these physiological characteristics or on anthropometric characteristics but on excellent technical, tactical and psychological skills. Tactical skill, i.e., performing the right action at the right moment seems the most discriminating variable, followed by motivation. Although sub-elite players score high on motivation, elite players score even higher. This motivation seems essential for both their current and future performance.

In chapters 4 and 5, the talented players from the study presented in chapter 3, were followed over time by applying a longitudinal study design. The research question was: how do elite and sub-elite youth field hockey players develop their multidimensional performance characteristics across time? Results show that during the phase of talent development, players improve on all anthropometric, physiological, and technical performance characteristics. Except for the anthropometric characteristics where the development of elite and sub-elite

124

players is similar, elite players improve more rapidly than sub-elite players. With a longitudinal model for interval endurance capacity, scores on the Interval Shuttle Run Test can be predicted for elite and sub-elite boys and girls in field hockey in the age-band of 12-19 years. During adolescence both male and female elite youth players have a more promising development pattern of their interval endurance capacity than sub-elite youth players. A lower percentage body fat, more hours of additional training, and a greater motivation account for a more desirable development of the interval endurance capacity. Questionnaires were used as the measuring instrument for tactical and psychological skills in absence of other valid, reliable, and practible instruments to measure these skills. To provide an external reference point, each player was to be compared to top players in their age-category. As a result of the nature of measuring instrument, however, it was difficult to draw conclusions about the development of those skills.

The studies on the relation between multidimensional performance characteristics and level of performance yield a hierarchy in the multidimensional performance characteristics important for success in field hockey. Tactics, motivation, and slalom dribble performance are the most important performance characteristics in distinguishing between elite and sub-elite youth players at the age of about 14 years. This hierarchy is in agreement with the results of a recent study on talent identification and development of talented water-polo players (Falk et al., 2004). Elite water-polo players at the age of 14-15 years were superior on most of the swim tasks, as well as on dribbling and game intelligence. This superiority was maintained throughout 2 years. The longitudinal data in chapter 4 show that at the age of 15 years and, again, at the age of 16 years, the performance characteristics from the hierarchy are still important. However, not only is the gap in test scores between elite and sub-elite players greater by that time, other performance characteristics as the interval endurance capacity additionally play a more prominent role in distinguishing between the two performance groups. Towards excellence, players have to meet increasingly high standards of achievement. In field hockey, as in most sports, there is a clear necessity for youth players to improve their performance level across a limited period of time. Different from actualizing for example musical or intellectual talents into excellence, players only have a relatively short period of time to perform at the highest level before the aging process causes a decline in their performance (Rowley, 1995).

The hierarchy in the multidimensional performance characteristics important for success in field hockey might change with time according to the evolution of the sport. Over historical time the absolute performance level of sports substantially increases. In some sports, the world records have improved by around 50% in the last century (Schulz and Curnow, 1988). The public, media, business and industry attach great importance to world-class performances

Chapter I II III IV V VI VII VIII

125

and this leads to growing sophistication of training, equipment, and facilities. In the last decade in field hockey for example, new stick material, artificial playing surface (water-based pitches) and the interchange rule made the game faster, increasing and changing the demands of the players. This development of the sport may also change the relationship between multidimensional performance characteristics and performance level.

8.3 Additional studies on talented athletes

The studies on the relation between multidimensional performance characteristics and performance level in talented youth field hockey players show that psychological characteristics distinguish elite from sub-elite talented youth field hockey players. One of the most discriminating variables between both performance groups is motivation. Motivation is not only essential for an optimal performance at a certain moment, i.e. in a match, training, or test, but throughout the long process of developing a successful sports career. To investigate whether this finding is specific for field hockey or can be generalized to other sports, a study to reveal the relationship between psychological skills and performance level within talented youth athletes in field hockey, basketball, volleyball, speed skating, and swimming was presented in chapter 6. Results show that, in general, psychological profiles differed between team sport athletes and individual sport athletes. Psychological characteristics seem more related to performance level in female than in male athletes. However, for discrimination of elite and sub-elite youth athletes, motivation and mental preparation were useful indicators that are independent of gender and type of sport.

In chapter 7, attention was paid to measuring tactical skills. The studies described in chapters 3 and 4 show that the most discriminating variable between elite youth field hockey players and sub-elite youth field hockey players is tactical skill. Future elite players seem to excel in tactical skills by the age of 14 already. However, in these studies tactical skills were measured by the opinion of the trainers. Although these trainers are experts in the field and their opinion is highly valued, one might argue that their judgment of a player’s tactical skills is influenced by their knowledge of that player’s performance level. Therefore, we conducted a study with the purpose of developing a practical, reliable, and valid self-report instrument to measure tactical skills in sports. Results show that the Tactical Skills Inventory for Sports is suitable for measuring tactical skills in youth field hockey players in sports practice.

126

8.4 Theoretical considerations

The present thesis contributes to a clearer understanding of the relation between (the development of) multidimensional performance characteristics and the performance level in talented youth field hockey players, and is a relevant step in unraveling the mechanisms of how one achieves greatness in sports. However, the definition of talent used in the present thesis is still vague: what exactly does it mean when a player is ‘better than peers during training and competition’ and how can we measure ‘the potential to become an elite performer in the future’? We used the accumulated know-how of field hockey experts to select the participants for this study. Players were considered talented in cases where they were part of a talent development program of a field hockey club of national prestige, and were playing at the highest level for their age category. The distinction between elite and sub-elite youth players was made on the basis of players additionally being part of a youth selection team of the Dutch Field Hockey Association (KNHB) or not. Consequently, the present thesis only gives insight into the process of talent development of already identified talented players; but what if not the most talented youth players were detected and identified? Field hockey, as all team sports, is characterized by a lack of objective performance measurements, making it hard to decide which player is the superior one. Although several trainers and coaches claim to be able to ‘recognize a talented player when they see one’ it would be valuable to measure more objectively what exactly it is that they ‘see’.

