Davidsson 2003 - The Business Platform: Developing an Instrument to Gauge and to Assist the...

DAVIDSSON AND KLOFSTEN 1

Assessing the state of newly startedfirms in order to predict their fate or in order to initiate appropriate action for increasing their probability of sur-vival and growth is no easy task (Hall1995). Such firms are in a turbulentphase of development where businessactivities are carried out in a short-termperspective and where it can be difficultto perceive more fundamental shortcom-

ings that are overshadowed by everydayproblems (Adizes 1988). Much researchhas been devoted to establishing associ-ations between various kinds of pre-sumed causal factors on the one handand the ability of a company to attain stability and growth on the other (forexample, Cooper 1981; van de Ven,Hudson, and Schroeder 1984; Kazanjian1988; Cooper, Gimeno-Gascon, and Woo,

Journal of Small Business Management 2003 41(1), pp. 126

The Business Platform: Developing an Instrument to Gauge and to Assist the Development of Young Firmsby Per Davidsson and Magnus Klofsten

The research presented here addresses the following problems we perceive inresearch on the development of young firms. First, we feel there is a lack of holisticyet quantifiable and generalizable ways in which to assess the state of newly startedfirms. Quantitative research typically relies on additive models that are unable toexplain more than half of the outcome variance at the most. Holistic approaches tendto be qualitative and therefore have unknown generalizability. Second, we feel thereis a lack of action-oriented assessment models that are firmly anchored in research.Research models typically take a passive prediction position and often build on rela-tionships that give little hands-on advice to managers. Numerous practical tools forassessing and developing firms during their early development can be found in how-to literature, but these typically are not anchored in systematic research and there-fore have unknown validity. Hence, what we set out to do in the research presentedin this article is to develop a quantifiable, holistic, and research-based instrumentfor assessing and assisting the development of young firms.

Per Davidsson is professor of entrepreneurship at the Jnkping International BusinessSchool ( JIBS), Sweden.

Magnus Klofsten is professor of entrepreneurship and director of the Centre for Innova-tion and Entrepreneurship (CIE) at Linkping University, Sweden.

1994; Davidsson 1991; Jones-Evans and Klofsten 1997; Gimeno et al. 1997;Morris 1998; Littunen, Storhammar, andNenonen 1998; Dahlquist, Davidsson,and Wiklund 2000). A general conclusionfrom this research is that a range offactors on the individual, firm, and envi-ronmental levels of analysis shape thefirms development. At the same time, theresults suggest that there is no individ-ual factor that universally and by itselfhas a strong determining influence.Further, explanatory models based onadditive effects of comprehensive lists ofpresumed causes provide far from fullexplanations of the outcome variance,supporting the notion that holistic and tosome extent idiosyncratic configurationsof factors jointly determine the successprobabilities of young firms.

Models for assessing and assisting thedevelopment of firms are a key interestfor research, teaching, and business practice. Numerous manuals have beenwritten that more concretely treat differ-ent aspects of new business developmentand that provide considerable amountsof practical advice on how a firm canincrease its chances for survival andsuccess. In this type of literature, com-prehensive problems in the firm are dis-cussed, but no serious attempts are madeto anchor the advice in systematic empir-ical research or to develop testable the-ories. The suggestions are based more onexperience and on what for the momentis considered to be good managementpractice (Hall 1995). Although researchin the last 10-year period has grown inscope and although our knowledge ofearly-growth and development processesin firms is considerably greater thanbefore, it is still fairly unusual forresearchers to try to generate practicaltools in this field.

In the present research we attempt to develop a quantifiable and research-based instrument for assessing and as-sisting the development of young firms.The implication of successfully develop-

ing and validating such an instrument is that it would provide managers andbusiness consultants with a better toolwith which to assess and assist youngbusinesses. In short, the tools that areavailable today tend to fall into one of the following three categories: (1)based on research, generalizable but not action oriented (quantitative, pre-diction-oriented models); (2) based onresearch but with unknown generaliz-ability and varying degree of action ori-entation (qualitative or understanding- or action-oriented research); or (3) notresearch based but action oriented (forexample, how-to literature). We attemptto get closer to a situation where theassessment and assisting of the devel-opment of young firms can be guided by a tool that is research based andgeneralizable and action oriented. Al-though we will not reach all the way toa fully developed and validated toolwithin the frame of this research, we feel that steps in this direction areneeded urgently. The instrument we try to develop is based on KlofstensBusiness Platform Model, which buildson previous research and has been usedextensively in qualitative research andbusiness consulting.

The Business PlatformModel

It has been shown that firms that havesucceeded in surviving two to three yearsand that have gone through a number ofcrucial phases attain a stable foundationfrom which they can continue to develop(Mayer and Goldstein 1961; Freeman,Carrol, and Hannan, 1983; Hall 1995).Gibb and Scott (1986) introduced theconcept base for potential developmentas an expression of this stabilizing con-dition. According to the authors, a devel-opment base has been attained when thenewly started firm is developed suffi-ciently concerning resources (capacity),experiences, control, leadership, andidea. With this basis, the firm then has

2 JOURNAL OF SMALL BUSINESS MANAGEMENT

the possibility to develop and to managefuture environmental changes andthereby can be considered to haveachieved stability.

Pursuing a similar idea, Klofsten(1992) conducted a comprehensive liter-ature review and defined eight firm-levelcornerstones that determine a firmsearly development process. These eightcornerstones are the business idea, theproduct, the market, the organization,core group expertise, core group drive/motivation, customer relations, and otherrelations, all of which make up the business platform model. The corner-stones are further explained in Table 1,which also displays the previousresearch on which each cornerstonebuilds or with which each cornerstoneaccords. The purpose of the cornerstonesis to describe the early developmentprocess in a holistic manner on themicrolevel. It comprises the developmentprocess itself (idea, product, market, andorganization); key actors such as thefounders, chief executive officer (CEO),and board members (expertise anddrive/motivation), and the flow of exter-nal resources (customer relations andother firm relations).

The fundamental thesis of this modelis that firms are vulnerable in their earlylife and continually run the risk of dis-appearing from the market. Success inthe firm is determined by how well thisvulnerability is overcome, and the earlydevelopment process is one of the mostimportant periods in the life of the firm.The thoughts and the driving forces inthe firm and the actions being taken atthis time can be decisive for the contin-ued growth and development of the firm(cf. Kimberly and Miles 1980). However,the model is not a passive predictionmodel. Its purpose is to assess so as tofind remedy, if need be.

As long as there is no very dramaticchanges in the firms environmentchanges that the firm cannot prevent,such as the loss of a key person or a

rapid decline in market potentialthetheory is that success will be determinedby how well the firm builds and main-tains its business platform. A firm attainsa platform by satisfying two criteria:securing an input of resources anddeveloping an ability to manage andutilize such resources (cf. Barney 1997,ch. 5). After the platform has beenachieved, a firm has a good deal ofleeway in generating and managing itsresources.

