THE EFFECTS OF ANXIETY, HOSTILITY,

AND DEPRESSION ON RESPONSES

TO THE ROTTER INCOMPLETE

SENTENCES BLANK

APPROVED:

/)

cf h L, J xA-z* /vy-b\H/ / 'l y\ 1 J xj-tt /y.-y -v'~f~7g:^ Major Professor

1/

Minor Professor ^

^ • • •

Chairman of the Department of Psychology

De'atf of the Graduate School

THE EFFECTS OF ANXIETY, HOSTILITY, AND DEPRESSION

ON RESPONSES TO THE ROTTER INCOMPLETE

SENTENCES BLANK

THESIS

Presented to the Graduate Council of the

North Texas State University in Partial

Fulfillment of the Requirements

For the Degree of

MASTER OF SCIENCE

By

Margaret Ann Boutte, B.A.

Denton, Texas

August, 1970

TABLE OF CONTENTS

Page

LIST OF TABLES iv

Chapter

I . INTRODUCTION 1

I I . REVIEW OF THE LITERATURE 14 His tory of the Sentence Completion Method L i t e r a t u r e p e r t i n e n t to the Ro t t e r Incomplete

Sentences Blank L i t e r a t u r e p e r t i n e n t to the Mul t ip le A f f e c t

Ad jec t ive Check L i s t

I I I . METHOD 40

Descr ip t ion of Sub jec t s and Apparatus Procedure

IV. RESULTS 45

V. DISCUSSION 54

The E f f e c t s of an Examination Threat on the Mul t ip le A f f e c t Ad jec t ive Check L i s t

The E f f e c t s of Anxiety and H o s t i l i t y on Responses to the Ro t t e r Incomplete Sentences Blank

Impl ica t ions of P resen t Study

VI. SUMMARY AND CONCLUSIONS 62

APPENDIX 64

BIBLIOGRAPHY 66

i n

LIST OF TABLES

Table Paqe

I . Means and Standard Deviations for the F i rs t Three Administrations of the Mult iple Affect Adjective Check L is t and the Rotter Incomplete Sentences Blank 46

I I . Mult iple Affect Adjective Check L is t and the Rotter Incomplete Sentences Blank Baseline t Tests 48

I I I . Mean Mult iple Affect Adjective Check L is t Scores of Experimental and Control Groups Before and After the Examination Threat 49

IV. Mean Rotter Incomplete Sentences Blank Scores of Experimental and Control Groups Before and After the Examination Threat 51

IV

CHAPTER I

INTRODUCTION

In a c l in ica l setting a psychologist attempts to learn as much

about his c l ient as he can. He seeks information from the c l ients '

fr iends, family, employer, work associates, and the c l ient himself.

The obtained information is organized in such a manner that a c l ients '

roles, his functioning sk i l l s and ab i l i t i e s , his emotional characteristics

and control, his intrapsychic conf l icts and defense techniques, his

self-concept, and values readily appear to the trained mind of the

psychologist (9). In essence one may say that a psychologist attempts

to get a global picture of his c l ients, one which includes both

strengths and weaknesses.

Tests are one of the tools which a psychologist may use in the

formulation of this global picture. The interpretations obtained from a

test are influenced by the test 's va l id i t y , i t s r e l i a b i l i t y , i t s

fu l f i l lment of purpose and objectives as stated in the test manual, and

in addition in the case of projective techniques — the s k i l l and

experience of the psychologist doing the interpretation. Consequently

the more information one has concerning a particular tes t , the easier

i t is for one to make a decision as to when and under what circumstances

to use a test and how much emphasis to place on the results of that

part icular test .

1

2

Sentence Completion Blanks are one method of bbtaining a global

picture of the c l ient . Generally these tests have been cr i t ic ized as

being too subjective. As a resul t , the r e l i a b i l i t y of such tests

fluctuate with the training and experience of the psychologist. In an

attempt to rec t i fy the si tuat ion, Rotter and Rafferty devised a sentence

completion blank which u t i l i zed an objective scoring system (5).

However the research l i terature has since indicated that Rotter and

Rafferty's objective scoring system either was not as objective as i t

appeared and/or the test responses varied with the amount of anxiety,

hos t i l i t y , and depression f e l t by the c l ient taking this test (1,8).

Hence this experiment seeks to determine the effects of such variables.

The present study is an attempt to determine the effect of

anxiety, hos t i l i t y , and depression on responses to the Rotter Incomplete

Sentences Blank when the Incomplete Sentences Blank is scored according

to the manual instructions. I f the results of this study indicate that

the adjustment score obtained from the Rotter Incomplete Sentences Blank

fluctuates s igni f icant ly with the c l ients ' manifested anxiety, hos t i l i t y ,

or depression, then the implication is that the psychologist should

take this fact into account when using the Rotter Incomplete Sentences

Blank as a diagnostic instrument especially in the i n i t i a l sessions with

the c l ient . On the other hand, i f the Rotter Incomplete Sentences Blank

is not influenced by anxiety, hos t i l i t y , or depression when scored

according to the objective scoring system provided by Rotter and

Rafferty in their manual, then the psychologist may rely on the results

of the test to a greater extent in his f i r s t impression of the c l i en t ,

his personality, and his problems.

3

A promising method of measuring anxiety, hos t i l i t y , and depression

is the Today form of the Multiple Affect Adjective Check L is t . The

Multiple Affect Adjective Check List " . . . was designed to f i l l the need

for a self-administered test which would provide val id measures of three

of the c l in ica l l y relevant negative affects: anxiety, depression, and

hos t i l i t y , " (13, p. 3). The Multiple Affect Adjective Check List is

easily administered, requiring no more than f ive minutes, and has i t s

own correction formula for checking response set. According to Zuckerman

and Lubin (13) the Multiple Affect Adjective Check List is especially

useful in stress experiments requiring before and after measures of

af fect , and for measures of natural stress such as operations and exam-

inations.

The Multiple Affect Adjective Check List has two forms, the General

and the Today form; hence, the Multiple Affect Adjective Check List has

a di f ferent r e l i a b i l i t y pattern depending on whether a "general" time

set or a "today" time set is used. The Multiple Affect Adjective Check

List (General form) and the Multiple Affect Adjective Check List (Today

form) are identical except for instructions. Both tests show a high

internal r e l i a b i l i t y of .72 and .85 respectively. The difference between

the two time sets occurs in test-retest r e l i a b i l i t y where the "general"

time set yields a r e l i a b i l i t y of .68 and the "today" time set's r e l i -

ab i l i t y is .31. The lower r e l i a b i l i t y of the "today" time set is expected

for the following reason: A test which attempts to measure affect should

not be s ta t i s t i ca l l y rel iable from day to day i f i t is t ru ly sensitive to

individual differences in mood (14).

The Multiple Affect Adjective Check List was standardized on 534

4.

subjects obtained from the following populations: (a) seventy-five

students obtained from an introductory psychology class at Adelphi

College in New York, (b) 133 psychiatric patients from Larue D. Carter

Memorial Hospital in Indiana, (c) forty-three patients from Central

I s l ip Hospital in New York, (d) eighty-three patients from the V.A.

Hospital in Washington, D. C., and the St. Cloud V.A. Hospital in

Minnesota, and 200 job applicants drawn from a population of 1200

job applicants at the Indiana Medical Center personnel o f f ice. These

la t ter 200 subjects were s t ra t i f ied by age, sex, and education to

match the census distr ibut ion of these variables. The mean ages for

a l l subjects ranged from eighteen to forty-two with a mean calculated

on each sample of males and females.

The means and standard deviations for the three variables which

the test seeks to measure — anxiety, depression, and hos t i l i t y — are

reported in the test manual for each group and sex (13). Only those

means and standard deviations pertinent to the college students'

population are presented here. Thus the sex variable which was

insignif icant was disregarded. Means of 6.6, 14.1, 7.7, and standard

deviations of 3.3, 7.1, 3.9, were found respectively for the variables

of anxiety, depression, and hos t i l i t y .

Since the origination of the test , numerous studies have been

conducted concerning i t s va l id i ty . Zuckerman (11) gave the scale to

a class of thir ty-two college students on consecutive class meetings a

week apart. In each case he found that the anxiety scale showed a

signif icant increase, relat ive to the baseline, when given on days

prior to an examination. A s igni f icant ly greater r ise occurred for

for students who obtained low grades on the exams than those who

obtained high grades. Similar results were obtained by Winter,

Ferreira, and Ranson (10).

Zuckerman and Biase (12) replicated the study and in addition

obtained worry ratings af ter the Mult iple Affect Adjective Check L is t

scale was given and before the exam was taken. These researchers

found that students who rated themselves a§ being worried about the

examination showed s ign i f i can t l y greater increases than students who

rated themselves as less worried about the examination.

Zuckerman, Lubin, Vogel, and Valerius (14) administered the

Mult iple Affect Adjective Check L is t on three baseline days, a week

apart. On the fourth week the experimentor entered with an unannounced

examination threat which was intended to e l i c i t h o s t i l i t y as well as

anxiety. The examination was not actual ly given at that time, but

la ter at a previously scheduled time. Fals i f ied low grades were given

pr ior to taking the Mult iple Affect Adjective Check L is t during the

f ina l week. The results showed that on the real examination day

Anxiety scale scores were the only ones s ign i f i can t l y elevated above

the baseline level . The Anxiety, Hos t i l i t y , and Depression scales were

s ign i f i can t l y affected by the examination threat and the low grades

stress. The Hos t i l i t y scale showed a s ign i f i can t ly greater response

to the examination threat than the Depression scale.

Lieberman (4) found that a l l three Mult iple Affect Adjective Check

L is t scores were s ign i f i can t l y elevated on days when students were

expecting an examination., In addition a comparison between good and

poor students disclosed that poor students scored higher on the General

6

form of the Multiple Affect Adjective Check List and the Today's

Host i l i ty scale on non-test days. Consequently, on the basis of

Lieberman (4) and other studies previously mentioned, one may conclude

that the Multiple Affect Adjective Check List does measure what i t

purports to measure; that i s , the Multiple Affect Adjective Check

List is an adequate measure of anxiety, hos t i l i t y , and depression.

The Rotter Incomplete Sentences Blank is one of the best known and

most widely used tests. Like some of the other sentence completion

methods used in studying personality, the Rotter Incomplete Sentences

Blank is a semi-structured projective technique in which the subject

completes part ial sentences. Rotter and Rafferty (5) assumed that not

only are the subjects' own wishes, desires, fears, and attitudes

reflected in the resulting sentences, but in addition tendencies to

block and to twist the meaning of the stimulus also occur.

The contemporary Rotter Incomplete Sentences Blank originated from

a revised form used by Rotter and Willerman (6) in the Army. This

contemporary form in turn originated from a revision of blanks used by

Hutt (3), Shor (7), and Holzberg, Teicher, and Taylor (2) in the

diagnosis, treatment, and disposition of veterans.