In the present thesis elite youth players were compared to sub-elite youth players revealing information on the performance characteristics that are important for current and future success in field hockey. Despite that, this thesis does not specify exactly the underlying processes that enable players to, for example, perform the right action at the right moment and future research on this topic is highly recommended. In addition, with the exception of the interval endurance capacity, no mechanisms underlying the development of multidimensional performance characteristics have been studied yet.

8.5 Nature-nurture controversy

Although not included in this present thesis, the environment of talented players must not be underestimated. When a talented youth player attempts to develop his or her talent to reach elite status, this has major consequences for lifestyle. The process is long, averaging at least 10 to 12 years, and during this interval, significant others, particularly parents and coaches, play an important role (Côté, 1999; Visscher et al., 2004). Bloom (1985) also stressed the role of the environment by indicating that the development of exceptional talents requires family

Chapter I II III IV V VI VII VIII

127

support, excellent teaching, and appropriate motivational reinforcement at any stage of their development. Because the present thesis focuses on individual performance characteristics and not on the environment, it is hard to draw conclusions on the real determinants of expert performance. Nevertheless, the results show that for the attainment of expert performance the undertaking of extensive amounts of practice is essential, which is in line with other research (Janelle and Hillman, 2003). Experts simply do not become experts without an enormous investment in training. The stars of tomorrow are the talented players of today and they have a long way to go to the top.

Howe et al. (1998) suggest that differences in early experiences, preferences, opportunities, habits, training, and practice are the real determinants of excellence. The deliberate practice theory of expert performance also takes the perspective that it is practice and experience rather than innate talent that is the real determinant of expert performance (Ericsson, 1998; 2003a; 2003b). This perspective is one that gains support from those advocating environmental determinants of exceptional performance (e.g., Sloboda et al., 1994a, 1994b; Howe, 2001) but it is at odds with the perspectives advanced by behavioral geneticists (e.g., Rowe, 1998).

Abernethy and colleagues (2003) critique the deliberate practice framework. They argue that genetic factors play a critical role in determining the limits to the impact of training and therefore the ultimate levels to performance in many sports (Singer and Janelle, 1999; Skinner, 2001). In support of this hypothesis, a longitudinal study on growth and development of young gymnasts, swimmers, soccer and tennis players, showed that continued success in sport of young athletes appeared to be related to inherited traits (Baxter-Jones et al., 1995). Further research with the field hockey players in the present thesis might give advancing insight in the determinants of excellence.

8.6 Recommendations for future research

Although this present thesis revealed performance characteristics that can distinguish elite youth field hockey players from sub-elite youth field hockey players, it is still unclear to what extent and how these performance characteristics can be trained. If more insight can be given into this question, sports in general and field hockey in particular can benefit enormously. Without detracting from the current results, this study will greatly increase in value when the talented field hockey players are followed until adulthood and some of them actually reach expert status. This study is a first step in bridging the gap between science and sports practice. The next important step to take is to specify the underlying processes of the multidimensional performance characteristics, to measure these in training and competition, and to evaluate

128

current training programs. Furthermore, when necessary, to develop, implement, and evaluate new training programs with the goal of improving the multidimensional performance characteristics of talented field hockey players. Examples are field hockey-specific programs for tactical and perceptual training. 8.7 Conclusions and implications for field hockey

The aim of the thesis was addressed by conducting research within a group of all talented field hockey players, measuring multidimensional performance characteristics in a sports-specific way, and following talented players across time by adopting a longitudinal study design. With caution because the talented players from this study have not yet reached expert performance in adulthood, and with acknowledging the limitations of this study, it is concluded that a talented field hockey player with the greatest chance of succeeding is a player with a relatively high level of performance in field hockey specific physiological characteristics, excellent technical skills, excellent tactical skills, and a very high motivation at the age of fourteen already. This, however, is not enough. A player also has to have potential to reach elite status in the future. Elite players need less time to develop better performance characteristics, meaning that a talented player has to increase his or her performance characteristics at a relatively fast pace for many years in a row. To sustain the long road to the top, investing enormous amounts of time preparing for the international sporting arena, again motivation is essential. From this thesis, relevant information for trainers, coaches, scouts, players, parents and other field hockey enthusiastics can be given:

- Acknowledge the multidimensional nature of a field hockey performance: a talented player is more than a technically gifted player.

- Motivation plays an essential role in the development of a successful career in field hockey.

- Technical and especially tactical skills have to be excellent if a player is to succeed. - In addition, a talented player needs a relatively high level of field hockey specific

physiological characteristics: sprinting fast over short distances, perform these sprints repeatedly, is agile while sprinting, and has a great interval endurance capacity.

For each talented player, it is recommended to construct a performance profile repeatedly, i.e. every year, during the entire process of talent development. In this way, the player’s level of performance characteristics can be compared to other talented peers. Even more, his or her development of the performance characteristics across time can be recorded. This

Chapter I II III IV V VI VII VIII

129

performance profile can be constructed by measuring the multidimensional performance characteristics with the field hockey test battery from this study. In Vakblad Hockey, an electronic journal of the Dutch Field Hockey Association, reference data of the multidimensional performance characteristics of talented boys and girls under 14 years, under 16 years, and under 18 years are published (Elferink-Gemser et al., 2004a; 2004b). By comparing the test results of each player with these reference data, it is possible to acknowledge strong and weaker performance characteristics, and, consequently, use this information in training.

130

References

Abernethy, B., Farrow, D., and Berry, J. (2003). Constraints and issues in the development of a general theory of expert perceptual-motor performance. A critique of the deliberate practice framework. In Expert performance in sports: Advances in research on sport expertise (edited by J. L. Starkes and K. A. Ericsson), pp. 349-369. Champaign, IL: Human Kinetics.

Baxter-Jones, A.D.G., Helms, P., Muffulli, N., Baines-Preece, J., and Preece, M. (1995). Growth and development of male gymnasts, swimmers, soccer and tennis players: a longitudinal study. Annals of Human Biology, 22, 381-395.

Bloom, B.S. (1985). Developing talent in young people. New York: Ballantine. Côté, J. (1999). The influence of the family in the development of talent in sport. The Sport

Psychologist, 13, 395-417. Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. (2004a). Prestatieprofielen van jeugdig

getalenteerde hockeyers: deel 1. [Performance profiles of talented youth field hockey players: part 1]. Vakblad Hockey.