Early development in a firm is definedby the business platform cornerstonesprogress and can take as little or as muchtime as is needed, with some firms neveractually getting beyond this stage. Takingsteps to realize a business concept by ini-tiating activities intended to lead to thecreation of a firm is the beginning ofearly development. It ends when the firmhas established a business platform. Atthe risk of being categorical, it could beargued that firms aspiring to grow andto become significant actors in themarket must sooner or later attain abusiness platform.

The reasoning, then, is that the cor-nerstones that make up a business platform must reach certain minimumlevels if a business platform is to beattained (see Table 1). For example, anidea must be communicable both withinand outside the firm (Kazanjian 1988;Timmons 1994). A hobby firm will notexhibit the same levels of activity andprogressiveness as a firm with strongdriving forces to grow and develop(Shaver and Scott 1991; Naffziger,Hornsby, and Kuratko 1994). Client con-tacts are not enough; there actually must be customers who are buying continuously (Zeithaml and Bitner 1996;Prenkert 1998). Thus, the model is not a simple additive and compensatorymodel. Reaching high levels on severaldimensions is not assumed necessarily to make up for severe shortcomings on another dimension, and sometimesachieving average levels across all


4JO

URN

AL O

F SMA

LL BU

SINESS M

AN

AG

EM

EN

T

Table 1Cornerstones of the Business Platform

Cornerstone Previous Research Minimum Levels to Attain

Formulation and Clarification of Kazanjian (1988) The idea must be clarified so that the special the Business Idea Bjerke (1989) know-how that makes up the commercial

Hills (1994) springboard is understandable and can be Timmons (1994) communicated internally and externally.Cooper, Fotta, and Woo (1995)Kohen and Kohli (1998)

Development to Finished Adizes (1988) Once the product is available, it must gain Product Kazanjian (1988) acceptance by one or more reference

Roberts (1991) customersthe firm has then proven that it is Rothwell (1992) capable of satisfying markets needs and Autio (1997) wants.Dyer and Gupta (1999)

Definition of Market Miller and Friesen (1984) The firm must define a market that is large Hisrich (1992) enough and profitable enough to ensureWesthead and Birley (1993) survival.Hall and Adams (1994)Weinstein (1994)OGorman (1997)

Development of an Operational Mintzberg (1973) The running of business operations requires Organization Greiner (1972) the existence of an organizational structure

Kazanjian (1988) that facilitates functional coordinationthisKazanjian and Drazin (1990) structure should take advantage of the firmsChandler and Hanks (1994) inherent flexibility and innovative ability and Barney (1997) should be fairly effective at internalShepherd and Shanley (1998) coordination and at maintaining and

developing external relations.

DAV

IDSSO

N A

ND

KLO

FSTEN

5

Table 1Continued

Cornerstone Previous Research Minimum Levels to Attain

Core Group Competence Susbauer (1967) A business firm must have technological and Cooper and Bruno (1977) commercial competence to develop itsBrockhaus (1980) products and marketit is crucial to have Chandler and Hanks (1994) access to expertise for solving the firms real Bird (1995) problems.Walsh and Kirchhoff (1998)

Commitment of the Core Group McClelland (1961) A basic requirement for development is that at and the Prime Motivation of Smith (1967) least one person is highly motivated andEach Actor Birley and Norburn (1985) that the other key actors are committed to

Gartner (1985) the business idea.Shaver and Scott (1991)Naffziger, Hornsby and Kuratko

(1994)

Customer Relations Roberts and Wainer (1968) A customer base must be qualitatively and Utterback and Reitberger (1982) quantitatively strong enough to generateAijo (1996) operating revenue.Zeithaml and Bitner (1996)Prenkert (1998)Krieger (1998)

Other Relations Bollinger, Hope and Utterback The firm may sometimes need additional (1981) capital, management know-how, or other oil

Knight (1986) in its machinerythese relations complement Bruno and Tyebjee (1984) the customer relationships.Olofsson and Wahlbin (1984)Autio (1995)Heydebreck (1997)

cornerstones does not suffice either. Con-versely, it sometimes is necessary tostand outto have a real edgeon atleast some dimension. This is consistentwith the resource-based view of the firms(Penrose 1959; Barney 1991, 1997).

Based on three comprehensive andlongitudinal case studies, Klofsten (1992)showed that it was possible to analyzethe state of the eight cornerstones at dif-ferent points in time and to determinewhether a business platform has beenattained. He also found reason to arguethat if a business platform it is notattained, the firm will sooner or later gounder and will disappear from themarketat least as independent actor.The model originally was applied totechnology-based firms but also hasproven to be applicable to other types offirms (Klofsten 1998).

What distinguishes the business plat-form model from other business modelsis the fact that in a clear and simple way,it specifies the cornerstones that definethe early development process of thefirm and what is demanded of every cor-nerstone in order for the firm to over-come its initial vulnerability. Businessmodels created by Gibb and Scott (1986),for example, and even those more practically oriented how-to models (busi-ness plan models) conduct an almostexclusive discussion on variable or fac-tor level. These models fail to indicatewhat is demanded of them in order tomake the firm overcome certain barriersand in order to take the next step in itsevolution.

The business platform model hasbeen disseminated (predominantly inEurope) in research, education, and tradeand industry (Klofsten 1992, 1994, 1997,1998). The experience to date is that

the platform model is a useful tool forgauging and assisting the developmentof young firms. An obvious limitation isthat its validity has not been proven yet in broadly based research. A neces-sary first step toward such validation is to develop a standardized opera-tionalization of firms standing on thecornerstones of the platform model.Developing an instrument that could be used for that purpose is what we set out to do in the present study. It is our hope that the development of such an instrument will facilitate theadoption and sound application of the business platform model by businessconsultants.

Our aspiration for the present paperis to develop an instrument that canquantify firms state on each cornerstonedimension. It should be clear from theprevious discussion that simple summa-tion of scores from such a quantificationwill not suffice for determining whethera firm has attained a business platformor not. The appropriate minimum levelfor each dimension, the possible need forexcellent scores on some dimensions,and the variation by industry or type offirm all require further research.

The SampleThe data that are presented in this

study derive from technology- andknowledge-based firms located atSwedish technopoles1 (Heydebreck,Klofsten, and Maier 2000). Addresses tothe firms were obtained from organiza-tions at each technopole that work withdifferent forms of business support, suchas science parks and innovation centersat universities. We were of the opinionthat by help of these technopole organi-zations, we should in an effective way get


1Technopoles are regions that exhibit strong technology- and knowledge-based new busi-ness development. They are a fertile combination of a university with a technical profile, aresearch institute, and science parks as well as a number of large and small firms. In Sweden,Gteborg, Linkping, Lund, Stockholm, and Uppsala are the main regions classified astechnopoles.

access to companies for our study. As theaim of the paper was to study growthand development in young firms, firmsmore than 10 years old were excluded.We thought those companies older than10 years would have difficulties in recall-ing details from their early developmentprocess. At the same time, it was con-sidered an advantage if the firms hadsome history and if a lower age limit oftwo years was set. This left 313 firms, allof which were sent a questionnaire(addressed to the CEO or other decision-making executive) either in electronicform (260 firms) or with traditional mail(53 firms). After a telephone reminder, atotal of 114 firms returned the question-naire, giving a response frequency of 36percent. While these response rates are adisappointment compared to what hasbeen obtained in some Swedish studies(cf. Davidsson 1989a; Wiklund 1998),they are high compared to most pub-lished mail survey research internation-ally. Importantly, in the present contextwe are not trying to obtain populationestimates, such as norm values, for awell-defined population. What we needis a large enough sample from a relevantpopulationthat is, one whose membersare likely subjects for real application ofthe business platform modelin order toassess the technical properties of theinstrument we are trying to develop.Therefore, the levels of nonresponse arenot severe for our purposes. Internalnonresponse to individual parts or itemsis a worse problem, because it may indi-cate some type of inadequacy of theinstrument. This is an issue to which wewill have reason to return.