According to Rotter and Rafferty (5), the purpose of the Rotter

Incomplete Sentences Blank is " . . . an attempt to standardize the

sentence completion method for use at the college level . . . " (p. 3)

and also provide a gross screening instrument whose total score would

give an index of maladjustment. I t is not surprising, therefore, that

the following objectives were kept in mind throughout the test 's

development: (1) to provide an objective technique useful for screening

7

and experimental purposes, be economical to administer and score, and

have some of the advantages of projective methods, (2) the obtaining of

specific diagnostic information for treatment purposes. This does not

mean the test was constructed to measure or describe "whole personality"

or expose the "deep layers" of personality of the fundamental structure.

I t was designed as a time saving device for the c l in ic ian, allowing him

to structure his f i r s t interview advantageously before he spends a

great deal of time with a patient.

The Rotter Incomplete Sentences Blank is available in three forms:

College, High School, and Adult. Ages for the various forms are not

numerically specified since the cr i ter ion is determined by group

membership (5).

The Rotter Incomplete Sentences Blank was standardized on 299

entering freshmen (85 females and 214 males) at Ohio State University.

The mean scores were 127.4 for the females and 127.5 for the males with

a standard deviation of 14.4 and 14.5 respectively. Rotter and Rafferty

(5) feel that this sample was typical of the freshman class since i t

compared favorably with the median percentile ranks on the Ohio State

Psychological Examination, showing a difference of approximately two

percentile points.

Validation data on the Rotter Incomplete Sentences Blank at the

college level was collected on 82 females and 125 male students from

Ohio State University who were classif ied as either adjusted or

maladjusted (5). The data was obtained from the two sexes separately

since the scoring manuals are d i f ferent .

The subjects were segregated by sex and each sex divided into two

8

groups. Group 1 for both sexes consisted of subjects c lass i f ied as

adjusted or maladjusted in a forced c lass i f ica t ion category by thei r

instructors . The Group 2 females consisted of ten students who had

previously been classif ied as adjusted or maladjusted by advance

student clinicians while the Group 2 males consisted of for ty-s ix

students who were considered maladjusted ei ther by se l f - r e f e r r a l s to

the psychological c l in ic for treatment or referred by vocational

advisors to the Occupational Service for personal counseling.

The data for Group 1 females showed considerable overlap between

those individuals rated "adjusted" and those rated "maladjusted."

This was not surprising since the c lass i f ica t ion by the instructors was

probably f a l l i b l e . Despite the overlap, however, a cutting score of

135 on the Incomplete Sentences Blank correctly identif ied 78% of the

adjusted cases and 59% of the maladjusted cases. This yielded a

biserial correlation coeff icient of .50.

Similar results occurred with Group 1 males. However, a cutting

score of 135 correctly identif ied 52% of the maladjusted cases, and 89%

of the adjusted cases thereby yielding a correlation coeff ic ient of

.62 (5).

A corrected spl i t -hal f r e l i ab i l i t y of .84 (based on 124 male

college students) and .83 (seventy-one female college students) was

obtained by dividing the Rotter Incomplete Sentences Blank items into

halves of approximate equal d i f f i cu l ty (5).

The t e s t - r e tes t r e l i ab i l i t y for college students ranged from

.38 to .54. These coeff ic ients indicate, according to Churchill and

Crandall (1), that no great time elapse between administration and

9

measurement of experimental variables should occur. Since all

correlations were s ignif icant ly d i f ferent from zero beyond the .01

level of confidence, Churchill and Crandall f e l t this indicated that

the Rotter Incomplete Sentences Blank measures more than momentary moods

or reactive s ta tes . However, when t e s t - r e t e s t r e l i ab i l i t y of the Rotter

Incomplete Sentences Blank performance in the group of mothers was cal-

culated a correlation of .70 was obtained. (The group of mothers con-

sisted of married adult women between the ages of 35-45 who had at leas t

one child) . Although the authors could not s ta te an exact reason, they

f e l t there were two l ikely explanations for the discrepancy between the

t e s t - r e t e s t r e l i a b i l i t y of the two groups: F i r s t , the t es t and re tes t

si tuations for the mothers were exactly the same; whereas, the college

students wrote their original Rotter Incomplete Sentences Blank's as a

part of a required series of placement t es t s given when they arrived on

campus while their participation in the re tes t administration was vol-

untary. Second, the possibi l i ty of a difference in the environments of

the two samples exis ts . I t can be assumed that most of the mothers'

lives were relat ively s table , while the college students were moving

from home environments to college and job environments which entailed

many new experiences and adjustments.

I t seems plausible that there is a third possibi l i ty which one

may consider. I t is possible that the required series of placement

tes t s given to the college students when they arrived on compus and the

turmoil of orientation served to create for these college students an

emotional s ta te which differed from their normal ones. If one accepts

this third poss ib i l i ty , then i t seems possible that the Rotter

10

Incomplete Sentences Blank may be influenced by momentary anxiety,

hos t i l i t y , or depression. Therefore the purpose of this investigation

is to determine exactly what effect momentary anxiety, hos t i l i t y , or

depression has on responses to the Rotter Incomplete Sentences Blank

i terns.

A Statement of the Problem

In order to determine the effect of momentary anxiety, hos t i l i t y ,

or depression, a baseline is needed. Consequently i t seems that i f

the Multiple Affect Adjective Check List and the Rotter Incomplete

Sentences Blank were given two or three times a week apart and then

averaged, a baseline could be determined. This Multiple Affect Adjective

Check List baseline would indicate how a subject generally f e l t with res-

pect to anxiety, hos t i l i t y , and depression. The Rotter Incomplete Sentences

Blank, on the other hand, would give a stable indication of a subjects

total adjustment score.

On the fourth week i f a treatment is administered in the form of

an examination threat, the Multiple Affect Adjective Check List should

adequately ref lect this signif icant difference between the baseline and

the examination threat in the experimental group. However, the control

group which does not receive an examination threat should not ref lect

any signif icant differences between the mean of the baseline and the

mean of the last administration of the Multiple Affect Adjective Check

List and the Rotter Incomplete Sentences Blank.

The Purpose of the Study

The primary purpose of this study is to determine the effect of

n

anxiety, hos t i l i t y , and depression on responses to the Rotter

Incomplete Sentences Blank.

Hypotheses

Based on the previous research concerning the effectiveness of

the Multiple Affect Adjective Check List in measuring anxiety, hos t i l i t y ,

and depression and the var iab i l i t y of the total adjustment score

obtained from the Rotter Incomplete Sentences Blank in the Churchill and

Crandall study, the following are hypothesized for the present study:

Hypothesis 1. In the experimental group on the examination threat

day a l l three seales-r—-anxiety, hos t i l i t y , and depression—of the

Multiple Affect Adjective Check List w i l l be s igni f icant ly higher than

their base l ines.

a. The greatest increase w i l l be on the hos t i l i t y scale.

b. The increase on the hos t i l i t y scale w i l l be s igni f icant ly

greater than the increase on the depression scale, but not

s igni f icant ly greater than the increase on the anxiety scale.

Hypothesis 2. The control group w i l l not have signif icant changes

between the means of i t s baseline data and the last administration of

the Multiple Affect Adjective Check L is t .

Hypothesis 3. In the experimental group, the total adjustment

score of the Rotter Incomplete Sentences Blank on the examination

threat day w i l l not be s igni f icant ly di f ferent than the total adjustment

score of i t s baseline.

Hypothesis 4. In the control group, the total adjustment score of

the Rotter Incomplete Sentences Blank on the examination threat day w i l l

not be s igni f icant ly di f ferent than the total adjustment score of i t s

baseline.

CHAPTER BIBLIOGRAPHY

1. Churchil l, Ruth and V. J. Crandall, "The Rel iab i l i ty and Val idi ty of the Rotter Incomplete Sentences Test," Journal of Consulting Psychology, XIX (October, 1955), 345-350.

2. Holzberg, J . , A. Teicher, and J. L. Taylor, "Contributions of Clinical Psychology to Mi l i tary Neuropsychiatry in an Army Psychiatric Hospital," Journal of Clinical Psychology, I I I (January, 1947), 84-95.

3. Hutt, M. L. , "The Use of Projective Methods of Personality Measurements in Army Medical Instal lat ions," Journal of Clinical Psychology, I (Apr i l , 1945), 134-140.

4. Lieberman, L. R., "Effects Upon Anxiety, Depression, and Host i l i ty of Postponement of an Examination," Perceptual and Motor Sk i l l s , XXIII (December, 1966), 1051-1054.

5. Rotter, J. B. and Janet E. Rafferty, Manual: The Rotter Incomplete Sentences Blank, New York: Psychological Corporation, 1950.

6. Rotter, J. B..and B. Willerman, "The Incomplete Sentences Test," Journal of Consulting Psychology, XI (January, 1947), 43-48.

7. Shor, J . , "Report on a Verbal Projective Technique," Journal of Clinical Psychology, I I (July, 1946), 279-282.

8.Stephens, Mark W., "The Incomplete Sentences Blank: Sources of Variance in Retest Re l iab i l i t y , " Journal of Clinical Psychology, XVI (July, 1960), 331-333.

9. Sundberg, Norman D„ and Leona E. Tyler, Clinical Psychology, New York, Appleton - Century - Crofts, 1962.

10. Winter, W. D., A. J. Ferreira, and R. Ranson, "Two Measures of Anxiety: A Validation," Journal of Consulting Psychology, XXVII (December, 1963), 520-524.

11. Zuckerman, M., "The Development of an Affect Adjective Check List for the Measurement of Anxiety," Journal of Consulting Psychology, XXVI (October, 1960), 457-462.

12. Zuckerman, M. and D. V. Biase, "Replication and Further Data on the Affect Adjective Check List Measure of Anxiety," Journal of Consulting Psychology, XXVI (June, 1962), 291.

12

13

13.

14.

Zuckerman, M. Check L i s t , 1965.

and B. Lubin, Manual San Diego:

_ _ _ _ _ The Mul t ip le A f f e c t Ad jec t ive Educational and I n d u s t r i a l Tes t ing Serv ice ,

Zuckerman, M., B. Lubin, L. Vogel, and E. V a l e r i u s , "Measurement of Exper imental ly Induced A f f e c t s , " Journal of Consul t ing Psychology, XXVIII (October , 1964), 418-425,

CHAPTER II

REVIEW OF THE LITERATURE

A review of the l i t e ra tu re is presented in three sections: (a)

the history of the Sentence Completion Method, (b) studies which

ut i l ized the Rotter Incomplete Sentences Blank, and (c) studies which

uti1ized the Multiple Affect Adjective Check List .

The History of the Sentence Completion Method

According to Rotter (46), the f i r s t incomplete sentence blank

for the study of personality was developed by A. F. Payne in 1928. Two

years l a t e r , in 1930, A. D. Tendler (55) published his emotional insight

t es t which consisted of twenty items devised to measure trends, fixed

a t t i tudes , conflict ing desires, sa t i s fac t ions , and attachments to

persons. His t e s t was designed in such a manner as to evoke emotional

response, allow freedom of response, and not provoke discrimination or

choice.