Elferink-Gemser, M.T., Visscher, C., and Lemmink, K.A.P.M. (2004b). Prestatieprofielen van jeugdig getalenteerde hockeyers: deel 2. [Performance profiles of talented youth field hockey players: part 2]. Vakblad Hockey.

Ericsson, K.A. (1998). The scientific study of expert levels of performance: General implications for optimal learning and creativity. High Ability Studies, 9, 75-100.

Ericsson, K.A. (2003a). Development of elite performance and deliberate practice. An update from the perspective of the expert performance approach. In Expert performance in sports: Advances in research on sport expertise (edited by J.L. Starkes and K.A. Ericsson), pp. 49-87. Champaign, IL: Human Kinetics.

Ericsson, K.A. (2003b). How the expert performance approach differs from traditional approaches to expertise in sport. In Expert performance in sports: Advances in research on sport expertise (edited by J.L. Starkes and K.A. Ericsson), pp. 370-402. Champaign, IL: Human Kinetics.

Falk, B., Lidor, R., Lander, Y., and Lang, B. (2004). Talent identification and early development of elite water-polo players: a 2-year follow-up study. Journal of Sports Sciences, 22, 347-355.

Howe, M.J.A. (2001). Genius explained. Cambridge, UK: Cambridge University Press. Howe, M.J.A., Davidson, J.W., and Sloboda, J.A. (1998). Innate talents: Reality or myth? Behavioral

and Brain Sciences, 21, 399-442. Janelle, C.M. and Hillman, C.H. (2003). Expert performance in sport: Current perspectives and current

issues. In Expert performance in sports: Advances in research on sport expertise (edited by J.L. Starkes and K.A. Ericsson), pp. 19-47. Champaign, IL: Human Kinetics.

Rowe, D.C. (1998). Talent scouts, not practice scouts: Talents are real. Behavioral and Brain Sciences, 21, 421-422.

Rowley, S. (1995). Identification and development of talent in young athletes. In Actualizing talent: A lifelong challenge (edited by J. Freeman, P. Span, and H. Wagner), pp. 128-143. London: Cassell.

Schulz, R. and Curnow, C. (1988). Peak performance and age among super athletes: Track and field, swimming, baseball, tennis, and golf. Journal of Gerontology: Psychological Sciences, 43, 113-120.

Singer, R.N. and Janelle, C.M. (1999). Determining sport expertise: From genes to supremes. International Journal of Sport Psychology, 30, 117-150.

Skinner, J.S. (2001). Do genes determine champions? Sports Science Exchange, 14, 83-90. Sloboda, J.A., Davidson, J.W., and Howe, M.J.A. (1994a). Is everyone musical? The Psychologist, 7,

349-354. Sloboda, J.A., Davidson, J.W., and Howe, M.J.A. (1994b). Musicians: Experts not geniuses. The

Psychologist, 7, 363-364. Visscher, C., Elferink-Gemser, M.T., and Lemmink, K.A.P.M. (2004). The role of parental support in

sports success of talented young Dutch athletes. In Children and Youth in Organized Sports (edited by M. Coelho e Silva and R.M. Silva), pp. 123-135. Coimbra: Coimbra University Press.

Chapter I II III IV V VI VII VIII

131

Summary

134

On which performance characteristics distinguish elite youth field hockey players themselves from sub-elite youth field hockey players? The goal of this thesis is to gain an understanding of the relation between (the development of) multidimensional performance characteristics and the performance level in talented youth field hockey players. This goal was addressed by conducting research within a group of all talented field hockey players, measuring multidimensional performance characteristics in a sports-specific way, and following talented players across time by adopting a longitudinal study design.

In order to measure performance characteristics, tests had to be developed. In chapter 2, the reliability of two field hockey specific sprint and dribble tests was evaluated: the Shuttle Sprint and Dribble Test (ShuttleSDT) and the Slalom Sprint and Dribble Test (SlalomSDT). The shuttle sprint and dribble performances of 22 young male and 12 young female field hockey players were assessed on two occasions within 4 weeks. Twenty one young female field hockey players took part in the slalom sprint and dribble test twice in a 4 week period. The ShuttleSDT requires the players to perform three 30 m shuttle sprints while carrying a hockey stick alternated with shorts periods of rest and, after a 5 minute rest, three 30 m shuttle sprints alternated with rest while dribbling a hockey ball. The SlalomSDT requires the players to run a slalom course and, after a 5 minute rest, to dribble the same slalom with a hockey ball. It was concluded that the ShuttleSDT and the SlalomSDT are reliable measures of sprint and dribble performances of young field hockey players.

To determine the relation between multidimensional performance characteristics and level of performance in talented youth field hockey players, in chapter 3 elite youth players (n = 38, mean age 13.21, sd = 1.26) were compared with sub-elite youth players (n = 88, mean age 14.17, sd = 1.26) on anthropometric, physiological, technical, tactical and psychological characteristics. Multivariate analyses with factors of performance level and gender, and with age as a covariate, showed that the elite youth players scored better than the sub-elite youth players on technical (dribble performance in a peak and repeated shuttle run), tactical (general tactics; tactics for possession and non-possession of the ball) and psychological variables (motivation) (p < 0.05). The most discriminating variables were tactics for possession of the ball, motivation and performance in a slalom dribble. Thereby, age discriminated between both performance groups, indicating that the elite youth players were younger than the sub-elite players

To reveal performance characteristics, which may have power for predicting future elite field hockey players, in chapter 4 we made a comparison between 30 elite and 35 sub-elite youth players in terms of anthropometric, physiological, technical, tactical and psychological characteristics measured on three occasions, each separated by a time interval of one year. Mean age of the players on the first measurement was 14.2 years (sd = 1.1). Repeated

135

measures analyses of covariance with factors of performance level and measurement, and with age as a covariate, showed that the elite players scored better than the sub-elite players on technical and tactical variables. Female elite youth players also scored better on interval endurance capacity, motivation and confidence. Future elite players seem to excel in tactical skills by the age of 14 already. They also stand out in specific technical skills and develop these together with the interval endurance capacity better than sub-elite youth players in the two subsequent years.