The average and median ages of thefirms were 6.7 and 7.0 years, respec-tively. During 1998, the average turnoverwas Swedish crones (SEK) 9.6 million(US$ 1.1 million) and the median SEK 9.0million (US$ 1.0 million) with a range ofSEK 071 million (US$ 07.9 million). Anaverage of 4.6 persons (median 3.0) wasemployed in the firms (range 0100). A

majority of 66 firms (58 percent) wereoriented toward providing services while40 firms (35 percent) primarily producedproducts. Six firms (5 percent) reportedother forms of business but did notspecify what kind. Slightly more than aquarter of the firms, 30 (26 percent),reported at this time that they hadchanged the focus of their business sincestartup. Most of the firms (55 firms or 48 percent) stated that they have a background from some universityforexample that the founders had beenresearchers or students immediatelyprior to startup. Those who reported otherwise (34 firms or 30 percent) camefrom firms or research institutes.

The InstrumentA team of four people, led by the

authors of this paper, developed theoperationalizations. Klofsten developedthe business platform model (Klofsten1992, 1994, 1997, 1998) and has sub-stantial consulting experience from itsuse, and Davidsson has considerableexpertise in survey research and devel-opment of measuring instruments (seeBellu, Davidsson, and Goldfarb 1990;Brown, Davidsson, and Wiklund forth-coming; Davidsson 1989a, 1989b, 1989c,1991, 1995a, 1995b; Davidsson andWiklund 1997; Wiklund et al. 1997).

In developing the instrument, weadopted the standard procedure of gen-erating multiple-item batteries of ques-tions for each intended dimension. Inorder to arrive at a manageable instru-ment, we aimed at approximately fiveitems for each dimension. We first gen-erated a pool of about five to 10 rawitems for each dimension. Each sug-gested item was discussed among theteam from substantive as well as fromtechnical standpoints. Through thisprocess of selecting, revising, and delet-ing items, we arrived at the final versionthat was mailed out, which includes atotal of 36 items, or between three andsix for each dimension.


Following the recommendations ofConverse and Presser (1986) we chose aforced-choice format. The respondentswere asked to indicate their relativedegree of agreement on a five-point scalebetween the two statements that werecontrasted. All items in the sent-outversion are listed in Table 1. Importantly,for each item the respondent was askedto give separate responses for the firmscurrent situation and for its status twoyears earlier. In that way we could maketwoalbeit not two independenteval-uations of the technical properties of theinstrument.

The package of questions aimed atcapturing the various dimensions of thebusiness platform constituted the lionsshare of the mail/email questionnaire.The remainder of the questionnaire con-cerned background facts about the firmand its founders as well as a questionconcerning the relative ease or difficultyof answering the business platform items.

Analysis MethodIf our operationalization is successful,

the items aimed at assessing the samecornerstone dimension (that is, Idea,Product, Market, Organization, and soforth) should have high intercorrelations,thus allowing summing the items to anindex and thereby reducing the influenceof random measurement error. In orderto assess our measures suitability forindex construction we ran the StatisticalPacke for Social Science (SPSS) reliabilityanalysis routine. In the interpretation ofthis analysis we applied Nunnallys ruleof thumb that a chronbach alpha above0.70 indicates satisfactory measurementquality from a technical standpoint (Nunnally 1967; Nunnally and Bernstein1994). We checked interitem correlations,item-to-total index correlations, and theeffect on alpha of dropping one item ata time. Items were dropped from therespective index. It turned out that theitem would lower rather than increase the Alpha value for the index.

In order to be useful, the resultingmeasures should not only be reliable butalso should be distinct from one another.There should be proof that packages ofitems intended to measure differentdimensions really do sothat is, theresulting indices should have discrimi-nant validity. In order to assess whetherthe dimensions proposed by the businessplatform model are empirically distinct,we ran SPSS factor analysis with princi-pal component extraction and varimax(orthogonal) rotation. The logic behindthis is that if items within a dimensionhave higher intercorrelations than theircross-correlations with items of otherdimensions, then the different item pack-ages should have a tendency to form sep-arate factors in factor analysis. However,we did not assume a priori that the dif-ferent dimensions (or cornerstones) ofthe business platform model would beuncorrelated (orthogonal). For instance,they should all be a function of time tosome extent. Therefore, a crystal-clearfactor pattern was not expected, so wealso evaluated the question of discrimi-nant validity via correlation analysis.Another reason for this is that summatedindices are not identical to factors in afactor analysis, and the summatedindices are what we propose be usedeventually.

The operationalization should notonly be technically adequate but alsoshould reflect the empirical reality thatthe conceptualization aims to capture. Atrue test of this would require the exis-tence of another known-to-be-validmethod for assessing the business plat-form model cornerstones. Obviously,such a yardstick does not exist, but asthe items were developed by a collectiveof four people, all of whom had experi-ence with the platform model, we feelconfident that the items reflect theintended conceptual meaning of eachcornerstone. In order to allow the readerto make his or her own assessment, weoffer the translated instrument verbatim


(translated from the Swedish original) inTable 2. The table also specifies (inparentheses) which items were excludedas a result of decreasing rather thanimproving the reliability of the respectivescale.

ResultsTable 3 displays Cronbachs alpha

values and other statistics for the result-ing indices, both concerning the currentsituation and the firms status two yearsearlier. We will comment the results foreach cornerstone.

Regarding the business idea, three outof the four original items could beretainedthat is, for one item it turnedout that its inclusion would lower ratherthan increase the reliability of the indexand consequently was dropped (see alsoTable 2). The resulting three-item scalehas high to very high reliability accord-ing to the Cronbachs alpha criterion;the assessment of the business idea thus appears to be successful. This is also indicated by the very low internalnonresponse (one and seven cases,respectively). However, some furtherimprovement of the measurement of thisdimension would be valuable. The dis-tribution is skewed positively with anacross-item average score of 4.25 out offive, and a full 26 percent of the samplehas the maximum score (15) on the Nowversion of the scale. This suggests thatsome minor rephrasing of the currentitems and the additions of one or twonew items would help capture the fullrange of variation better than the currentversion does. For a first try, however, thecurrent version performs very well.