Wheeler (59) designed an incomplete sentence blank which he

referred to as a similies t e s t . Similes of an original nature for a

series of adjectives were provided by the subject and analyzed in terms

of thematic content following principals of H. A. Murray's theory of

personality.

Cameron (7) in 1938 devised a sentence completion blank which was

presented orally to subjects. The subjects oral responses were

analyzed for logicalness, integration, d is tor t ion, and fixed thinking.

In analyzing his data qual i ta t ively , Cameron stated that the nature of

language and the thinking process in abnormal subjects were d i f fe ren t

14

15

from those of normal subjects.

In 1941 Lorge and Thorndike (38) published a timed incomplete

sentence test whose responses were interpreted in terms of individual

t ra i ts and interests. The test in general was not productive for

studying individual differences in personality and did not y ie ld the

personal attitudes of the subject regarding his problems.

In 1941 Rhode (45) revised Payne's original test and published

a sixty-four item test which could be used with high school students.

"Rhode's items were short and generally f a i r l y unstructured." The

instructions were general and required that the subject only complete

the sentence. Responses were then analyzed in terms of Murray's

scheme using needs, press, inner states, and cathex objects.

A th i r t y item incomplete sentences test which was more structured

than others used for c l in ical purposes was described by Sanford (51)

in 1943. The Sanford test was adaptable for a variety of purposes

and had three types of items; a series stating a press, an action

pattern of some need, and ambiguous statements for which the subject

supplied the need. The Sanford test was applicable for students from

grades three to eight.

During the war the sentence completion method was greatly

accelerated. I ts use in Army general hospitals was described by Hutt

(31), Holzberg (30), and Shor (52). Shor devised a test which in his

opinion was an adaptation of the word association method. He described

i t as a "self idea completion test . " His instructions were designed to

encourage freedom and spontaneity of responses. Clinical interpretations

of the test were based on areas of reject ion, areas of resistance,

16

other methods of evasion, and f r ee c l in ica l in te rpre ta t ion in re la t ion

to the examiners' psychological t ra in ing .

Rotter and WiHerman (50) attempted to develop a sentence

completion method which could be used as a screening t e s t and be

scored by re la t ive ly untrained scorers . In developing the i r t e s t ,

Rotter and Wi Herman re-evaluated items on the t e s t s used by Hutt (31),

Shor (52), and Holzberg (30). Duplications and items which e l i c i t e d

stereotypes and reduced individual responses were eliminated. In

addition the ins t ruct ions were changed in order to eliminate the time

s e t .

Rotter , Raffer ty , and Schachtitz (49) adapted the Army t e s t

fo r college students and developed object ive scoring manuals for a

measure of maladjustment. "Separate scoring manuals were developed for

male and fo r female s tudents ." The t e s t became known as the Rotter

Incomplete Sentences Blank.

Studies Which Utilized the Rotter Incomplete Sentences Blank

For the sake of convenience, Rotter Incomplete Sentences Blank

studies will be presented under the following categories: (a) va l id i ty

and r e l i a b i l i t y s tud ies , (b) adjustment s tud ies , (c) s tudies with

scoring as a var iab le , (d) surveys and correlat ional s tud ies , and

(e) studies with anxiety as a var iable .

Validity and Rel iab i l i ty Studies

Berg (3) found that high school protocols revealed s tudents '

needs, a t t i t u d e s , a sp i ra t ions , and adjustment. An in te rpre ta t ion made

17

by a qual i f ied person without knowledge of the students' ident i ty was

ver i f ied as accurate behavior in a s ign i f icant number of cases by

teachers who had observed the student in the school environment.

Getter and Weiss (19) found that maladjusted subjects u t i l i zed the

inf irmary s ign i f i can t ly more often than adjusted subjects and

s ign i f i can t ly more maladjusted males u t i l i zed the c l i n i c than maladjusted

females.

Gardner (17) concluded that a cutt ing score of 135 correct ly

ident i f ied 80% of the heroin using males, 90% of the p i l l using males

and 100% of the heroin using females. Reported interscorer r e l i a b i l i t y

for three judges rat ing ten protocols was .88.

Kennedy, Co t t r e l l , and Smith (34) found that very superior

adolescents tended to have elevated adjustment scores on the Rotter

Incomplete Sentences Blank. They suggested that superior individuals

should have higher cut -o f f scores before being,considered maladjusted.

Rotter, Rafferty, and Lotsof (48) gave the Rotter Incomplete

Sentences Blank to approximately half the population of a small high

school. The norms which ensued were able to discriminate with high

ef f ic iency selected groups of maladjusted children from those children

of a presumable normal population. High inter-scorer r e l i a b i l i t y fo r

trained scorers, .96 for the boys' manual and .97 for the g i r l s '

manual, were found. Boys' and g i r l s ' s p l i t half r e l i a b i l i t i e s were

.74 and .86 respectively.

Rotter, Rafferty, and Schachtitz (49) reported va l i d i t y and

r e l i a b i l i t y studies conducted on 82 female and 124 male subjects. Test

interscorer r e l i a b i l i t y for two experienced scorers was .96 for the

18

female manual and .91 for the male manual. "Split-half r e l i a b i l i t y

was .83 for the female manual and .84 for the male manual." Biserial

validi ty coeff icients of .64 for females and .77 for males were reported.

I t was concluded that the t es t appeared promising for use with

college students as a screening device when a measure of confl ict

or maladjustment was required.

Churchill and Crandall (8) obtained high interscorer agreement,

above .90, among scorers who had relat ively l i t t l e psychological

training and who were trained exclusively on the Rotter Incomplete

Sentences Blank manual. Moderate t e s t - r e t e s t r e l i a b i l i t y was found for

periods of up to three years, suggesting that the Rotter Incomplete

Sentences Blank measured more than momentary mood.

Windle (61) discovered that there was a tendency for t e s t - r e t e s t

adjustment scores on the Rotter Incomplete Sentences Blank to lean

towards better adjustment when the interval between the tes t s was less

than two months. Among the factors he cited as contributing to higher

adjustment scores were the following: (a) the pull of social norms

plus the gaining of insight into the nature of the t e s t , (b) ageing,

(c) arousal and loss of in te res t , and (d) relaxation or hos t i l i ty .

On the other hand, Bendig and Bruder (2) f e l t that the tendency for the

adjustment scores to regress to normalcy was due to the u t i l iza t ion of

the same items or forms at each administration. A third explanation for

variance in re tes t si tuations was advanced by Stephens (54). He

concluded that the variance was due to the ins tab i l i ty of t e s t scores,

changes in adjustment, and variance accrued to each item.

19

Adjustment Studies

Goldberg and Stark (22) used the Rotter Incomplete Sentences

Blank along with several other t e s t s to measure personali ty and

a t t i t ude corre la tes of po l i t i ca l preference in the elect ion of 1964.

Johnson supporters were s ign i f i can t ly more maladjusted as measured

by the Rotter Incomplete Sentences Blank. Nixon supporters, on the

other hand, tended to be bet ter adjusted, more au thor i t a r i an , more

r e l ig ious , and more orthodox in the i r views.

Morton (42) used the Rotter Incomplete Sentences Blank to measure

maladjustment a f t e r psychotherapy emphasizing ins ight was u t i l i z e d .

His experimental group showed a mean decrease in maladjustment at the

.001 leve l . His control group also showed a mean decrease in malad-

justment but at the .05 leve l . He concluded tha t adjustment scores

obtained from the Rotter Incomplete Sentences Blank regressed toward

the mean of the population t e s t score with repeated usage of the Rotter

Incomplete Sentences Blank.

Studies with Scoring as a Variable

Walker and Linden (57) had four judges varying in level of

psychological sophis t icat ion score Rotter Incomplete Sentences Blank

protocols. In terms of consistency and accuracy of judgements made, a l l

judges were successful . The following conclusions were drawn: (a) the

Rotter Incomplete Sentences Blank was evaluated with consistency by even

inexperienced judges although experienced judges did s l i gh t ly be t ter in

terms of exact agreement in ra t ings , (b) a l l judges c l a s s i f i ed protocols

with an accuracy be t te r than chance would allow, and (c) marked

20

individual differences in judging abi l i ty appeared among judges of all

levels of sophistication.

Guertin (24) examined sentence completion protocols for gross

errors. A factor analysis demonstrated that there was consistency in

the nature of the errors made and that the type of error made was

related to other material of clinical significance. "The four factors

obtained were labeled: Word Carelessness, Irrelevance, Poor Spelling,

and Telegraphic Phrasing."

Fiske and Buskirk (14) judged sentence completion protocols in

terms of a Q sort of need variables. The experimental design was such

that i t was possible to compare the agreement between interpretat ions

of each subjects ' three protocols with the agreement between interpre-

tations for d i f ferent subjects. In one-fourth of the comparisons, the

agreement was higher with protocols for d i f fe ren t subjects than with

other protocols from the same subject. "Thus a single protocol may be

an insuff ic ient basis for an interpretation that d i f fe ren t ia tes one

person from other people," (14, p. 180).

Surveys and Correlational Studies

Mills (41) ranked tes t s used in experimental research according to

their frequency of occurrence in the Journal of Projective Techniques.

Sentence completion tes t s ranked f i f t h for the period from 1960-1964.

Crenshaw (9) extended the survey to include ten journals and an

eighteen year time period. He found that the most frequently used

projective techniques had changed l i t t l e over time and sentence completion

tes ts were ranked f i f t h .

Goldbergs' (21) survey ranked Sentence Completion Tests s ixth, but

21

stated that the Rotter Incomplete Sentences Blank was the tes t most

frequently used of the Sentence Completion Methods. Unanimous agreement

held that the method was useful in assessing interpersonal a t t i tudes .

In addition, sentence completions are used more as a clinical than

a research instrument.

Bieri , Blacharsky, and Reid (4) found the following correlations

when the Rotter Incomplete Sentences Blank was used as a cri terion of

adjustment and the Manifest Anxiety Scale was used as the predicting

t e s t : (a) a correlation of .46 between the Rotter Incomplete Sentences

Blank and the Taylor Manifest Anxiety Scale, (b) an insignif icant

positive relationship between maladjustment and the ab i l i ty to predict

differences correctly and, (d) a s ignif icant negative correlation

between assimilative projection and maladjustment.

Denenberg (13) obtained a correlation of 0.388 between scores on

the Rotter Incomplete Sentences Blank and maze performance. After

cross-validation the correlation rose to 0.459.

Arnold and Walker (1) found a moderate relationship between scores

obtained by the same individuals on the self and other-reference

sentence completion blank.

Fitzgerald (15) found that correlations between frequency of

response ratings on the Rotter Incomplete Sentences Blank were low but

s igni f icant . In addition, these ratings correlated s ignif icant ly with

need value and neecj potent ial , but not to "conflict" score. After

obtaining a near zero correlation between the ratings on the Thermatic

Apperception Test and on the Rotter Incomplete Sentences Blank,

Fitzgerald (15) concluded that each tes t measured d i f fe ren t facets of

22

the personali ty. The Rotter Incomplete Sentences Blank tended to

provide information about overt dependent behavior.