Chapter 5 describes a study with the goal to better understand the mechanisms that underlie the development of the interval endurance capacity in talented youth field hockey players in the 12-19 age band. In a period of three years 393 measurements were taken and longitudinal changes in interval endurance capacity were investigated using the multilevel modelling program MlwiN. A polynomial model of order 2 with gender-specific intercepts, linear terms and level 1 variances adequately represents the variance structure of the data. Differential effects of sub-elite for age and gender significantly improved the model. During adolescence both male and female elite youth players have a more promising development pattern of their interval endurance capacity than sub-elite youth players. Explaining variables besides age are percentage body fat, additional training, and motivation.

From the previous studies it becomes clear that psychological skills are important in distinguishing elite and sub-elite youth field hockey players. To find out if this finding is specific for field hockey players or can be generalized to other sports, in Chapter 6 attention is paid to psychological skills of talented athletes in field hockey, basketball, volleyball, speed skating, and swimming. To reveal the relationship between psychological skills and performance level within a talent group, 458 talented youth athletes (mean age 14.80, sd = 1.52) filled in the Dutch Youth Version of the Psychological Skills Inventory for Sports with scales for motivation, confidence, anxiety control, mental preparation, team emphasis, and concentration. A performance level (elite versus sub-elite) by type of sport (team sports versus individual sports) by gender multivariate analysis of covariance (2 X 2 X 2) with age as a covariate resulted in significant effects. Psychological skills distinguished between more and less successful talented athletes, especially in females. In general, psychological profiles differed between males and females and between team sport athletes and individual sport athletes. However, for discrimination of elite and sub-elite youth athletes, motivation and mental preparation were useful indicators that are independent of gender and type of sport.

The most discriminating variable between elite and sub-elite youth field hockey players is tactical skill. However, in the previous studies tactical skills were measured by the opinion of the trainer. To avoid depending solely on the trainer, purpose of the study described in chapter 7 was to develop a practical, reliable, and valid instrument to measure tactical skills in sports

136

directly by the player. Nineteen trainers and 415 competitive youth field hockey and soccer players (mean age 15.9, sd = 1.6; 283 boys and 132 girls) selected by their age, gender, and performance level participated. With the trainers, 34 questions were formulated involving tactical skills. Factor analysis resulted in the Tactical Skills Inventory for Sports (TACSIS). Scales were labeled Positioning and Deciding, Knowing about Ball Actions, Knowing about Others, and Acting in Changing Situations, covering all aspects of tactical skills regarding declarative versus procedural knowledge, and attack and defense. Internal consistency and test-retest measures for reliability (except Knowing about Ball Actions) were within acceptable limits. Elite players scored better than non-elite players, supporting construct validity. In conclusion, the TACSIS is suitable for measuring tactical skills in youth field hockey and soccer players in sports practice.

Based on the results, it is concluded that this thesis contributes to a clearer understanding of the relation between (the development of) multidimensional performance characteristics and the performance level in talented youth field hockey players, and is a relevant step in unraveling the mechanisms of how one achieves greatness in sports. However, the definition of talent used in the present thesis is still vague: what exactly does it mean when a player is ‘better than peers during training and competition’ and how can we measure ‘the potential to become an elite performer in the future’? With caution because the talented players from this study have not yet reached expert performance in adulthood, and with acknowledging the limitations of this study, it is concluded that a talented field hockey player with the greatest chance of succeeding is a player with a relatively high level of performance in field hockey specific physiological characteristics, excellent technical skills, excellent tactical skills, and a very high motivation at the age of fourteen already. This, however, is not enough. A player also has to have potential to reach elite status in the future. Elite players need less time to develop better performance characteristics, meaning that a talented player has to increase his or her performance characteristics at a relatively fast pace for many years in a row. To sustain the long road to the top, investing enormous amounts of time preparing for the international sporting arena, again motivation is essential. In conclusion, an elite player distinguishes him/herself from a sub-elite player not by physiological or anthropometric characteristics but by excellent technical, tactical and psychological skills. In the guidance of young talented players to the top more attention has to be paid to these skills.

Samenvatting

138

Nederland heeft een lange hockeyhistorie en is één van de toonaangevende landen als het om tophockey gaat. De meeste tophockeyers zijn begonnen met hun sport toen ze 7 jaar oud waren en allemaal hebben ze veel tijd en energie geïnvesteerd in hun hockeyloopbaan voordat ze de top bereikten. Huidige jeugdspelers hebben dan ook een lange weg te gaan naar de top. De meeste tophockeyclubs hebben een jeugdopleiding om de prestatiebepalende kwaliteiten van getalenteerde spelers vanaf ongeveer 12 jaar verder te ontwikkelen. In dit onderzoek wordt een getalenteerde hockeyer gedefinieerd als een hockeyer die beter presteert dan leeftijdsgenoten en bovendien de potentie heeft om de top te halen. De beste jeugdhockeyers, in dit onderzoek de jeugdige toppers genoemd, spelen niet alleen in de jeugdopleiding van hun club maar ook in een districts- of nationale jeugdselectie van de Koninklijke Nederlandse Hockey Bond (KNHB). Jeugdige subtoppers hockeyen daarentegen alleen in de jeugdopleiding van hun eigen club. Zowel jeugdige toppers als subtoppers spelen met hun team op het hoogste nationale niveau van hun leeftijdscategorie. Van alle talenten zullen uiteindelijk maar weinig daadwerkelijk in staat zijn om ook bij de senioren het hoogste niveau te halen. Relevante vragen zijn: wat zijn kenmerken van getalenteerde hockeyers? Wie haalt de top en wie niet? Op welke prestatiebepalende kwaliteiten onderscheiden jeugdige toppers zich van jeugdige subtoppers? Het doel van het huidige onderzoek is het geven van meer inzicht in de relatie tussen (de ontwikkeling van) prestatiebepalende kwaliteiten en het prestatieniveau bij jeugdige getalenteerde hockeyers. Het gaat dus om het krijgen van meer inzicht in de kwaliteiten die getalenteerde hockeyers moeten bezitten en ontwikkelen om door te kunnen groeien naar de top. De multidimensionele prestatiebepalende kwaliteiten zijn de kwaliteiten die de hockeyprestatie bepalen: de antropometrische eigenschappen (lengte, gewicht, vetpercentage), de fysiologische kwaliteiten (maximale shuttle sprint, herhaalde shuttle sprint, slalom sprint en interval uithoudingsvermogen), de technische kwaliteiten (maximale shuttle dribbel, herhaalde shuttle dribbel en slalom dribbel), de tactische kwaliteiten (algemene tactiek, tactiek bij balbezit en tactiek bij niet-balbezit) en de mentale kwaliteiten (motivatie, zelfvertrouwen, angstcontrole, mentale voorbereiding, teamoriëntatie en concentratie). Om het bovengenoemde doel te bereiken, zijn de prestatiebepalende kwaliteiten op een sportspecifieke manier gemeten binnen een groep getalenteerde hockeyers in de leeftijd van 12-18 jaar. Tevens zijn de talenten gevolgd in de tijd door middel van een longitudinaal onderzoeksdesign.