Only three items were developed forthe second scale, product, and all threecould be retained. The Cronbachs alphavalues here are extremely high for bothversions of the scale, indicating very highreliability. The measurement of this cor-nerstone is not entirely unproblematic,though. First, internal nonresponse isnonnegligible. For some reason some

respondents appeared to think that thequestions were not suitable to describetheir firm. Second, the distributions ofresponses are bimodal, with overrepre-sentation at both extremes. This suggeststhat the respondents tended to view thisdimension as dichotomous: either theyhad a developed product ready for saleor they did not. Again, then, some furtherimprovement seems possible althoughthis first attempt must be judged as rela-tively successful.

For the third cornerstone, market, sat-isfactory to high Cronbachs alpha valueswere obtained after deletion of one ofthe original items. High reliability thus is attributed to the resulting four-itemscale. The distributions also look gooddespite some positive skewing. Theproblem here is instead the high internalnonresponse. About 20 percent of thesample chose to skip one or severalitems. In order to check whether therewas any difference between manufactur-ing and service firms relative propensityto give valid responses, we computedvalid/missing dummies for the productand market indices and cross-tabulatedthem against service versus manufactur-ing firm. There were no significant dif-ferences and were not even a tendencytoward any industry differences (Chi2 =0.31, p = 0.58, and Chi2 = 0.46, p = 0.50,respectively; d.f. = 1 and N = 106 in bothcases). The reasons for the internal non-response require further investigation.For those who answered, the scale per-forms well.

The next cornerstone, organization, isthe one for which measurement has beenmost successful. The Cronbachs alphavalues are high, the distribution is veryclose to normal, and internal nonre-sponse is modest. From a reliability pointof view, this dimension works almostperfectly as it is and needs no furtherdevelopment. After elimination of oneitem, the five-item competence scale alsohas very satisfactory properties, with lowinternal nonresponse, high Cronbachs


10JO

URN

AL O

F SMA

LL BU

SINESS M

AN

AG

EM

EN

T

Table 2Forced Choice Items for Each Cornerstone of the Business Platform Model

Low (1) High (5)

Business IdeaThe idea about what the firms operations vs. There exists a very clearly specified idea for what the firms

should be is not particularly specified. operations should be.Within the firm there is some lack of vs. Everybody in the firm is completely clear about what ideas to

clarity as to what ideas should be pursued. prioritizeIt is relatively unclear what type of need of vs. It is completely clear what need for what type to customer the

what type of customer the firms idea firms idea can satisfy.might satisfy.

(It is rather difficult to say what is special vs. (It is completely identified what is special and unique about the or unique about the firms idea.) firms idea.)

ProductThere is no developed product. vs. There is at least one well developed product that is entirely ready

for sale.No user has tested the product. vs. The product has been tested by a number of potential users.No reference customer can verify the vs. A number of reference customers exist, who can verify the

usefulness of the product. usefulness of the product.

DAV

IDSSO

N A

ND

KLO

FSTEN

11

Table 2Continued

Low (1) High (5)

MarketThe firm has no limitations as to what vs. The firm turns itself to a very specific customer category.

customers it turns to.It is not possible to say what characterizes vs. There exist a number of criteria, which precisely define the

the firms customers. potential customers that have the highest probability to buy.(What value the firms product can offer the vs. (Customer value for the firms product is completely specified

customers is built on assumptions within based on contacts with customers)the firm.)

The market is worked up mainly through vs. The firm follows a structured strategy for working up the market.random contacts.

The firm sees a large number of customer vs. The firm gives clear priority to certain customer categories over categories all of which are deemed others.equally important to cultivate.

OrganizationAll staff do most types of work. vs. All staff have clearly delimited tasks.There are no specified organizational units. vs. It is possible to clearly describe the firm in an organizational

chart.Members of the organization are recruited vs. Members of the organizations are there because of conscious

on the basis of the founders personal recruitment of specific competencies.network.

The firms operations are adapted to the vs. A disciplined and goal oriented effort towards developing the firm situation and governed by events that is being implemented.come up.

Everybody in the firm have responsibility vs. There is a strict division of authority and responsibilities within and authority within most areas. the firm.

12JO

URN

AL O

F SMA

LL BU

SINESS M

AN

AG

EM

EN

T

Table 2Continued

Low (1) High (5)

CompetenceTo some extent the firm lacks knowledge vs. The firm is very well equipped with knowledge about the market

about the market for its products. for its products.To some extent the firm lacks competence vs. The firm is very well equipped with competence in marketing

in marketing and selling. and selling.To some extent the firm lacks expert vs. The firm is very well equipped with expert knowledge within its

knowledge within its domain. domain.To some extent the firm lacks experience vs. The firm is very well equipped with experience and competence

and competence in the area of leadership. in the area of leadership.The competence the firm has will not cover vs. The competence the firm has will completely cover its needs for

its needs for the future. the future.(Every member of the organization is vs. (A systematic plan for enhancement of competence is

responsible for the enhancement of implemented for every member of the organization.)his/her own competence.)

Drive/MotivationThe founders primary goal with the firm is vs. The founders primary goal with the firm is to amaze the world;

to provide employment for him/herself build a growth company.and perhaps some friends.

The founder regards the firm as one of vs. The founder is completely geared towards a future as business several possible ways of earning his/her owner-manager.living.

Work must not intrude on the leisure of the vs. All time is invested in the firm.people involved.

For all parties concerned their involvement vs. All parties concerned have a very large involvement with the firm.with the firm can be characterized asmodest.

DAV

IDSSO

N A

ND

KLO

FSTEN

13

Table 2Continued

Low (1) High (5)

Customer RelationsThe firm has as yet not sold any product to vs. The firm has a large number of customers who have bought its

a customer. products.It is not likely that any customer will repeat vs. It is very common that the firms customer make repeat

purchase the product. purchases.It is difficult to create sales to new vs. The firm constantly gets loads of new customers.

customers.The customers are sometimes dissatisfied. vs. The customers are always very satisfied.(One person keeps all customer contacts.) vs. (Customer contacts are dispersed among all members of the

organization).

Other RelationsThere exist no relationships with banks or vs. There exist very good and stable relationships with banks and

investors. investors.The firm has a shortage of financial capital. vs. Access to financial capital for the firms operations is no problem

whatsoever.There exist no contacts that can provide vs. The firm has well developed contacts with other actors who give

credibility for the firm in the eyes of the the firm the complementary resource credibility.market.

There exist no contacts that can provide the vs. The firm has well developed contacts with other actors which firm with additional management provides the firm the with additional management competence.competence.


Table 3Scale Construction Results for Each Cornerstone in the

Business Platform Model

Number of Number of Mean S.D. Cronbachs Items Cases Alpha

Business IdeaScale Now 3 113 12.74 2.12 0.79Scale Two Years 3 107 10.77 3.41 0.90Ago

ProductScale Now 3 101 12.30 3.85 0.94Scale Two Years 3 95 10.07 4.83 0.96Ago

MarketScale Now 4 91 15.96 2.80 0.70Scale Two Years 4 88 12.81 3.97 0.84Ago

OrganizationScale Now 5 108 16.63 5.28 0.83Scale Two Years 5 104 13.35 5.22 0.84Ago

CompetenceScale Now 5 111 18.28 3.75 0.79Scale Two Years 5 105 15.26 4.42 0.81Ago

Drive/MotivationScale Now 4 109 15.29 3.38 0.71Scale Two Years 4 105 13.52 3.68 0.72Ago

Customer RelationsScale Now 4 107 15.15 2.77 0.60Scale Two Years 4 104 13.30 3.65 0.73Ago

Other RelationsScale Now 4 112 15.23 3.10 0.63Scale Two Years 4 106 12.91 4.01 0.80Ago

alpha values, and only a mild positiveskewing. Similarly, the results for drive/motivation are also satisfactory, althoughthe alpha values are only slightly above0.70, and the positive skewing is some-what more pronounced.