Tempone and Lamb (56) found a relat ionship of .73 between repression

sensi t izat ion and adjustment scores obtained on the Rotter Incomplete

Sentences Blank.

Rotter, Raffer ty, and Lotsof (48) found low posit ive correlations

between adjustment scores obtained from the Rotter Incomplete Sentences

Blank and interview ratings made by a psychologist.

Si ipola (53) found a d i rect relat ionship between the amount of

ego-alien content e l i c i t ed by the Rotter Incomplete Sentences Blank and

the amount of time pressure imposed. In addit ion anxious subjects

showed a greater degree of incongruence than non-anxious subjects.

Studies in Which Anxiety was a Variable

Renner, Maher, and Campbell (44) constructed a scoring system for

the Rotter Incomplete Sentences Blank which assessed anxiety.

"Correlations of .24 and .34, respectively, for male and female subjects

were reported between the i r Sentence Completion Test anxiety scores and

val idat ion c r i t e r i a based on sel f -descr ipt ion and peer ratings of

anxiety.," (43, p. 197).

Neuringer and Orwick (43) found s t a t i s t i c a l l y s ign i f i cant

differences between high anxiety subjects when they were compared with

middle and low anxious groups. A l l protocols were scored according to

the anxiety c r i t e r i a found in Rotter and Rafferty (47).

I t appears that a great deal of research has been conducted using

t h e Rotter Incomplete Sentences Blank as a variable. However from a

23

review of the l i t e ra ture some doubt s t i l l remains as to i t s s t ab i l i t y .

Studies Which Utilized the Multiple Affect Adjective Check List

For the sake of convenience Multiple Affect Adjective Check List

studies will be presented under the following categories: (a) normative

data, (b) hypnotically induced emotional s t a tes , (c) army basic t raining,

(d) behavioral therapy, (e) psychotherapy, (f) sens i t iv i ty t ra ining,

(g) relaxation t raining, (h) drugs and alcohol, ( i) correlations with

other data, and ( j ) miscellaneous studies.

Normative Data

Giddon (20) gave the Anxiety Affect Adjective Check List to

seventy-one female dental hygiene students once a week for fourteen

weeks prior to clinical practice periods. Individual subjects ' means

ranged from 2.5 to 10.7 with an overall group mean of 7.44±1.64.

Harrow, Colbert, Detre, and Bakeman (25) gave the Today form of the

Anxiety Affect Adjective Check List to seventeen schizophrenic and for ty-

three depressive pat ients . The means and standard deviations of the

anxiety scores were 11,95±5.16 for the schizophrenics and 14.93±3.71

for the depressives. The schizophrenics had a s ignif icant ly lower mean

score (p< .02) . Zuckerman and Biase (61) concurred and Zuckerman, Lubin,

and Robins (62) obtained similar resul ts with respect to the depressive

scale.

Hayes (28) administered the Anxiety Affect Adjective Check List

twenty-four times to sixty-seven sophomoiue nursing students during the

f i r s t trimester of clinical nursing. His resul ts indicated that the

24

subjects reported themselves as being more anxious from day-to-day

than generally. A considerable increase in anxiety was reported before

course examinations and after the f i na l .

Knapp, Zimmerman, and Roscoe (35) administered the Multiple

Affect Adjective Check List to 306 entering freshmen prior to a f ive

hour admissions test. He noted that the depressive scores and the males

anxiety scores were s l ight ly higher than the norms reported in the

manual. He attr ibuted this rise in scores to the imminence of the

testing session.

Hypnotically Induced Emotional States

Levi t t , DenBreeijen, and Persky (36) used the Anxiety Affect

Adjective Check List to obtain measures of anxiety prior to and during

the induction of anxiety. These researchers noted that twenty-one of the

twenty-four adjectives which dif ferent iated anxiety patients from

normals also di f ferent iated the hypnotic anxiety state from the control

state. Baseline scores under hypnosis, i t was also noted, were lower

than under normal conditions.

Zuckerman, Persky, and Link (65) attempted to test the hypothesis

that affect states just prior to hypnotic induction are related to

subsequent hypnotizabil i ty. Their results indicated that the Host i l i ty

Scale of the Multiple Affect Adjective Check List was s igni f icant ly and

negatively correlated with hypnotizabil i ty for subjects tested in small,

highly motivated groups, but were unrelated to hypnotizabil i ty in a

large, less motivated group.

25

Army Basic Training

Date!, Gieseking, Engle, and Dougher (12) gave the Mult iple

Affect Adject ive Check L is t once a week for nine weeks to th i r ty -n ine

subjects undergoing basic combat t ra in ing. Their results showed a

decline during the week pr ior to the th i rd test ing occasion, but an

increase in h o s t i l i t y and depression during the fol lowing two weeks.

The authors noted that the Mult iple Af fect Adjective Check L is t scores

corresponded with known stresses. More spec i f i ca l l y , depression and

h o s t i l i t y were affected by the general stress of basic t ra in ing while

KP e l i c i t ed a speci f ic h o s t i l i t y reaction.

Datel and Engle (10) replicated the above study from the beginning

to the end of basic t ra in ing. Their results suggested that h o s t i l i t y

may replace anxiety over time during basic combat t ra in ing .

Datel, Engle, and Barba (11) extended the above two studies to

include a whole company. The results showed that re l iab le af fect

changes occurred in basic combat t ra in ing.

Bournes, Col i , and Datel (6) found that anxiety was the dominant

a f fec t expressed by medics in Viet Nam. Two years la ter Bourne, Col i ,

and Datel (5) found that h o s t i l i t y was the predominate af fect expressed.

Behavioral Therapy

Folkins, Lawson, Opton, and Lazarus (16) sought to determine i f

desensit ization could serve as a stress reducer and i f i t d id , was the

complete procedure necessary or were ei ther or both components jus t as

ef fect ive as the f u l l procedure in reducing stress reactions. Subjects

in the experimental procedure received one of four t ra in ing procedures

26

for three t ra ining sessions a week apart before exposure to a s t r e s s fu l

f i lm, an analogue of therapeutic desens i t iza t ion , re laxa t ion , cognitive

rehearsa l , or no-training control . The subjects completed the

Anxiety Affect Adjective Check List questionaire following each of the

f i r s t two t ra ining sessions and immediately a f t e r the s t r e s s fu l movie

scene. The experimental r e su l t s indicated tha t subjects who had

received cognit ive-rehearsal in the three t ra ining sessions reported

the l eas t amount of anxiety on the Affect Adjective Check List Scale.

On the other hand, subjects who received complete desensi t izat ion

reported more anxiety on the Affect Anxiety Check List Scale than

everyone else except the control group. As a consequence of the

experimental r e s u l t s , Folkins, Lawson, Opton, and Lazarus (16)

concluded tha t cognitive rehearsal was the most e f f e c t i v e s t ress

reducer while desens i t iza t ion , which combined relaxat ion and cognitive

rehearsa l , was the l eas t e f f e c t i v e .

An in teres t ing by-product of th i s experiment was the discovery tha t

males who reported high s t ress to a female experimentor described

themselves on the Anxiety Affect Adjective Check List as "highly

anxious, with low self-confidence and personal adjustment, low needs for

achievement and dominance, and high needs fo r abasement and deference."

Females, on the other hand "who reported high s t ress to a male experi-

mentor described themselves as highly anxious, with low needs for

achievement and dominance and high needs fo r succorance, abasement, and

deference," (16> p. 109).

The authors suggested that the above r e su l t s may be interpreted

as e i ther a combination of sex-role insecur i ty added to the experi-

27

mentally induced stress or that males and females found i t easier

to admit a personality disturbance to a member of the opposite sex.

Greer and Turteltaub (18) assessed the e f fec t of a subjects '

subsequent reaction a f t e r observing the behavior of other subjects in

a fear arousing s i tuat ion. The Multiple Affect Adjective Check List was

used to describe the emotional responses to the fear arousing s i tuat ion.

A signif icant difference in responses to the Multiple Affect Adjective

Check List existed between the subjects who saw their model respond

fear fu l ly in a feared situation and those subjects whose model was

brave in a fearful s i tuat ion.

Psychotherapy

Goldstein (23) investigated the e f fec ts of nonspecific factors in

psychotherapy and several characteris t ics or psychological treatment

which contributed to the placebo e f f ec t . He administered the Multiple

Affect Adjective Check List and several other t es t s which measured

anxiety, moods, and psychological adjustment to 96 neuropsychiatry

hospital patients capable of returning to the community. The subjects

were divided into pairs and each pair was placed into one of the

following four treatment conditions: (a) positive set concerning

treatment outcome with an alleged therapist congruent with their

expectations, (b) positive set concerning treatment outcome with an

alleged therapist incongruent with their expectations, (c) neutral set

concerning treatment outcome and an alleged therapist congruent with

their expectations, and (d) neutral set concerning treatment outcome

and an alleged therapist incongruent with thei r expectations. One

28

member of each pair spoke to a therapist in the form of a tape recorder

while the other member of the pair waited.

The results indicated that subjects who spoke into the tape

recorder reported lowered anxiety and had changed certain moods in

"therapeutic directions." Furthermore, subjects who were given a

therapist congruent with their expectations responded to treatment

with a greater positive placebo effect or measured by the Multiple

Affect Adjective Check List and the other dependent variables. Subjects

who received a positive set and a therapist congruent with their

expectations showed the greatest placebo ef fect .

Sensit ivi ty Training

Lubin and Zuckerman (40) used the Multiple Affect Adjective Check

List to contrast pre and post conference scores or 43 subjects who

completed sensi t iv i ty training. His results indicated that there was

a signif icant relationship between the affect variable and the degree

to which feelings are openly shared in the group. Participants reported

experiencing less anxiety, hos t i l i t y , and depression as feelings were

more openly shared. Lubin and Zuckerman (40) concluded that their

findings supported a basic tenet of small group training philosophy, that

the open showing of feelings tended to have a releasing effect upon

group processes and the behavior of individual subjects.

Lubin (39) found a low correlation between pre-conference sel f -

disclosure inventory and the mean of the three scales of the Multiple

Affect Adjective Check List administered after each of the eight

sensi t iv i ty group sessions.

29

Relaxation Training

Johnson and Spielberger (33) used the Multiple Affect Adjective

Check List to investigate the e f fec ts of muscle relaxation training

procedures and the passage of time on the empirical measurement of s ta te

and t r a i t anxiety. After relaxation training procedures s ta te anxiety

declined s ignif icant ly whereas t r a i t anxiety did not s ignif icant ly

decline. In addition, correlations between t r a i t anxiety measures

were high and stable over time whereas correlations between s ta te

anxiety measures were moderate or negligible. The results of this

experiment were interpreted as supporting the view that s ta te and t r a i t

anxiety are two separate and d is t inc t constructs.