Voor het meten van prestatiebepalende kwaliteiten zijn valide en betrouwbare testmethoden nodig. In hoofdstuk 2 wordt de ontwikkeling van twee hockeyspecifieke sprint- en dribbel tests beschreven: de Shuttle Sprint en Dribbel Test (ShuttleSDT) en de Slalom Sprint en Dribbel Test (SlalomSDT). Om de betrouwbaarheid van de tests te bepalen, hebben 34 jeugdige hockeyers (12 meisjes en 22 jongens; gemiddelde leeftijd 14.9 jaar,

139

standaarddeviatie 1.5) twee keer deelgenomen aan de ShuttleSDT. Aan de SlalomSDT hebben 21 hockeymeisjes twee keer deelgenomen. De conclusie is dat de ShuttleSDT en de SlalomSDT betrouwbare testen zijn voor het meten van de sprint en dribbel kwaliteiten van jeugdige hockeyers.

In hoofdstuk 3 is de relatie tussen de multidimensionele prestatiebepalende kwaliteiten en het prestatieniveau bij getalenteerde hockeyers nader bestudeerd. Allereerst is de testbatterij beschreven waarmee de verschillende prestatiebepalende kwaliteiten gemeten kunnen worden. De testbatterij bestaat uit bepaling van de lengte, het gewicht en het vetpercentage, de ShuttleSDT, de SlalomSDT, de Interval Shuttle Run Test (ISRT), de ‘Tactiek in Sport’ vragenlijst en de Nederlandstalige jeugdversie van de Psychological Skills Inventory for Sports (PSIS-jeugd). Er is een vergelijking gemaakt tussen 38 jeugdige toppers (17 meisjes en 21 jongens; gemiddelde leeftijd 13.2 jaar, standaarddeviatie 1.3) en 88 jeugdige subtoppers (46 meisjes en 42 jongens; gemiddelde leeftijd 14.2 jaar, standaarddeviatie 1.3) voor wat betreft hun antropometrische, fysiologische, technische, tactische en mentale kwaliteiten. Een multivariate analyse met prestatieniveau (toppers versus subtoppers) en geslacht als factoren en met leeftijd als covariaat laat zien dat jeugdige toppers beter scoren dan jeugdige subtoppers op technische (maximale en herhaalde shuttle dribbel), tactische (algemene tactiek, tactiek bij balbezit en tactiek bij niet-balbezit) en mentale kwaliteiten (motivatie). Uit een discriminant analyse blijken tactiek bij balbezit, motivatie en slalom dribbel de meest discriminerende variabelen tussen jeugdige toppers en subtoppers. Omdat de toppers jonger waren dan de subtoppers maakt leeftijd eveneens onderscheid tussen beide groepen.

Om de prestatiebepalende kwaliteiten te achterhalen welke wellicht toekomstig hockeysucces kunnen voorspellen, zijn de getalenteerde hockeyers in de tijd gevolgd. In hoofdstuk 4 is een vergelijking gemaakt tussen 30 jeugdige toppers (15 meisjes en 15 jongens; gemiddelde leeftijd op het eerste meetmoment 13.9 jaar, standaarddeviatie 1.0) en 35 jeugdige subtoppers (18 meisjes en 17 jongens; gemiddelde leeftijd op het eerste meetmoment 14.4 jaar, standaarddeviatie 1.2) voor wat betreft hun antropometrische, fysiologische, technische, tactische en mentale kwaliteiten. Er zijn metingen verricht gedurende drie wedstrijdseizoenen en er is een multivariate analyse met herhaalde metingen uitgevoerd voor jongens en meisjes afzonderlijk met prestatieniveau (toppers versus subtoppers) en meetmoment (t1 versus t2 versus t3) als factoren en met leeftijd als covariaat. Deze wijst uit dat jeugdige toppers beter presteerden op technische en tactische tests dan jeugdige subtoppers. Bij de meisjes scoorden de toppers daarnaast beter dan de subtoppers op het interval uithoudingsvermogen, de motivatie en het zelfvertrouwen. Bij zowel de jongens als de meisjes lijken toekomstige tophockeyers al op veertienjarige leeftijd uit te blinken in tactiek. Ze vallen tevens op door

140

hun uitstekende techniek en bovendien verbeteren ze hun prestatiebepalende kwaliteiten beter dan subtoppers.

Uit de resultaten van hoofdstuk 4 blijken jeugdige toppers onder meer hun interval uithoudingsvermogen beter te ontwikkelen dan jeugdige subtoppers in de leeftijd van 14 tot 16 jaar. In hoofdstuk 5 wordt nader ingegaan op de achterliggende mechanismen in de ontwikkeling hiervan. Met behulp van een multilevel analyse zijn ontwikkelingscurven voor jeugdige toppers en subtoppers in de leeftijdscategorie van 12 tot 19 jaar tot stand gekomen. Deze curven zijn gemaakt voor zowel jongens als voor meisjes. Het interval uithoudingsvermogen kan daarmee aan de hand van leeftijd, vetpercentage, extra trainingsuren en motivatie voorspeld worden. Tijdens de adolescentie laten, zowel bij de jongens als bij de meisjes, de toppers een positievere ontwikkeling van hun interval uithoudingsvermogen zien dan de subtoppers.