The results for the final two dimen-sions, customer relations and other relations, are very similar. Both are basedon four items in the final analysis, andboth are somewhat positively skewedwith across-item averages of 3.79 and

3.81 (out of five), respectively. Neitherhas any severe problems with internalnonresponse, but for both dimensionsthe Now version of the scale comes outwith a less than satisfactory alpha.Adding one more appropriate item eachto the assessment of these two corner-stones is recommendable.

Taken together, this initial attempt tocreate a formal operationalization of thebusiness platform model must be judgedas very successful from a reliability pointof view. The instrument seems to workreasonably well as is, although improve-ment would be desirable for some of thecornerstones. Encouraging for the instru-ments validity is also that the means areuniformly higher and that the standarddeviations are uniformly lower for theNow version of the scales relative to thetwo years ago versions. This is the devel-opment that should be expected overtime from a surviving sample of youngfirms. The only more serious problemdetected so far is, arguably, the rathersubstantial internal nonresponse for themarket and product dimensions. A closerlook at this problem may suggest limita-tions for what types of firm these dimen-sions of the model are applicable.Alternatively, it may suggest a more positive solution to the problem. It isconceivable that with face-to-face or telephone interviewing the uncertaintybehind the internal nonresponse couldbe sorted out. As to mail surveys,improved instructions to the respondentmight help.

The results for the control questionthat were asked after the platform itemssuggest that few respondents found the questions difficult to answer. In theNow version, two-thirds of the sampleagreed completely or partly that it waseasy to determine what the answersshould be. For natural reasons relativelyfewer respondents49 percentfoundit easy to give responses for the firms situation two years earlier. The propor-tions that refuted that the questions were

easy stayed at eight and 12 percent,respectively.

Having established reliability, we nowturn to the issue of discriminant validity.For this purpose we ran an exploratoryfactor analysis with the default criterionthat eigenvalues for extracted factorsshould be higher than unity. We used theNow version of each item and includedonly those 32 items that were retainedafter the reliability analysis reported.No further items were dropped in thevarious factor analyses we performed. Asthe number of cases is important for thestability of factor analysis solutions, wechose the pairwise alternative, thus usingall available correlational information inthe data. The initial analysis yielded ninefactors, which accounted for 68 percentof the total variation. As the ninth factorhad an eigenvalue just over one and inorder to better assess the fit between theeight-cornerstone conceptualization andthe factor analysis results, we performeda second run with the extraction of eightfactors as a forced choice, which accountfor 65 percent of the variance. Thevarimax-rotated version of this analysis isdisplayed in Table 4. Note also that forease of interpretation the factors havebeen renumbered, and loadings below|0.30| have been suppressed.

All items intended to measure thesame dimension loaded on the samefactor, and only in two cases out of 32does an item have a higher loading onanother factor than its own. The numberof side-loadings above |0.30| is lowoverall (seven). Recall that we did notassume a priori that the different cor-nerstones of the business platform modelwould be uncorrelated, as they are allpartly a function of time. In spite of thefact that the factor analysis extractsorthogonal (uncorrelated) factors, thecornerstones emerged as distinct in thisanalysis. This clearly suggests that thedifferent item packages measure dif-ferent dimensions. That is, discriminantvalidity appears to be high.


16JO

URN

AL O

F SMA

LL BU

SINESS M

AN

AG

EM

EN

T

Table 4Factor (Principal Components) Analysis Loadings for Cornerstone Items

(Now Data; Pairwise Deletion of Missing Values; n = 92114)

Factor F1: F2: F3: F4: F5: F6: F7: F8:(expl. var) Idea Product Market Organization Competence Drive/ Customer Other

(7.2 (8.9 (6.8 (12.0 (9.6 Motivation Relations RelationsItems percent) percent) percent) percent) percent) (7.5 (5.9 (6.7

percent) percent) percent)

Idea 1 .84Idea 2 .86Idea 3 .75Product 1 .83 .38Product 2 .86Product 3 .92Market 1 .81Market 2 .61Market 4 .34 .63 .31Market 5 .74Org. 1 .81Org. 2 .70Org. 3 .70Org. 4 .68Org. 5 .81

DAV

IDSSO

N A

ND

KLO

FSTEN

17

Table 4Continued




Comp. 1 .77Comp. 2 .57 .31Comp. 3 .66Comp. 4 .72Comp. 5 .72Drive 1 .77Drive 2 .69Drive 3 .67Drive 4 .59 .42Cust. 1 .69Cust. 2 .71Cust. 3 -.30 .53Cust. 4 .57Other 1 .36 .58Other 2 .58Other 3 .75Other 4 .78

Note: Loadings less than .30 have been suppressed. Total explained variance = 65 percent. Displayed variance explained perfactor is after rotation. The displayed solution is a forced eight-factor solution, as nine factors with eigenvalue >1 can be extractedfrom these data (var expl. = 68 percent). The screeplot shows a curve that falls asymptotically with no clear breaks or bends.

When the analysis was reran for thetwo years ago versions of the items, theresults were very similar. In some waysthey were better and in other ways wereworse than those displayed in Table 4.The analysis yields eight factors by thedefault criterion (better), and these eightfactors account for 70 percent of the vari-ance (better). The factor pattern is similarto that displayed in Table 4 and thus isvery clear all dimensions.

However, the number of side-loadingsis higher (9 to 7; worse), and for twoitems the highest loading is on the wrongfactor (same). The results are displayedin Appendix 1.

The factors in the factor analysis canbe regarded as weighed indices of allitems included in the analysis, withgreater weight given to items with highloadings. Summated indices based on thereliability analysis, on the other hand, areinfluenced solely by those items thatwere intended to assess that particulardimension, and these items are givenequal weight in the summation. There-fore, a summated index is not identicalto its corresponding factor. While thefactors are constructed to be uncorre-lated, the indices may overlap. We there-fore supplement the factor analysis witha correlation analysis to give more directevidence on the relative distinctivenessof the cornerstone indices. This analysisis displayed in Table 5.

The results show that the correlationsare positive for the most part. This wasexpected, since all dimensions have incommon that they are in part a functionof time. Only one correlation is negative,and that correlation is not statisticallysignificant. Interestingly, two of thedimensions one might have suspectedwere relatedmarket and customer relationsare not correlated at all in thissample. Most correlations are modest.The fact that the organization dimensionhas a couple of correlations in the .40sdoes not prove it not to be distinctenough. A correlation on 0.44 reflects

that the two factors have less than 20percent of the variance in common. Thisis very far from being identical, and ourconclusion that discriminant validity issatisfactory remains valid.