Drugs and Alcohol

Hankoff and Rudorfer (26) administered pyrbenzindole and a placebo

to 37 adults , ambulatory pat ients , in a randomized double blind design

for two weeks. The data collected from these subjects consisted of a

psychiatr is ts ' progress notes and scores from the Anxiety Affect

Adjective Check List and the Target Symptom Rating Scale. The resul ts

indicated that pyrbenzindole had no more e f fec t than the placebo in

relieving anxiety and depression or in producing global clinical

improvement. However patients who were rated by a psychiatr is t as being

more anxious or more depressed on the Target Symptom Rating Scale showed

higher Affect Adjective Check List scores. Similarly, patients who

manifested an a t t i tude of greater denial toward their mental i l lness

showed lower Affect Adjective Check List scores.

Hankoff, Rudorfer, and Paley (27), using the same experimental

30

design as the Hankoff and Rudorfer study (26), assessed the ef fec t iveness

of Ataraxics; chlordiazepoxide, chlorpromazine, and meprobamate. These

researchers found tha t the number of words indicat ive of anxiety as

checked by the pat ients on the Affect Adjective Check List was reduced

in a l l treatment groups, but was s t a t i s t i c a l l y s ign i f i can t by t t e s t

only with chlordiazepoxide and chlorpromazine. However when the data

were subjected to an analysis of co-variance, the r e su l t s indicated

a s t a t i s t i c a l l y s ign i f i can t d i f ference between treatment groups fo r

placebo, chlordiazepoxide, and chlorpromazine.

Williams (60) investigated the psychological reasons fo r normal

social drinking and t r i ed to determine why some people drink frequently

and excessively. He held f ive cocktail par t ies in which 91 students

from two colleges par t ic ipa ted . The data consisted of a problem-

drinking sca le , which measured proneness to alcoholism, and anxiety

and depression adjec t ive check l i s t s which weee given before the par ty ,

a f t e r 4 ounces of alcohol had been consumed, and a t the end of the

party. "Pre-party resu l t s indicated that problem drinking was

posi t ively associated with anxiety, depression, and the amount of

alcohol consumed." Anxiety and depression were found to decrease

s ign i f i can t ly a t low levels of alcohol consumption, but to reverse and

increase r i s ing to pre-party levels a f t e r eight ounces of alcohol had

been consumed.

Zuckerman, Kolin, Price and Zoob (64) developed a Sensation

Seeking Scale which correlated s ign i f i can t ly with the Anxiety Affect

Adjective Check Lis t . Correlations with the Depression and Host i l i ty

Scales were low, pos i t ive , and nonsignif icant .

31

Correlations With Other Data

Giddon (20) noted several correla t ions between the Anxiety Affect

AdjectivejXheck List and other rat ing devices, personal habi t s , and

physiological measurements. He found that the mean of the Anxiety

Affect Adjective Check List was s ign i f i can t ly correlated d i rec t ly with

f inger temperature, the Anxiety Scale of the Forced Choice Anxiety

Scale and the number of cups of coffee consumed in a day. At the

same time the mean of the Anxiety Affect Adjective Check List was

s ign i f i can t ly correlated inversely with f i r s t semester grade point

average, d i a s to l i c blood pressure and almost inversely correlated with

the Social Desi rabi l i ty Scale of the Forced Choice Anxiety Scale. The

Edwards Preference Scale had no re la t ionship with the Anxiety Affect

Adjective Check Lis t .

Hayes (28) reported a s ign i f i can t negative correla t ion between the

Multiple Affect Adjective Test and course grades. Poor grades were

frequently associated with h o s t i l i t y . This finding i s supported by;

Lieberman (37) who noted tha t subjects who do poorly on an examination

are more apprehensive when expecting the examination and on the actual

day of the examination.

Miscellaneous Studies

Weaver (58) gave Zuckermans' Affect Adjective Check List and B i l l s '

Index of Adjustment of Values to 181 high school seniors . On the basis

of the adjustment scores the students were divided into three groups;

high, medium and low adjustment groups. One-third of each group received

a posi t ive personali ty evaluation, one-third received a negative

32

evaluation, and one-third served as the control group and were told tha t

the i r evaluations would be ready in two days. The r e su l t s indicated the

following: f i r s t , subjects who were given negative personali ty

evaluations and who had in r ea l i t y high self concepts re jected the

negative personali ty information and did not manifest e i ther any

s ign i f i can t self-concept change nor any anxiety. Second, female

subjects who were given a posi t ive self concept evaluation and who

had in r e a l i t y a low concept raised the i r self concepts. Male subjects

in the same s i tua t ions showed no s ign i f i can t se l f concept changes. Both

males and females accepted posi t ive evaluations and did not indicate the

presence of any anxiety. Third, subjects who were given a posi t ive

self concept evaluation and a posi t ive self concept image or ig ina l ly

showed no s ign i f i can t self concept change, no s ign i f i can t change in

anxiety scores, and accepted the posi t ive se l f concept evaluation.

Fourth, subjects who were given a low self concept image and in r e a l i t y

had a low self concept image rejected the information, raised the i r

self concept scores and remained r e l a t ive ly s table with respect to t he i r

anxiety score. F i f t h , changes in anxiety scores following treatment

were very d i f f e r e n t fo r subjects who d i f fe red in i n i t i a l anxiety l eve l .

Highly anxious subjects had lower anxiety scores following any treatment

whereas low anxious subjects raised the i r anxiety scores when given a

posi t ive evaluation.

In reviewing the research conducted with the Multiple Affect

Adjective Check List ,one may conclude that fo r a t e s t so recently

developed, a great deal of research has been conducted in the u t i l i z a t i o n

of th i s instrument. Prac t ica l ly al l of the research concerning th i s

33

instrument has been favorable. The only exception was Herron (29), who

concluded that the major component of each of the Multiple Affect

Adjective Check List Today Scales was a function of the response s e t .

Consequently he stated tha t the Today form of the Multiple Affect

Adjective Check List should be used with caution. Therefore taking al l

of the research into consideration, i t appears that the Multiple Affect

Adjective Check List i s an adequate instrument designed to measure

anxiety, h o s t i l i t y and depression.

CHAPTER BIBLIOGRAPHY

1. Arnold, Frank C. and Verne A. Halter, "The Relationship Between a Self and Other-Reference Sentence Completion Test," Journal of Counseling Psychology, IV, (Spring, 1957), 65-70.

2. Bendig, A. VI. and Gail Bruder, "The Effect of Repeated Testing on Anxiety Scale Scores," Journal of Consulting Psychology, XXVI (August, 1962) 392.

3. Berg, Wilbert Arthur, "Determining Va l id i ty of the Incomplete Sentences Blank through Appraisal of Qual i tat ive Interpretat ions, Dissertation Abstracts, XXII, (January, 1952), 504-505.

4. B ie r i , James, Edward Blacharsky, and J. William Reid, "Predictive Behavior and Personal Adjustment," Journal of Consulting Psychology, XIX, (October, 195b), 351-366.

5. Bourne, P. G., W. M. Col i , and W. E. Date!, "Affect Levels of Ten Special Forces Soldiers Under Threat of Attack," Psycholoqical Reports, XXII, (Ap r i l , 1968), 363-366.

6. Bourne, P. G., W. M. Col i , and VI. E. Datel, "Anxiety Levels of Six Helicopter Ambulance Medics in a Combat Zone," Psychological Reports, XIX, (December, 1966), 821-822.

7. Cameron, t i . , "Reasoning, regression, and communication in schizophrenia," Psychological Monogranh, L, (whole #221, 1938), 1-34.

8. Churchi l l , Ruth and V. J. Crandall, "The Re l iab i l i t y and Va l id i ty of the Rotter Incomplete Sentences Test," Journal of Consulting Psychology, XIX, (October, 1955), 345-350.

9. Crenshaw, David, Suzanne Bohn, Marlene R. Hoffman, John M. Matheus, Stefan G. Offenbach, "The Use of Projective Methods in Research: 1947-1965," Journal of Projective Techniques and Personality Assessment, XXXII, (February, 1968), 3-9.

10. Datel, W. E. and E. 0. Engle, "Affect Levels in Another Platoon of Basic Trainees," Psychological Reports, XIX, (October, 1966), 407-412.

11. Datel, U. E., E. 0. Engle, and M. A. Barba, "Af fect Levels in a Company of Basic Trainees," Psychological Reports, XIX, (December, 1966), 903-909.

34

35

12. Datel, U. E., C. F. Gieseking, E. 0. Engle, and M. J. Douqher, "Affect Levels in a Platoon of Basic Trainees," Psycholoqical Reports, XVII I , (February, 1966), 271-285.

13. Denenberq, Victor H., "The Relationship Between a Measure of Kinesthesis and Two Indices of Adjustment," Journal of General Psychology, LXII , (January, 1960), 43-52.

14. Fiske, Donald W. and Charles Van Buskirk, "The S tab i l i t y of Interpretations of Sentence Completion Tests," Journal of Consulting Psychology, XXII I , (Apr i l , 1959), 177-180.

15. Fitzgerald, Bernard Joseph, "The Relationship of Two Projective Measures to a Sociometric Measure of Dependent Behavior," Dissertation Abstracts, XX, (December, 1959), 2380-2381.

16. Folkins, C. H., K. D. Lav/son, E. M. Opton, J r . , and R. S. Lazarus, "Desensitization and the Experimented Reduction of Threat," Journal of Abnormal Psychology, LXXIII, (Ap r i l , 1968), 100-113.

17. Gardner, James M., "The Adjustment of Drug Addicts as Measured by the Sentence Completion Test," Journal of Projective Techniques and Personality Assessment, XXXI, (June"!967), 28-29.

18. Geer, J. H. and A. Turteltaub, "Fear Reduction Following Observation of a Model," Journal of Personality and Social Psychology, VI, (July, 1967) 32/-331. ^ 1 ^

19. Getter, Herbert and Stephan D. Weiss, "The Rotter Incomplete Sentences Blank Adjustment Score as an Indicator of Somatic Complaint Frequency," Journal of Projective Techniques and Personality Assessment, XXXII/""["June. 1968), 266.

20. Giddon, Donald B., "Individual Differences in Average Affect Adjective Check L is t (AACL) Scores: Their Relation to Psycholoqical and Physiological Variables," Psychological Renorts, XIV, (Apr i l , 19(54), 541-542.

21. Goldberg, P h i l l i p , "The Current Status of Sentence Completion Methods," Journal of Projective Techniques and Personality and Assessment, XXXII.TJune. 215-221.

22. Goldberg, Ph i l l i p and Milton J. Stark, "Johnson or Goldwater?: Some Personality and Att i tude Correlates of Po l i t i ca l Choice," Psychological Reports, XVII, (October, 1951), 627-631.

23. Goldstein, H., "Placebo, Psychotherapy, and Change in Anxiety, Mood and Adjustment," Dissertation Abstracts, XXVI, (September, October, 1965), 1775-177F:

36

24. Guertin, Wilson H., "An Analysis of Gross Errors on a Sentence Completion Test," Journal of Cl inical Psvcholoqy, XV, (October, 1959), 415-416.