Uit hoofdstuk 3 en 4 kan geconcludeerd worden dat mentale kwaliteiten een belangrijk verschil vormen tussen jeugdige hockeytoppers en -subtoppers. Om te achterhalen of deze bevinding hockeyspecifiek is of gegeneraliseerd kan worden naar meerdere sporten, wordt in hoofdstuk 6 dieper ingegaan op de mentale kwaliteiten van jeugdige getalenteerde hockeyers, basketballers, volleyballers, schaatsers en zwemmers. Om inzicht te krijgen in de relatie tussen mentale kwaliteiten en het prestatieniveau binnen een talentengroep, hebben 458 getalenteerde jeugdige sporters (gemiddelde leeftijd 14.8 jaar, standaarddeviatie 1.5) de PSIS-jeugd ingevuld. Deze vragenlijst bevat schalen voor motivatie, zelfvertrouwen, angstcontrole, mentale voorbereiding, teamoriëntatie en concentratie. Een multivariate analyse met prestatieniveau (toppers versus subtoppers), geslacht en type sport (teamsport versus individuele sport) als factoren en met leeftijd als covariaat resulteert in significante effecten. In het algemeen is het mentale profiel van jongens anders dan dat van meisjes en het mentale profiel van teamsporters anders dan dat van individuele sporters. Desalniettemin maken mentale kwaliteiten onderscheid tussen meer en minder succesvolle sporters, vooral bij meisjes. Op motivatie en mentale voorbereiding scoren jeugdige toppers beter dan subtoppers. Deze mentale kwaliteiten zijn, ongeacht geslacht of type sport, goede indicatoren voor het onderscheid tussen jeugdige toppers en subtoppers.

De meest discriminerende kwaliteit tussen jeugdige hockeytoppers en –subtoppers is tactiek. In hoofdstuk 3 en 4 is tactiek gemeten aan de hand van het oordeel van de trainer. Dit oordeel is echter mogelijk beïnvloed door het prestatieniveau van de speler. Om deze beïnvloeding te ondervangen, wordt in hoofstuk 7 de ontwikkeling beschreven van een praktisch toepasbaar, betrouwbaar en valide meetinstrument om tactiek te meten. In samenwerking met 19 trainers is een vragenlijst met 34 vragen betreffende tactiek opgesteld. Nadat 415 jeugdige wedstrijdhockeyers en –voetballers (283 jongens en 132 meisjes;

141

gemiddelde leeftijd 15.9, standaarddeviatie 1.6) deze vragenlijst hebben ingevuld, is een factor analyse uitgevoerd. Dit resulteerde in de Tactical Skills Inventory for Sports (TACSIS), de vragenlijst voor tactische vaardigheden van sporters. De vragenlijst bevat vier schalen: ‘Positie kiezen en besluitvorming’, ‘Inzicht in acties met de bal’, ‘Inzicht in anderen’ en ‘Omgaan met veranderingen’. Deze schalen bevatten alle aspecten van tactiek voor wat betreft het onderscheid tussen declaratieve kennis (‘weten wat je moet doen’) en procedurele kennis (‘het doen’) en het onderscheid tussen aanval en verdediging. Interne consistentie en test-hertest betrouwbaarheid (behalve de schaal ‘Inzicht in acties met de bal’) zijn acceptabel tot goed. De construct validiteit wordt ondersteund door de bevinding dat jeugdige toppers hoger scoorden dan jeugdige subtoppers. De conclusie was dat de TACSIS geschikt is om in de praktijk tactische vaardigheden te meten bij jeugdige hockeyers en voetballers.

Op basis van de resultaten wordt geconcludeerd dat dit onderzoek meer inzicht geeft in de relatie tussen de (ontwikkeling van) multidimensionele prestatiebepalende kwaliteiten en het prestatieniveau bij jeugdig getalenteerde hockeyers en een relevante stap is in het ontrafelen van het mysterieuze begrip talent. Er zijn echter nog veel onduidelijkheden. Zo is bijvoorbeeld de definitie van ‘talent’ nog steeds vaag en verdient het de aanbeveling om de onderliggende prestatiebepalende kwaliteiten nader te bestuderen. Voorzichtig, omdat de talenten de top nog niet gehaald hebben en met inachtneming van de beperkingen van het onderzoek, wordt geconcludeerd dat een getalenteerde hockeyer de grootste kans heeft om te slagen als hij of zij al op veertienjarige leeftijd een hoog niveau heeft van hockeyspecifieke fysiologische kwaliteiten, een uitmuntende techniek heeft en vooral een uitmuntende tactiek combineert met een zeer hoge motivatie. Dit is echter nog niet genoeg. Een speler moet ook de potentie hebben om de top te halen. Omdat jeugdige toppers in vergelijking met jeugdige subtoppers minder tijd nodig hebben om betere prestatiebepalende kwaliteiten te ontwikkelen, betekent dit dat een talent zijn of haar kwaliteiten gedurende vele jaren in een relatief hoog tempo moet ontwikkelen. Om de lange weg naar de top vol te kunnen houden is motivatie wederom essentieel. Kortom, een jeugdige tophockeyer onderscheidt zich van een jeugdige subtopper niet zozeer door antropometrische of fysiologische kwaliteiten, maar juist door een uitstekende techniek, tactiek en mentale kwaliteiten. Binnen talentontwikkeling zou vooral aandacht moeten worden besteed aan deze kwaliteiten.

142

Voor trainers, coaches, spelers, ouders en andere hockey enthousiastelingen biedt dit onderzoek relevante aanknopingspunten:

- Houd rekening met het multidimensionele karakter van hockey: een talent is meer dan een technisch goede speler.

- Motivatie speelt een essentiële rol in de ontwikkeling van een succesvolle hockeyloopbaan.

- Om de top te kunnen halen, moeten de techniek en vooral de tactiek uitmuntend zijn.