DiscussionIn all, we would hold that our attempt

to create a formal operationalization ofthe business platform model has beensuccessful. We have established moder-ate to very high reliability and haveestablished high discriminant validity forthe different cornerstone indices. This isan important step toward making thebusiness platform model a quantifiable,holistic, and action-oriented instrumentfor assessing and for assisting the devel-opment of young firms and hence towardincreased and more well-founded use ofthe model.

Much remains to be done before afully satisfactory tool has been devel-oped. As noted above, some revision of the instrument may be needed inorder to improve its technical properties.These technical properties need to betested also for a translated version, sothat the applicability of the instrumentcan be generalized to other countries.We have tested the instrument on technology-based firms, a category for which the platform model originally wasdeveloped. Subsequent research hasindicated broader applicability (Klofsten1998). However, it cannot be ruled out that the type of firm or the average higher education of the CEOsinvestigated made the instrumentperform better than it would have in amore diverse sample of firms. Furtherresearch on large and broad samplesshould use the technically improvedversion of the instrument to assess thestate of firms at various points in timeduring their early development andshould relate these assessments to out-comes. This is needed in order to estab-lish predictive validity of the model asoperationalized with our instrument.


DAV

IDSSO

N A

ND

KLO

FSTEN

19

Table 5Pearson Correlations between the Cornerstone Summated Indices (based on Now data;

n = 87110)

Idea Product Market Organization Competence Drive/Motivation Customer Relations Other Relations

Idea 1.00

Product -.07 1.00

Market .05 .36** 1.00

Org. .13 .44** .44** 1.00

Comp. .13 .22* .13 .31** 1.00

Drive .19* .07 .15 .23* .42** 1.00

Cust. .04 .22* .00 .17 .29** .19 1.00**

Other .11 .13 .16 .07 .19* .10 .25** 1.00

*p < 0.05; **p < 0.01.

Other issues include establishing mini-mum levels for each cornerstone andinvestigating the possible need of out-standing levels on some dimensions, aswell as checking for differences in theseregards by industry or type of firm.

The fact that the model is based on anextensive literature review and qualita-tive research should ascertain someexternal validity. However, further in-depth work is needed in order to deter-mine whether assessments based on thestandardized instrument accord withclinical judgment. Experiments would bethe ideal for evaluating the models andthe instruments suitability not only for prediction but also as a basis for cor-rective action. Such experiments wouldbe difficult to set up in practice, and if possible, conducting them would beethically questionable, as it wouldinvolve refraining from giving advice themodel predicts is essential for the firmssurvival. However, quasi-experimentsshould be possible. That is, all cases inthe study would get proper adviceaccording to the model. The experimen-tal manipulation would be provided bythe firms themselvesthat is, the extentto which they choose to implement theactions suggested by the advice. Evi-dence that those who followed theadvice fared better than those that didnot would be very strong support for thevalidity of the model and its opera-tionalizations. Ultimately, that is the typeof evidence we need in order to applythis tool with great confidence.

Finally, when it comes to the analysisof the firms development processes(earlier as well as later) one will neverescape from the difficulties to make relevant judgments in real time, a situa-tion very different from analysis ofalready written cases (for teaching pur-poses) where one knows in advance theresults of decisions and actions in thefirm. In order to show the difficulties,one can exemplify with some of the cornerstones of the business platform:

How does the entrepreneur know when the market is defined? How doesthe entrepreneur know when there areenough driving forces for growth? Howdoes the entrepreneur know when there is quality in the customer relations?Of course there is no perfect method for handling such philosophical issues,although a most relevant procedure is to ask relevant qualitative questions concerning the eight cornerstones con-tinuously within the firms manage-ment team and then to try to achieveconsensus concerning at what level thefirm is at present. Relevant questions on the market development couldinclude the following: Describe the firmsmethod of developing its marketdoesthe firm develop its market in a struc-tured manner, for example by its ownpersonnel, representatives, or built-incustomers? What criteria are there for the choice of market, segment, or niche?Concerning prime mover and commit-ment, one can ask the following:Why was the firm started? Who is themotor in the firm? What are the differ-ences in opinion that exists over how thefirms activities should be run? Hopefully,these questions can answer how far thefirm has progressed in its developmentof the business platform. With knowl-edge of the contributing factors, or cor-nerstones, which are necessary to buildup a business platform, it is possible todetermine if one has succeeded in attain-ing one, and if not, how far there is leftto go.

ReferencesAdizes, I. (1988). Corporate Lifecycles:

How and Why Corporations Grow and Die and What to Do about It,Englewood Cliffs, NJ: Prentice Hall.

Aijo, T.S. (1996). The Theoretical andPhilosophical Underpinnings of Mar-keting, European Journal of Market-ing (30)2, 818.

Autio, E. (1995). Four Types of Innova-tors: A Conceptual and Empirical


Study of New Technology-Based Com-panies and Innovators, Entrepreneur-ship and Regional Development 7(3),233248.

Autio, E. (1997). New Technology-BasedFirms in the Innovation Networks,in Technology, Innovation, and EnterpriseTthe European Experi-ence. Ed. D. Jones-Evans and M.Klofsten. London, England: MacmillanPublishers, 209235.

Barney, J.B. (1991). Firm Resources andSustained Competitive Advantage,Journal of Management 17(1), 99120.

Barney, J.B. (1997). Gaining and Sus-taining Competitive Advantage. MenloPark, Calif.: Addison Wesley.

Bellu, R., P. Davidsson, and C. Goldfarb(1990). Toward a Theory of Entre-preneurial Behavior: Empirical Evi-dence from Israel, Italy, and Sweden,Entrepreneurship and Regional Devel-opment 2(2), 195209.

Bird, B. (1995) Toward a Theory of Entrepreneurial Competency, Ad-vances in Entrepreneurship, FirmEmergence and Growth 2, 5172.

Birley, S., and D. Norburn. (1985). Smallversus Large Companies: The Entre-preneurial Conundrum, The Journalof Business Strategy 6(1), 8187.

Bjerke, B. (1989). Att Skapa Nya Affrer[The Creation of New Business]. Lund,Sweden: Studentlitteratur.

Bollinger, L., K. Hope, and J. Utterback(1981). Technology and IndustrialInnovation in Sweden. A Review of Literature and Hypothesis on NewTechnology-Based Firms. Stockholm,Sweden: STU.

Brockhaus, R.H., Sr. (1980). Risk-TakingPropensity of Entrepreneurs, Aca-demy of Management Journal 23(3),509520.

Brown, T., P. Davidsson, and J. Wiklund(2001). An Operationalization ofStevensons Conceptualization ofEntrepreneurship as Opportunity-Based Firm Behavior. Strategic Man-agement Journal 22(10), 953968.

Bruno, A.V., and T.T. Tyebjee (1984).The Entrepreneurs Search forCapital, in Frontiers of Entrepreneur-ship Research. Ed. J.A. Hornaday, F.Tarpley, J.A. Timmons, and K.H.Vesper. Wellesley, Mass.: BabsonCollege, 1831.