25. Harrow, M., J. Colbert, T. Detre, and R. Bakeman, "Symptomatology and Subjective Denression in Current Depressive States," Archives of General Psychiatry, XIV, (February, 1966), 203-212.

26. Hankoff, L. I)., L. Rudorfer, and H. M. Paley, "A Reference Study of Ataraxics: A Tv/o Week Double Blind Outpatient Evaluation," Journal of New Drugs, I I , (May, June, 1962), 173-178.

27. Hankoff, L. D., L. Rudorfer, and H. M. Paley, "The Psychiatric use of Pyshenziondole: A Double Blind Out-Patients Study," Journal of New Drugs, I I , (Hay, June, 1962), 167-172.

28. Hayes, Carol V., "The Measurement of Anxiety in Sophomore Nursing Students Using Zuckerman's AACL," Nursing Research, XV, (Summer, 1966), 262-267. '

29. Herron, E. Wayne, Lewis Bernstein, and Harold Rosen, "Psychometric Analysis of the Multiple Affect Adjective Check L is t : MAACL Today," American Psychologist, XXII, (July, 1967), 591.

30. Holzberg, J . , A. Teicher, and J. L. Taylor, "Contributions of Cl inical Psychology to Mi l i ta ry Neuropsychiatry in an Army Psychiatric Hospital," Journal of Cl inical Psychology, I I I , (January, 1947), 84-95.

31. Hutt, M. L . , "The Use of Projective Methods of Personality Measurements in Army Medical Insta l la t ions," Journal of Cl inical Psychology, I , (Apr i l , 1945), 134-14(57"

32. Johnson, D. T . , "Effects of Interview Stress on Measures of State and Tra i t Anxietv," Journal of Abnormal Psychology, LXXIII, (June, 1968), 245-2511

33. Johnson, D. T. and C. D. Spielberger, "The Effects of Relaxation Training and the Passage of Time on Measures of State and Tra i t Anxiety," Journal of Cl inical Psycholoay, XXIV, (January, 1968), 20-23.

34. Kennedy, Wallace A., Ted Cot t re l l , and A1 Smith, "Norms of Gifted Adolescents on the Rotter Incomplete Sentences Blank," Journal of Cl inical Psychology, XIX, (July, 1963), 314-316.

35. Knapp, R. R., W. S. Zimmerman, W. S. Roscoe, and W. B. Michael, "The Effects of College Entrance Examination on Effects of Anxiety, Depression, and Hos t i l i t y , " Educational and Psychological Measurement, XXVII, (Winter, 1967), 112V-1T2(5T

37

36. Lev i t t , E. E., A. DenBreei.ien, arid H. Persky, "The Induction of Cl inical Anxiety by Means of a Standardized Hypnotic Technique," American Journal of Cl inical Hypnosis, I I , (Apr i l , 1960), 206-214.

37. Lieberman, L. R., "Effects Upon Anxiety, Depression, and Host i l i t y of Postponement of an Examination," Perceptual and Motor Sk i l l s , XXII I , (December, 1966), 1051-1054.

38. Lorges, I . and E. L. Thorndike, "The Value of the Responses in a Completion Test as Indicators of Personal Tra i ts , " Journal of Applied Psychology, XXV, (Apr i l , 1941), 191-199.

39. Lubin, B., "A Modified Version of the Self-Disclosure Inventory," Psychological Reports', XVII, (October, 1965), 498.

40. Lubin, Bernard and Marvin Zuckerman, "Affect ive and Perceptual-Cognitive Pattern in Sensi t iv i ty Traininq Groups," Psychological Reports, XXI, (October, 1967), 365-376.

41. M i l l s , David H., "The Research Use of Projective Techniques: A Seventeen Year Survey," Journal of Projective Techniques and Personality Assessment, XXIX, (December, 1965), 513-515.

42. Morton, Robert B. "An Experiment in Brief Psychotherapy," Psychological Monographs, LXIX, (January, 1955), 1-17.

43. Neuringer, Charles and Perry 0. Orwick, "The Measurement of Anxiety on the Sentence Completion Test," Journal of General Psychology, LXXVIII, (Apr i l , 1968), 197-207:

44. Renner, Edward, Brendan A. Maher, and Donald T. Campbell, "The Val id i ty of a Method for Scoring Sentence Completion Responses for Anxiety, Dependency, and Hos t i l i t y , " Journal of Applied Psychology, XXXXVI, (Auqust, 1962), 285-2W:

45. Rhodes, A. R., "Explorations in Personality by the Sentence Completion Method," Journal of Applied Psycholoqy, XXX, (Apr i l , 1946), 169-lFH

46. Rotter, Julian B., "Word Association and Sentence Completion Methods," In H. H. Anderson and G. L. Anderson (Eds.), An Introduction to Projective Techniques, New York, Prentice-HalT7 1951, 279-311.

47. Rotter, J. B. and Janet E. Rafferty, Manual: The Rotter Incomplete Sentences Blank, Mew York: Psychological Corporation, 1950.

48. Rotter, Julian B., Janet E. Rafferty, and Antoinette B. Lotsof, "The Val id i ty of the Rotter Incomplete Sentences Blank: High School Norm" Journal of Consulting Psychology, XVII I , (Ap r i l , 1954), io5- i IT ; "

38

49. Rotter, Julian B., Janet E. Rafferty and Eva Schachtitz, "Validation of the Rotter Incomplete Sentences Blank for College Screening," Journal of Consulting Psychology, X I I I , (October, 1949), 348-346.

50. Rotter, J. B. and B. Willerman, "The Incomplete Sentences Test," Journal of Consulting Psychology, XI, (January, 1947), 43-48.

51. Sanford, R. N., "Physique, Personality, and Scholarship," Monogram of Social Research and Child Development, V I I I , (January ,~T943), 1.

52. Shor, J . , "Report on a Verbal Projective Technique," Journal of Cl in ical Psychology, I I , (July, 1946), 279-282.

53. Si ipo la, Elsa M., "Incongruence of Sentence Completion Under Time Pressure and Freedom," Journal of Projective Techniques and Personal i ty Assessment, XXXlY, T0ecember,,"T^68), 562=570.

54. Stephens, Mark W., "The Incomplete Sentences Blank: Sources of Variance in Retest Re l i ab i l i t y , " Journal of Cl in ical Psychology, XVI, (July, 1960), 331-333.

55. Tendler, A. D., "A Preliminary Report on a Test for Emotional Ins ight , " Journal of Applied Psychology, XIV, (Apr i l , 1930), 123-136.

56. Tempone, Vincent J. and Wesley Lamb, "Repression-Sensitization and i t s Relation to Measures of Adjustments and Conf l i c t , " Journal of Consulting Psychology, XXXI, (Ap r i l , 1967), 131-136.

57. Walker, C. Eugene and James D. Linden, "Varying Degrees of Psychological Sophistication in the Interpretat ion of Sentence Completion Data," Journal of Cl in ical Psychology, XXII I , (Ap r i l , 1967), 229-231.

58. Weaver, T. T . , J r . , "Effects of Positive and Negative Personality Evaluations on the Self Concepts of High School Seniors," Dissertation Abstracts, XXVI, (September, October, 1965), 1785-1786.

59. Wheeler, D. R., "Imaginal Productivity Tests," in H. A. Murray, Exploration in Personality, New York, Oxford University Press, 1938, 545-55157 680-684.

60. Williams, A. F. , "Social Drinking, Anxiety, and Depression," Journal of Personality and Social Psychology, I I I , (June, 1966), 689-693.

61. Windle, C. "Further Studies of Test-Retest Effect on Personality Questional'res." Educational and Psychological Measurement, XV, (Autumn, 1955;, 246-253.

39

62. Zuckerman, M. and D. V. Biase, "Replication and Further Data on the Affect Adjective Check L is t Measure of Anxiety," Journal of Consulting Psychology, XXVI, (June, 1962), 291.

63. Zuckerman, Marvin, Bernard Lubin, and Sidney Robins, "Validation of the Mult iple Affect Adjective Check L is t in Cl in ical Si tuat ions," Journal of Consulting Psychology, XXIX, (December, 1965), 594.

64. Zuckerman, M., E. A. Kolin, L. Pine, and I . Zoob, "Development of a Sensation-Seeking Scale," Journal of Consulting Psychology, XXVIII, (December, 1964),

65. Zuckerman, M., H. Persk.y, and K. E. Link, "The Relation of Mood and Hypnot izabi l i ty: An I l l u s t r a t i on of the Importance of the State vs. Tra i t D is t inc t ion, " Journal of Consulting Psychology, XXXI, (October, 1967), 464-470:

CHAPTER I I I

METHOD

This chapter presents the method and procedure used to determine

the effect of anxiety, hos t i l i t y , and depression on responses to the

Rotter Incomplete Sentences Blank. The Today form of the Multiple

Affect Adjective Check List was used to measure anxiety, hos t i l i t y ,

and depression.

Description of Subjects and Apparatus

The subjects consisted of th i r ty-e ight students, sixteen females

and twenty-two males, who were enrolled in General Psychology I and

who met the following c r i te r ia : (1) were enrolled in one of the two

General Psychology I sections which met from 12:00 - 1:00 PM on Mon-

day, Wednesday, and Friday; (2) completed a minimum of two of the

three baseline tests and the f inal test .

The apparatus consisted of the following two tests: the Rotter

Incomplete Sentences Blank and the Today form of the Multiple Affect

Adjective Check L is t . An example of each test is provided in Appendix

A and Appendix B.

40

41

Procedure

The instructor read the following instructions to his class

before administering the Today form of the Multiple Affect Adjective

Check List and the Rotter Incomplete Sentences Blank: "The tes ts which

will be passed out are part of an experiment in which the examiner is

interested in measuring day to day changes in feel ings. You will be

given the tes t every Wednesday at the s t a r t of the class hour. There is

no time l imit , but you are urged to complete them as quickly as

possible." The instructor then passed out the stapled tes t s which had

been previously arranged in a counter-balanced order. Following the

distr ibution of the t e s t s , the instructor stated that the instructions

for taking each tes t were printed on the t es t and he proceeded to read

them to the class. The instructions were read aloud by the instructor

for only the f i r s t administration of the tes t s in order to make certain

everyone understood the directions and to answer any questions concerning

procedure. After taking the t e s t s , the students were requested to write

down on a separate sheet of paper any comments they had concerning the

experiment.

On the second and third administration of the t e s t s , which

occurred seven and fourteen days a f t e r the in i t i a l administration, the

subjects were again told that the purpose of the experiment was to

measure day to day changes in feel ings. In addition, the subjects were

reminded to read and follow the instructions which were printed on the

t e s t s . An additional sheet was provided for any additional comments

the subjects wished to make.

The fourth administration of the tes t s occurred twenty-eight days

42

a f t e r the administration of the in i t i a l t e s t s . The subjects in the

control group followed the same procedures which they had followed for

the previous three times. The subjects in the experimental group, however,

followed a d i f fe ren t procedure.