- Bovendien moet een getalenteerde speler over een relatief hoog basisniveau van hockeyspecifieke fysiologische kwaliteiten beschikken, wat wil zeggen dat hij of zij snel moet kunnen sprinten over korte afstanden, de sprints herhaaldelijk moet kunnen uitvoeren, wendbaar moet zijn en een zeer goed interval uithoudingsvermogen moet hebben.

Er wordt aanbevolen om tijdens het hele proces van talentontwikkeling regelmatig, bijvoorbeeld ieder jaar, een prestatieprofiel op te stellen van jeugdige hockeyers. Op deze manier kan het niveau van de prestatiebepalende kwaliteiten per speler worden vergeleken met andere getalenteerde hockeyers. Bovendien kan zijn of haar ontwikkeling van deze kwaliteiten in kaart gebracht worden en deze informatie kan toegepast worden in de trainingen. Voor het opstellen van het prestatieprofiel kan gebruik gemaakt worden van de testbatterij zoals die voor dit onderzoek ontwikkeld is. In Vakblad Hockey is de testbatterij beschreven en worden de prestatieprofielen van getalenteerde jongens en meisjes onder 14, onder 16 en onder 18 jaar gepresenteerd (Elferink-Gemser et al., 2004a; 2004b).

List of publications

144

Peer-reviewed articles Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., Mulder, Th. (2004). Relation

between multidimensional performance characteristics and level of performance in talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

Elferink-Gemser, M.T., Visscher, C., Richart, H., Lemmink, K.A.P.M. (2004). Development of the Tactical Skills Inventory for Sports. Perceptual and Motor Skills, 99, 883-895.

Lemmink, K.A.P.M., Elferink-Gemser, M.T., en Visscher, C. (2004). Evaluation of the reliability of two field hockey-specific sprint and dribble tests in young field hockey players. British Journal of Sports Medicine, 38, 138-142.

Chapters in books Visscher, C., Elferink-Gemser, M.T. en Lemmink, K.A.P.M. (2004). The role of parental

support in sports success of talented young Dutch athletes. In: M. Coelho e Silva en R.M. Malina (eds). Children and Youth in Organized Sports. Coimbra University Press, Portugal, 123-135.

Proceedings Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M. (2002). Relation between

predictors and performance level in talented young field hockey players. In: Proceedings of the 7th annual congress of the European College of Sport Science, Athens, Greece, 619.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M. (2003). Multidimensional performance characteristics in talented youth field hockey players – a longitudinal study. In: Proceedings of the 8th annual congress of the European College of Sport Science, Salzburg, Austria, 161.

Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M. (2004). Are today’s young top players the stars of tomorrow? In: Proceedings of the Young Researcher Seminar, University of Innsbruck, Austria. Young Researcher Award for best oral presentation.

Lemmink, K.A.P.M., Elferink-Gemser, M.T., Visscher, C. (2003). Interval sprint and interval endurance capacity of young soccer players. In: Proceedings of the 8th annual congress of the European College of Sport Science, Salzburg, Austria, 226.

Visscher, C., Elferink-Gemser, M.T., Lemmink, K.A.P.M. (2003). The role of parental support in sports success of talented young Dutch athletes. In: Proceedings of the 8th annual congress of the European College of Sport Science, Salzburg, Austria, 143.

145

Nederlandse vakbladen en rapporten NOC*NSF Elferink-Gemser, M.T. (2004). Talent ontrafeld? Onderzoek Rijksuniversiteit Groningen

afgerond. Hartsticke Bosch, 5, 16-17. Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M. (2004). Prestatieprofielen van

jeugdig getalenteerde hockeyers: deel 1. Vakblad Hockey. Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M. (2004). Prestatieprofielen van

jeugdig getalenteerde hockeyers: deel 2. Vakblad Hockey. Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M. (2005). De weg naar de top:

kwestie van mentaliteit? Mentale kwaliteiten van getalenteerde sporters in de leeftijd van 12-18 jaar. Een onderzoek bij voetballers, hockeyers, volleyballers, basketballers, schaatsers en zwemmers. NOC*NSF. Sector Topsport/publ./in press.

Elferink-Gemser, M.T., Knoop, J., Lemmink, K.A.P.M., Visscher, C. (2004). Topvoetbal: kwestie van mentaliteit? De VoetbalTrainer, 125, 40-41.

Visscher, C., Gemser, M.T. en De Greef, M. (1996a). Jeugdige topsporters. Invloed van ouders en onderwijs. Richting Sportgericht, 5, 263-268.

Visscher, C., Gemser, M.T. en De Greef, M. (1996b). Jeugdige topsporters. Invloed van ouders en onderwijs (2). Richting Sportgericht, 6, 329-332.

Visscher, C., Gemser, M.T. en De Greef, M. (1997). Talentontwikkeling. De invloed van ouders en onderwijs. NOC*NSF. Sector Topsport/publ./TO 005.

Visscher, C., Bakema, R., Elferink-Gemser, M.T. en Lemmink, K.A.P.M. (2003). Uitval binnen het jeugdturnen. Sportgericht, 3, 4-8.