Chandler, G.N., and S.H. Hanks (1994).Market Attractiveness, Resource-Based Capabilities, Venture Strategies,and Venture Performance, Journal ofBusiness Venturing 9(4), 331347.

Converse, J. M., and S. Presser (1986).Survey Questions: Handcrafting theStandardized Questionnaire. SageUniversity Paper series on Quantita-tive Applications in the Social Sci-ences, series no. 07063. Beverly Hills,Calif.: Sage Publications.

Cooper, A.C. (1981). Strategic Manage-ment: New Ventures and Small Busi-ness, Long Range Planning 14(5),3945.

Cooper, A.C., F.J. Gimeno-Gascon, andC.Y. Woo (1994). Initial Human andFinancial Capital as Predictors of New Venture Performance, Journal of Business Venturing 9(5), 371395.

Cooper, A.C., and A.V. Bruno (1977).Success among High-TechnologyFirms, Business Horizons 20(2),1622.

Cooper, A.C., A.C. Fotta, and C.Y. Woo(1995). Entrepreneurial InformationSearch, Journal of Business Ventur-ing 10(2), 107120.

Dahlquist, J., P. Davidsson, and J.Wiklund (2000). Initial Conditions as Predictors of New Venture Performance: A Replication andExtension of the Cooper et al. Study,Enterprise and Innovation Manage-ment Studies 1(1), 117.

Davidsson, P. (1989a). Continued Entrepreneurship and Small FirmGrowth. Stockholm, Sweden: Stock-holm School of Economics.

Davidsson, P. (1989b). Entrepreneur-shipand after? A Study of Growth Willingness in Small Firms, Journal


of Business Venturing 4(3), 211226.

Davidsson, P. (1989c). Need forAchievement and EntrepreneurialActivity in Small Firms, in Under-standing Economic Behavior. Ed. K.G.Grunert, and F. lander. Dordrecht,The Netherlands: Kluwer AcademicPublishers, 4764.

Davidsson, P. (1991). Continued Entre-preneurship: Ability, Need, andOpportunity as Determinants of SmallFirm Growth, Journal of BusinessVenturing 6(6), 405429.

Davidsson, P. (1995a). Culture, Structure,and Regional Levels of Entrepreneur-ship, Entrepreneurship and RegionalDevelopment 7(1), 4162.

Davidsson, P. (1995b). Determinants ofEntrepreneurial Intentions. Paper presented at the RENT IX Workshop,Piacenza, Italy, Nov. 2324, 1995.

Davidsson, P., and J. Wiklund (1997).Values, Beliefs, and Regional Varia-tions in New Firm Formation Rates,Journal of Economic Psychology18(2/3), 179199.

Dyer, B., A.K. Gupta, and D. Wilemon(1999). What First-to-Market Com-panies Do Differently, Research Tech-nology Management 42(2), 1521.

Freeman, J., G.L. Carrol, and M.T.Hannan, (1983). The Liability ofNewness: Age Dependence in Organi-zational Death Rates, American Soci-ological Review 48, 692710.

Gartner, W.B. (1985). A ConceptualFramework for Describing the Phe-nomenon of New Venture Creation,Academy of Management Review10(5), 696706.

Gibb, A., and M. Scott (1986). Under-standing Small Firms Growth, inSmall Business Growth and Develop-ment. Ed. M. Scott, A. Gibb, J. Lewis,and T. Faulkner. London, England:Gower, 81104.

Gimeno, J., T.B. Folta., A.C. Cooper, andC.Y. Woo (1997). Survival of the

Fittest? Entrepreneurial HumanCapital and the Persistence of Under-performing Firms, AdministrativeScience Quarterly 42(4), 750783.

Greiner, L.E. (1972). Evolution and Revolution as Organizations Grow, Harvard Business Review50(Jan/Feb), 3746.

Hall, G. (1995). Surviving and Prosperit-ing in the Small Firm Sector. London.England: Routledge.

Hall, G., and G. Adams (1994). Influ-ences on Growth of SMEs, Entrepre-neurship and Regional Development6(2), 84104.

Hills, G.E. (1994). Market Opportunitiesand Marketing, in The Portable MBA inEntrepreneurship. Ed. W.D. Bygrave.Toronto: John Wiley & Sons, 67112.

Heydebreck, P. (1997). TechnologicalInterweavement: A Means for NewTechnology-Based Firms to AchieveInnovation Success, in Technology,Innovation and EnterpriseThe European Experience. Ed. D. Jones-Evans and M. Klofsten. London,England: Macmillan Publishers,274302.

Heydebreck, P., M. Klofsten, and J. Maier(2000). Innovation Support for NewTechnology-Based Firms: The SwedishTeknopol Approach, R&D Manage-ment 30(1), 89100.

Hisrich, R.D., and M.P. Peters (1992).Entrepreneurship: Starting, Develop-ing, and Managing a New Enterprise.Homewood, Illinois: Irwin.

Jones-Evans, D., and M. Klofsten (1997). Technology, Innovation, andEnterpriseThe European Experi-ence. London, England: MacmillanPublishers.

Kazanjian, R.K. (1988). Relation of Dom-inant Problems to Stages of Growth in Technology-Based New Ventures,Academy of Management Journal31(2), 257279.

Kazanjian, R.K., and R. Drazin (1990).A Stage-Contingent Model of Design


and Growth for Technology-BasedNew Ventures, Journal of BusinessVenturing 5(3), 137150.

Kimberly, J.R., and R.H. Miles. (1980).The Organizational Life Cycle. SanFrancisco, Calif.: Jossey-Bass.

Klofsten, M. (1992). Tidiga Utveckling-sprocesser i Teknikbaserade Fretag[Early Development Processes in Tech-nology-Based Firms], Linkping Uni-versity, Department of Managementand Economics, Linkping Studies inManagement and Economics, No. 24,Ph.D dissertation.

Klofsten, M. (1994). Technology-BasedFirms: Critical Aspects of Their EarlyDevelopment, Journal of Enterpris-ing Culture 2(1), 535557.

Klofsten, M. (1997). Management of the Early Development Process in Technology-Based Firms, in Technol-ogy, Innovation, and EnterpriseTheEuropean Experience. Ed. D. Jones-Evans and M. Klofsten. London,England: Macmillan Publishers, 148178.

Klofsten, M. (1998). The Business Plat-form: Entrepreneurship and Manage-ment in the Early Stages of a FirmsDevelopment. Luxembourg: EuropeanCommission, TII.

Knight, R.M. (1986). Product Innovationby Smaller, High-Technology Firms inCanada, Journal of Product Innova-tion Management 3, 195203.

Kohen, P.A., and P. Kohli (1998).Idea Generation: Who Comes Upwith the Most Profitable Products?in Frontiers of EntrepreneurshipResearch. Ed. P.D. Reynolds, W.D.Bygrave, N. M. Carter, S. Manigart,C.M. Mason, G.D. Meyer, and D.G.Shaver. Wellesley, Mass.: BabsonCollege, 4445.

Krieger, E. (1998). Trust and Man-agement as Applied to InnovativeSmall Companies, in New Technology-Based Firms in the 1990s, IV. Ed. W.During and R. Oakey. London,

England: Paul Chapman Publishing,125139.