In the experimental group, the instructor entered the class

carrying the tes t s for the experiment as he had done in the previous

administrations. In addition, he carried a stack of t es t s and answer

sheets identical to those used for a major course examination.

Before distr ibuting the experimental t e s t s , the instructor told

his class that as soon as they had finished f i l l i n g out the tes t s for

the experiment, they would have their second major course examination.

This examination as well as all other examinations for this particular

course had been previously scheduled and given in the form of a

"handout" to the students. However, one week previously the instructor

had told the class that the course examination would be postponed until

the Wednesday a f t e r Easter, one week l a t e r .

The announcement of the unexpected course examination was met

with a roar of protest . One student reminded the instructor that the

course examination had been rescheduled for the following week. The

instructor replied that he had to turn in mid-semester grades to the

administration of f ice before Easter break and that he f e l t that the

results of the two course examinations would be a better indication of

their actual grade in this particular course.

Another student stated that he was not prepared to take a course

examination. The instructor replied that if he had been keeping up

with his reading and if his attendance in class had been regular then

43

he should not have any trouble with the course examination.

The instructor then passed out the tests for the experiment

and to ld the class to f i n i sh as quickly as possible. He indicated

that as soon as everyone was f inished he would pick up a l l the tests

before passing out the course's second examination. Subjects who

completed the experimental tests before the rest of the class were

permitted to spend the remainder of the time studying the i r books and

notes in preparation for the second course examination.

After a l l members of the class had completed the experimental

t es t , the class was to ld to wr i te down the i r comments on a blank sheet

of paper which had been provided with the tests. In addi t ion, each

subject indicated whether or not he believed that he was going to

receive the course examination which the instructor had in his hand.

The deception regarding the courses' second major examination

was eliminated immediately a f ter the col lect ion of the experimental

tests.

The Rotter Incomplete Sentences Blanks were hand scored according

to instruct ions in Rotter and Rafferty 's manual (1). Ten of the

Rotter Incomplete Sentences Blanks were randomly selected and scored

by a graduate student who served as an in te r ra te r . This graduate

student did not know which subjects had received the examination

threat and which ones were in the control group.

The Mult iple Affect Adjective Check Lists were hand scored

according to scoring prof i les obtained from the publishing companies (2).

CHAPTER BIBLIOGRAPHY

1. Rotter, J. B. and Janet E. Rafferty, Manual: The Rotter Incomplete Sentences Blank, New York: Psychological Corporation, 1950.

2. Zuckerman, M. and B. Lubin, Manual: The Multiple Affect Ad.jective Check L is t , San Diego: Educational and Industrial Testing Service, 1965.

44

CHAPTER IV

RESULTS

Evidence relevant to the four hypotheses i s provided in the

present chapter. The resu l t s of the experiment were in the form of a

before and a f t e r to ta l adjustment score obtained from the Rotter

Incomplete Sentences Blank and before and a f t e r anxiety, h o s t i l i t y , and

depression scores obtained from the Multiple Affect Adjective Check Lis t .

The basic purpose was to determine the e f f e c t of anxiety, h o s t i l i t y , and

depression on responses to the Rotter Incomplete Sentences Blank.

The scores from the f i r s t three administrations of the Multiple

Affect Adjective Check List and the Rotter Incomplete Sentences Blank

were averaged in order to obtain the following baselines or before

scores: (a) an anxiety basel ine, (b) a h o s t i l i t y basel ine, (c) a

depression basel ine, and (d) a to ta l adjustment basel ine. In cases

where the subjects were present for only two of the three baseline t e s t

administrat ions, the baseline consisted of an average of those two

administrat ions. A summary of the means and standard deviations for

each of the f i r s t three administrations of the Multiple Affect Adjective

Check List and the Rotter Incomplete Sentences B1ank and the resu l t ing

baseline is presented in Table 1.

In order to determine if the experimental and the control groups

were basical ly equivalent on the dependent variables - - anx ie ty ,

45

TABLE I--Continued

RISB

H Experimental Control Experimental Control M SD M SD M SD M SD

9.94 3.91 7.44 3.14 129.32 18.09 131.43 18.50

9.18 4.72 9.05 4.53 125.81 11.83 118.21 15.20

10.35 4.67 8.77 3.24 129.20 12.41 119.61 16.10

9.79 3.25 8.11 2.64 127.54 12.42 122.70 14.53

47

48

hos t i l i ty , depression—and the total adjustment score obtained

respectively from the Multipie Affect Adjective Check List and the

Rotter Incomplete Sentences Blank, t - t e s t s for equal N's (1) were

computed between the baseline means of the two groups. A summary of

the results is presented in Table I I .

TABLE II

MAACL AND RISB BASELINE t-TESTS

Source df t P

MAACL Anxiety Baseline 38 .9334 n .05 ns

MAACL Depression Baseline 38 1.9914 n .05 ns

MAACL Hosti l i ty Baseline 38 1.7023 n .05 ns

RISB Baseline 38 1.0674 n .05 ns

All t ' s were insignif icant (p> .05) , thus indicating that there were no

s ignif icant differences between the experimental and control group

before the treatment, in the form of an examination threa t , was applied

to the experimental group.

The f i r s t hypothesis stated that the experimental group, on the

examination threat day, would have s ignif icant ly higher anxiety,

hos t i l i t y , and depression scores than their baselines. The greatest

increase was expected on the Hostili ty scale. This increase on the Hostil-

i ty scale was hypothesized as being s ignif icant ly greater than the increase

49

on the Anxiety scale . In order to determine if the f i r s t hypothesis

was t rue , t - t e s t s fo r equal N's (1) were u t i l i zed between the means

obtained from the baseline data and those means obtained a f t e r the

examination th rea t .

TABLE III

MEAN MAACL SCORES OF EXPERIMENTAL AND CONTROL GROUPS BEFORE AND AFTER THE EXAMINATION THREAT

Anxiety Depression Host i l i ty

Condition Before After t Before After t_ Before After t

Experi-mental

Control

8.53 11.21 2.1407*

7.63 8.00 .3134

16.95 17.32 .1752

13.21 14.06 .3660

9.79 13.42 2.3979*

8.11 8.53 .3616

*p< .C 15.

Table III contains a summary of the mean Multiple Affect Adjective

Check List scores of both the experimental and control group during the

i n i t i a l baseline period and immediately a f t e r the examination th rea t .

The j t - tes ts indicated that the Anxiety scale and the Host i l i ty scales

of the Multiple Affect Adjective Check List were s ign i f i can t ly

d i f f e r e n t from i t s basel ine. Visual inspection indicated tha t the

Host i l i ty scale showed the grea tes t increase from i t s basel ine.

Two additional t - t e s t s were computed in order to determine if the

increase on the Host i l i ty scale was s ign i f i can t ly greater than the

increase on the Depression sca le , but not s ign i f i can t ly greater than

the increase on the Anxiety scale . In order to compute these two

50

t - tes ts , i t was necessary to f i r s t obtain a c r i t i ca l rat io consisting

of the differences between the mean baseline and the mean score

obtained after the examination threat for a l l three scales—Anxiety,

Host i l i t y , and Depression. The two t /s were then computed. The

difference on the Depression scale, but not s igni f icant ly more than

the difference increased on the Anxiety scale (t=2.4743, df=38, t=.9157,

df=38).

The second hypothesis stated that the control group would not

have any signif icant changes between the means of i t s baseline data and

the last administration of the Multiple Affect Adjective Check L is t .

In u t i l i z i ng t - tests for equal N's (1) in order to determine i f the

hypothesis was true, no signif icant differences were found in the

control group between the means of i t s baseline data and the last

administration of the Multiple Affect Adjective Check L is t . The means of

the Multiple Affect Adjective Check List and i t s t ' s for the control

group were presented in Table I I I .

The th i rd hypothesis stated that in the experimental group, the

total adjustment score of the Rotter Incomplete Sentences Blank on the

examination threat day would be s igni f icant ly greater than the total

adjustment score of i t s baseline. In order to determine i f this

hypothesis was true, t-tests for equal N's (1) were completed between the

mean baseline and the scores obtained after the examination threat.

Table IV contains a summary of the means of the Rotter Incomplete

Sentences Blank for both the experimental and control group during the

i n i t i a l baseline period and immediately af ter the examination threat.

The t - tes t between the before and after means was not s igni f icant , thus

51

indicating tha t no s ign i f i can t di f ferences were re f lec ted in the to ta l

adjustment score of the Rotter Incomplete Sentences Blank a f t e r the

examination th rea t .

TABLE IV

MEAN RISB SCORES OF EXPERIMENTAL AND CONTROL GROUPS BEFORE AND AFTER EXAMINATION THREAT

Rotter Incomplete Sentences Blank

Condition Before After t P

Experimental 127.5 124.1 .9113 p .05 ns

Control 122.7 117.6 .7324 p .05 ns

The fourth hypothesis s tated tha t in the control group, the mean

to ta l adjustment score obtained from the Rotter Incomplete Sentences

Blank on the examination day would not be s ign i f i can t ly d i f f e r e n t from

i t s baseline total adjustment score. In order to determine i f t h i s

hypothesis was t rue , a t> tes t fo r equal N's (1) was u t i l i zed between the

mean to ta l adjustment baseline and the l a s t obtained mean to ta l

adjustment score. No s ign i f i can t d i f ference was found in the control

group between the mean of i t s baseline and the l a s t administration of

the Rotter Incomplete Sentences Blank. The means of the Rotter

Incomplete Sentences Blank and i t s t, 's were presented in Table IV.

In order to determine how accurate the Rotter Incomplete Sentences

Blank had been scored, ten of the protocols were randomly drawn and

rescored by a graduate student . This graduate student served in the

52

capacity of an in te r ra te r , and he had no knowledge of the previous scores

nor of which protocols belonged to the experimental or control group.

A Pearson r yielded a correlation of .83 between the experimentors'

obtained total adjustment scores and the in te r ra te r s ' obtained total

adjustment scores.

CHAPTER BIBLIOGRAPHY

1. Spence, Janet T., Benton J. Underwood, Carl P. Duncan, and John W. Cotton, Elementary Statistics, New York, Appleton-Century-Crofts, 1968.

53

CHAPTER V

DISCUSSION

The present chapter consists of a comparison between what was

predicted in the form of hypotheses and what was obtained in the

form of results. Whenever possible, analogies are drawn to previous

research. The following chapter headings are ut i l ized: (a) The

Effects of an Examination Threat on Responses to the Multiple Affect

Adjective Check L is t , (b) The Effects of Anxiety and Host i l i ty on

Responses to the Rotter Incomplete Sentences Blank, and Implications

of this Study.