Dankwoord

148

Mijn proefschrift is af! DANK DANK DANK aan een ieder die me daarbij geholpen heeft (en dat zijn er een heleboel!). Maar…. ik ben nog lang niet klaar…. Ik weet niet meer precies wat ik verwachtte toen ik vijf jaar geleden aan dit talentonderzoek begon, maar wat ik nog wel weet is dat ik in de veronderstelling leefde dat ik antwoorden op vragen zou kunnen geven. Dat als het moment van promoveren was aangebroken ik toch zeker wel één of misschien zelfs wel twee vragen minder had over hoe het eigenlijk zit als sportieve talenten de top willen halen. Niets is echter minder waar; in ruil voor iets meer inzicht in de onvoorstelbare complexiteit van de ontwikkeling van een succesvolle sportloopbaan zijn er legio nieuwe vragen bij gekomen. Dit voortschrijdende inzicht heb ik vooral en in eerste instantie te danken aan de projectleider van het talentonderzoek, Chris Visscher. Chris, je creativiteit ligt ten grondslag aan alles wat dit project voorstelt. Je bent echter veel meer voor mij dan alleen mijn directe begeleider. Jij hebt als geen ander oog voor dingen die echt belangrijk zijn in het leven. Soms schieten woorden tekort, maar ik denk dat je wel weet wat ik bedoel. Het fundament van het sportonderzoek bij Bewegingswetenschappen wordt met Chris gevormd door Koen Lemmink. Samen vormen jullie een hecht team en ik ben blij dat ik door jullie allebei begeleid ben. Van jou, Chris, moest een artikel vaak weer helemaal over de kop omdat de rode lijn er toch nog niet goed genoeg inzat, terwijl jij, Koen, veel meer de ‘puntjes op de ‘i’ zette. Van jou heb ik enorm veel geleerd over het schrijven van een wetenschappelijk artikel. Maar waar ik jullie vooral voor wil bedanken is de sfeer waarin we samenwerken; altijd doelmatig maar vooral altijd gezellig. Natuurlijk wil ik ook mijn promotor, Theo Mulder, bedanken. Theo, ik moet altijd een beetje lachen om jouw afkeer van topsport en alles wat daarmee te maken heeft. Gelukkig heeft je schijnbaar onbegrensde interesse in het menselijk brein en -aanpassingsvermogen ook je enthousiasme voor het talentonderzoek opgelaaid. Ik hoop in de toekomst nog vaak met je in wetenschappelijk debat te gaan. Een onderzoek komt niet van de grond zonder sponsoren en mensen in de praktijk die mee willen denken en werken. In het licht hiervan wil ik allereerst graag Geert Slot van het NOC*NSF bedanken voor zijn grote persoonlijke betrokkenheid. Ook ben ik Martijn Schakel en Iwan Doyer van hockeyvereniging Rotterdam en Wim Kemps van hockeyvereniging ’s Hertogenbosch veel dank verschuldigd. Ook al was het waterveld bezet, zat het wedstrijdprogramma overvol en moest het trainingsprogramma er op aangepast worden, toch lukte het ieder jaar weer om de metingen te verrichten. Hiervoor mijn dank aan alle spelers en speelsters, trainers, coaches, begeleiders en ouders. Niet alleen van de hockeyverenigingen in Rotterdam en Den Bosch, maar ook van GHHC, GHBS, HC Eelde en het dr. Nassau College waar we pilot-onderzoeken hebben uitgevoerd. En van de voetbalverenigingen FC Groningen,

149

Sc Heerenveen, Be Quick en Helpman, want ook al staat het niet in dit proefschrift, alle metingen die we bij de hockeyers hebben verricht, zijn ook uitgevoerd bij voetballers. Mijn dank gaat ook uit naar Marc Lammers en de dames van het nationale hockeyteam die aan de tests hebben deelgenomen om ‘de afstand tot de top’ te bepalen. Sinds de start van het project zijn er twaalf studenten afgestudeerd binnen het talentonderzoek. Zij hebben meegeholpen met de metingen en meegedacht over afzonderlijke vraagstellingen. Bedankt Mieke, Inge, Marlies, Joke, Karen, Suzanne, Yvonne, Irene, Moniek, Nynke, Jesper en Ruud. Mieke, jou wil ik graag in het bijzonder bedanken omdat jij ook als onderzoeksassistent veel hebt bijgedragen aan het talentonderzoek. Verder zijn er natuurlijk nog de talloze testleiders die herhaaldelijk meehielpen om de organisatie op de testdagen rond te krijgen en zonder wie het allemaal niet gelukt was. Speciaal wil ik hierbij een aantal toptestleiders bedanken: Steven, Rienk, Eline en Alien. ‘Jongens’, bedankt! Het tot een goed einde brengen van een promotietraject is een hele klus. Een klus die een stuk leuker is met een groep enthousiaste collega’s om je heen. En daar heb ik het enorm mee getroffen. Eerst op de gang van de derde verdieping in het AZG (ja, ik weet het, het is nu UMCG), later op de derde bij BW. Ik heb wat afgelachen met Bianca, Leontien, Sandra, Anuschka, Rients, Wietske, Grieke en Juha. En nu met Esther, Mieke, Suzanne, Henri, de Robben, Helco, Geir en alle anderen. Als hele goede vriendinnen wil ik ik in het bijzonder Esther en Bianca bedanken omdat jullie zoveel meer zijn dan collega’s. Ik vind het heel bijzonder, Esther, dat we ook nog eens kamer- en huisgenootjes zijn en Bianc, ook al botsen we wel eens met anderen, ik wil nog graag met je in de auto naar het zwembad. Eigenlijk zou ik nog een hele tijd zo door willen gaan en iedereen willen bedanken die belangrijk voor me is. Nathalie, Rienk en kleine Tom bijvoorbeeld, of Robert en Iris (nog heel erg bedankt voor je hulp bij het lay-outen, en...nee, we laten Arjan echt niet nog een keer winnen!). En Edwin, ook jij bedankt voor je hulp bij de grafieken. En natuurlijk mijn familie. Papa en mama, Jildou en Volkert, Pieter-Jorn en Corina, pake niet te vergeten, maar natuurlijk ook Jan en Siny, Annemarie en Henri, Steven en Anna en Clemens en Laura die stuk voor stuk zo nu en dan hun wenkbrauwen optrokken omdat ik alweeeeer moest werken. En dan zijn er nog mijn lievelingsneefjes en –nichtjes die ervoor zorgen dat mijn hoofd zo nu en dan weer even leeg wordt: Sven, Vera-Lou, Nori, Stijn, Linde en Peike Joeri. Maar het allerbelangrijkste ben jij, Arjan. Jij weet als geen ander hoe belangrijk dit voor mij is. Zonder jou zou ik het niet redden. Zonder jou wil ik het niet eens redden. Lieve Arjan: we gingen voor goud en vonden elkaar! Ook al ben ik blij dat het proefschrift nu af is, ik beëindig dit dankwoord zoals ik het begonnen ben: mijn proefschrift is af, maar ik ben nog lang niet klaar.