Littunen, H., E. Storhammar, and T.Nenonen (1998). The Survival ofFirms over the First Three Years andthe Local Environment, Entrepre-neurship and Regional Development10(3), 189202.

McClelland, D. (1961). The AchievingSociety, 1st ed. Princeton, NJ: van Nostrand.

Mayer, K.B., and S. Goldstein (1961).The First Two Years: Problems of Small Firm Growth and Survival.Washington, D.C: Small BusinessAdministration.

Miller, D., and P.H. Friesen (1984). ALongitudinal Study of the CorporateLife Cycle, Management Science30(10), 11611183.

Mintzberg, H. (1973). Strategy Making inThree Modes, California Manage-ment Review, XVI(2), 4453.

Morris, M.H. (1998). EntrepreneurialIntensity: Sustainable Advantages forIndividuals, Organizations, and Soci-eties. London, England: QuorumBooks.

Naffziger, D.W., J.S. Hornsby, and D.F.Kuratko (1994). A Proposed ResearchModel of Entrepreneurial Motivation,Entrepreneurship Theory and Practice18(3), 2346.

Nunnally, J. (1967). Psychometric Theory.New York, NY: McGraw-Hill.

Nunnally, J.C., and I.H. Bernstein (1994).Psychometric Theory, 3rd ed. NewYork, NY: McGraw-Hill.

Oorman, C. (1997). Success Strategies inHigh-Growth SMEs, in Technology,Innovation, and EnterpriseTheEuropean Experience. Ed. D. Jones-Evans and M. Klofsten. London,England: Macmillan Publishers, 179208.

Olofsson, C., and C. Wahlbin (1984).Technology-Based New Venturesfrom Technical Universities: ASwedish Case, in Frontiers of Entre-


preneurship Research. Ed. J.A. Horna-day, F. Tarpley, J.A. Timmons, and K.H.Vesper. Wellesley, Mass.: BabsonCollege, 192211.

Penrose, E. (1959). The Theory of theGrowth of the Firm. Oxford, England:Oxford University Press.

Prenkert, F. (1998). Notes On How Some Swedish SMEs Do Businessthrough Patterns of Interaction, inMarketing Behaviour and Capabilitiesin Some Swedish SMEs. Ed. C.Hultman, C. Gunnarsson, and F.Prenkert. Stockholm: FSF, 4580.

Roberts, E.B., and H.A. Wainer (1968). New Enterprises on Route128, Science Journal 4(Dec.), 7884.

Shaver, K., and L. Scott (1991). Person,Process, and Choice: The Psychologyof New Venture Creation, Entrepre-neurship Theory and Practice 16(2),2346.

Sheperd, D.A., and M. Shanley (1998).New Venture Strategy: Timing, Envi-ronmental Uncertainty, and Perfor-mance. London, England: SAGEPublications.

Roberts, E.B. (1991). Entrepreneurs inHigh Technology: Lessons from MITand beyond. Oxford, England: OxfordUniversity Press.

Rothwell, R. (1992). Successful Indus-trial Innovation: Critical Factors forthe 1990s, R&D Management 22(3),221239.

Smith, N.R. (1967). The Entrepreneur and His Firm: The Relationshipbetween Type of Man and Type ofCompany. East Lansing: MichiganState University.

Susbauer, J.C. (1967). The Science Entre-preneur, Industrial Research 8(2),2330.

Timmons, J.A. (1994). OpportunityRecognition: The Search for HigherPotential Ventures, in The PortableMBA in Entrepreneurship. Ed. W.D.Bygrave. Toronto: John Wiley & Sons,2654.

Utterback, J.M., and G. Reitberger (1982).Technology and Industrial Innova-tion in SwedenA Study of New Technology-Based Firms. Cambridge,Mass.: MIT-CPA (CPA/82061A).

Van de Ven, A.H., R. Hudson, and D.M.Schroeder (1984). Designing NewBusiness Startups: Entrepreneurial,Organizational, and Ecological Con-siderations, Journal of Management10(1), 87107.

Walsh, S.T., and B.A. Kirchhoff (1998).Can Entrepreneurial Core Competen-cies and Capabilities Overcome PoorStrategies? in Frontiers of Entrepre-neurship Research. Ed. P.D. Reynolds,W.D. Bygrave, N.M. Carter, S. Mani-gart, C.M. Mason, G.D. Meyer, andD.G. Shaver. Wellesley, Mass.: BabsonCollege, 269282.

Weinstein, A. (1994). Market Definitionin Technology-Based Industry: A Comparative Study of Small versusNon-Small Companies, Journal ofSmall Business Management 32(4),2836.

Westhead, P., and S. Birley (1993).Employment Growth in New Inde-pendent Owner-Managed High-Technology Firms in Great Britain,Working paper, University of Warwick.

Wiklund, J. (1998). Small Firm Growthand Performance: Entrepreneurshipand beyond. Jnkping: JnkpingInternational Business School, Ph.Ddissertation.

Wiklund, J., P. Davidsson., F. Delmar, andM. Aronsson (1997). Expected Conse-quences of Growth and Their Effecton Growth Willingness in DifferentSamples of Small Firms, in Frontiersof Entrepreneurship Research. Ed. P.D.Reynolds, W.D. Bygrave, N.M. Carter,P. Davidsson, W.B. Gartnier, C.M.Mason, and P.P. McDougall. Wellesley,Mass.: Babson College, 116.

Zeithaml, V.A., and M.J. Bitner (1996).Services Marketing. New York, NY:McGraw Hill.


DAV

IDSSO

N A

ND

KLO

FSTEN

25

AppendixFactor (Principal Components) Analysis Loadings for Cornerstone Items (Two Years Ago

Data; Pairwise Deletion of Missing Values; n = 88109)




Idea 1 .90Idea 2 .92Idea 3 .86Product 1 .81Product 2 .88Product 3 .85Market 1 .84Market 2 .31 .69Market 4 .37 .56Market 5 .75Org. 1 .82Org. 2 .38 .60Org. 3 .74Org. 4 .68Org. 5 .79

26JO

URN

AL O

F SMA

LL BU

SINESS M

AN

AG

EM

EN

T

AppendixContinued




Comp. 1 .32 .75Comp. 2 .36 .56Comp. 3 .61Comp. 4 .78Comp. 5 .66Drive 1 .70Drive 2 .67 .32Drive 3 .84Drive 4 .63 .31Cust. 1 .69Cust. 2 .71Cust. 3 .62Cust. 4 .37 .44 .35Other 1 .73Other 2 .75Other 3 .80Other 4 .66

Note: Loadings less than .30 have been suppressed. Total explained variance is 70 percent. Displayed variance explained perfactor is after rotation. The displayed solution is the default solution, as only eight factors with eigenvalue >1 can be extractedfrom these data. The screeplot shows a curve with clear bends at two and eight factors.

Davidsson 2003 - The Business Platform: Developing an Instrument to Gauge and to Assist the...

Documents

Transcript of Davidsson 2003 - The Business Platform: Developing an Instrument to Gauge and to Assist the...