The Effects of an Examination Threat on Responses to the Multiple Affect Adjective Check List

One of the principal variables investigated was the effects of an

examination threat on anxiety, hos t i l i t y , and depression as measured

by the Multiple Affect Adjective Check List . As specified in Hypothesis

1_ the experimental group on the examination threat day would have

s igni f icant ly higher anxiety, hos t i l i t y , and depression scores than

their baselines. In addition the greatest increase would be on the

hos t i l i t y scale. This increase on the hos t i l i t y scale would be signi-

f icant ly greater than the increase on the depression scale, but not

s igni f icant ly greater than the increase on the anxiety scale. Studies

cited in Chapter I (3, 9, 10, 11, 12), plus a great many other related

studies, provided data and results which supported this hypothesis.

54

55

Differences between the base lines and the examination threat

indicated that in the experimental group the anxiety scale and the

hos t i l i ty scale were s iqnif icant ly d i f ferent from their baselines as

predicted. The mean differences of the depression scale,while con-

s is tent with the hypothesis,was too small to be s t a t i s t i c a l l y

s igni f icant . The hos t i l i ty scale showed the greatest increase of the

three scales. As predicted, this increase on the hos t i l i ty scale was

s ignif icant ly greater than the increase on the depression scale, but

not s ignif icant ly greater than the increase on the anxiety scale.

From a practical point of view,the s t a t i s t i c a l analysis of this

hypothesis confirmed the reactions of the subjects as observed by the

instructor of the class . Each subject used in the experiment indicated

on his opinion sheet the belief that he was going to receive the

course examination which the instructor held in his hands. In addition,

the roar of protest from the students which ensued a f t e r the announcement

of the unexpected examination indicated f i r s t a reaction of hos t i l i ty

towards the instructor for being unfair and unpredictable and second

as the realization of unpreparedness occurred, anxiety due to expected

fa i lure and the lowering of their previous grade point average.

In order to ascertain that the results predicted in Hypothesis 1

were in e f f ec t due to the experimental manipulation and not to an

extraneous variable, such as the number of times the Multiple Affect

Adjective Check List had been completed, a control group was u t i l i zed .

As stated in Hypothesis 2 the control group was not expected to d i f f e r

s ignif icant ly between i t s baselines and the l as t administration of the

Multiple Affect Adjective Check Lis t . As predicted, the control group

56

did not d i f f e r s ign i f i can t ly between i t s baselines and the l a s t

administration of the Multiple Affect Adjective Check Lis t . These

resu l t s support past studies usinq the Multiple Affect Adjective Check

List in which control qroups were not u t i l i z ed . The implication is

that control qroups were not used then and need not be used in the

fu ture when usinq the Multiple Affect Adjective Check Lis t in empirical

inves t iqa t ions . Data relevant to th i s variable in re la t ion to

Hypotheses 1 and 2_ can be found in Tables I and I I I .

The Effects of Anxiety and Host i l i ty on Responses to the Rotter Incomplete Sentences Blank

The e f f e c t of anxiety and h o s t i l i t y on responses to the Rotter

Incomplete Sentences Blank was the second variable invest igated. As

was s ta ted e a r l i e r the experimental group was given an examination

threat which s iqn i f i can t ly raised the i r scores on the anxiety and

h o s t i l i t y scales of the Multiple Affect Adjective Check L i s t . The

question asked was: Will th is increase in anxiety and h o s t i l i t y

s iqn i f i can t ly ra i se the to ta l adjustment score obtained from the Rotter

Incomplete Sentences Blank? As stated in Hypothesis 3 i t was expected

that the to ta l adjustment score of the Rotter Incomplete Sentences Blank

on the examination threa t day would be s iqn i f i can t ly qreater than i t s

previous basel ine. Contrary to what was expected,the to ta l adjustment

score of the Rotter Incomplete Sentences Blank did not s iqn i f i can t ly

increase, but showed a trend in the opposite d i rec t ion . Consequently,

one may conclude that temporary anxiety and h o s t i l i t y did not s i q n i f i -

cantly influence responses to the Rotter Incomplete Sentences Blank.

These resu l t s althouqh contrary to what one would expect from the

57

results of Churchill and Crandall (1) are in accordance with those

results found by other investigators (2, 4, 6, 7, 8).

Windle (8) stated that there was a tendency toward better

adjustment on retest of the Rotter Incomplete Sentences Blank when the

intervals between the test had been less than two months. Windle (8)

suggested that this phenomena of appearing better adjusted on retests

was due to the following factors: (a) the pull of social norms plus

the gaining of greater insight into the nature of the test , (b) ageing,

(c) arousal and loss of interest, and (d) increased relaxation of

hos t i l i t y .

The fact that test-retest may be influenced by the pull toward

social norm is supported by Morton (4). His evaluation of br ief

psychotherapy showed a mean decrease at the .001 level on the Rotter

Incomplete Sentences Blank total adjustment score for subjects who had

undergone therapy. In addition, subjects in his control group who had

not undergone therapy also showed a mean decrease in their total

adjustment scores which was signif icant at the .05 level.

Another investigator, Fisher (2), concluded that projective

techniques which included the Rorschach, The Thermatic Apperception Test,

a Figure Drawing, and Word Association were not s igni f icant ly sensitive

to the effects of embarrassment and anxiety induced by the experimental

subjects. Consequently i f generalization occurs, one may question the

sensi t iv i ty of the Rotter Incomplete Sentences Blank to anxiety and

depression. Stephens (6), however, indicated that the probabil i ty of

the Rotter Incomplete Sentences Blank being sensitive to anxiety and

hos t i l i t y is s l ight . He stated that the ins tab i l i t y of the adjustment

58

scores for the Rotter Incomplete Sentences Blank was due to certain

stems in the blank which varied directly with adjustment and the judges

interpretation of adjustment. Consequently, one may conclude that

although t e s t - r e t e s t chanqes occur in responses to the Rotter Incomplete

Sentences Blank, these changes are res t r ic ted to only a few stems which

generally do not a f fec t the overall total adjustment score to a

s ignif icant degree.

In order to ascertain that the results obtained with respect to

Hypothesis 3 v/ere not due to an extraneous variable, a control grouo was

u t i l ized . Hypothesis 4 had predicted that no s ignif icant changes would

occur with respect to the total adjustment score obtained on the l a s t

administration of the Rotter Incomplete Sentences Blank and i t s previous

baseline. This hypothesis was confirmed although a trend was noted

toward social norms. These results indicated that the control group

which did not experience an increase in anxiety and depression obtained

similar total adjustment scores to those of their baselines. Data

relevant to Hypotheses 3 and £ are presented in Table 4. Since both the

experimental and control groups remained relat ively stable with respect

to their baselines, one may conclude that the total adjustment score of

the Rotter Incomplete Sentences Blank is not s ignif icant ly influenced by

temporary a f fec t s tates of anxiety and hos t i l i t y .

Implications of Present Study

There are two primary implications which have been generated from

the results of this study. The f i r s t concerns the Rotter Incomplete

Sentences Plank with regards to therapy, and the second refers to the

59

necessary t ra in ing of scorers of the Rotter Incomplete Sentences Blank.

The Rotter Incomplete Sentences Blank is on the whole a stable

instrument which is a good indicator of pre and post therapeutic changes

when the interval between test and retest is greater than two months.

I f the blank is used consecutively at weekly in te rva ls , one is l i ke l y

to obtain adjustment scores which regress to the socia l ly accepted

norms. This phenomena may be corrected s t a t i s t i c a l l y by adopting a

higher signif icance level when doing experimental research.

With respect to i n i t i a l interviews, the Rotter Incomplete Sentences

Blank is unl ikely to be affected by temporary af fect ive states of

anxiety or h o s t i l i t y . However, i f the states are more permanent or long

last ing then the adjustment score w i l l be s ign i f i can t l y influenced.

Consequently, whatever contributed to the maladjustment score obtained

from the Rotter Incomplete Sentences Blank w i l l generally be more than

of a temporary nature. Extreme caution must be exercised, however, to

determine the primary factors leading to maladjustment and those which

are a by-product and of a temporary nature. Consequently, i t is wiser i f

the Rotter Incomplete Sentence Blank be u t i l i zed along with other tests

in a battery.

The moderately high in te r - ra ter scorer correlations indicated that

graduate students in psychology who have trained themselves by

fol lowing Rotter and Rafferty 's Manual (5) may be reasonably successful

in the scoring of Rotter Incomplete Sentences Blank protocols. This

idea is supported by Walker and Linden (7) who had judges of four

levels of psychological sophist ication successfully u t i l i z e sentence com-

plet ion data. "Success was defined in terms of consistency and accuracy

of judgement."

CHAPTER BIBLIOGRAPHY

1. Churchi l l , Ruth and V. J . Crandal l , "The R e l i a b i l i t y and Val id i ty of the Rotter Incomplete Sentences Tes t , " Journal of Consulting Psychology, XIX, (October, 1955), 345-350.

2. F i sher , Rhoda Lee, "The E f f e c t of a Disturbing S i tua t ion Upon the S t a b i l i t y of Various P ro jec t ive Tes t s , " Psychological Monograph, LXXXXII, (#467, 1958), 1-23.

3. Lieberman, L. R., "Ef fec t s Upon Anxiety, Depression, and H o s t i l i t y of Postponement of an Examination," Perceptual and Motor S k i l l s , XXIII, (December, 1966), 1051-1054.

4. Morton, Robert B. "An Experiment in Brief Psychotherapy," Psychological Monograph, LXXXIX, (January, 1955), 1-17.

5. Ro t t e r , J . B. and Janet E. R a f f e r t y , Manual: The Rotter Incomplete Sentences Blank, New York: Psychological Corporat ion, 1950.

6. Stephens, Mark W., "The Incomplete Sentences Blank: Sources of Variance in Retest R e l i a b i l i t y , " Journal of Cl inical Psychology, XVI, ( Ju ly , 1960), 331-333.

7. Walker, C. Eugene and James D. Linden, "Varying Degrees of Psychological Sophis t ica t ion in the I n t e r p r e t a t i o n of Sentence Completion Data," Journal of Cl in ica l Psychology, XXIII, (Apr i l , 1967), 229-231.

8. Windle, C. , "Further Studies of Tes t -Retes t E f f e c t on Persona l i ty Questional ' res," Educational and Psychological Measurement, XVI, (Autumn, 1955), 246-253.

9. Winter, W. D., A. J . F e r r e i r a , and R. Ranson, "Two Measures of Anxiety: A Val ida t ion ," Journal of Consulting Psychology, XXVII, (December, 1963), 520-524.

10. Zuckerman, Marvin, "The Development of an E f f e c t Adject ive Check L i s t f o r Measurement of Anxiety," Journal of Consulting Psychology, XXIV, (October, 1960), 457-462.

11. Zuckerman, Marvin and D. V. Biase, "Replication and Further Data on the Af fec t Adject ive Check L i s t Measure of Anxiety," Journal of Consulting Psychology, XXVI, (June, 1962), 291.

60

12. Zuckerman, Marvin, Bernard Lubin, L. Vogel, and E. V a l e r i u s , "Measurement of Experimental ly Induced E f f e c t s , " Journal of Consul t ing Psychology, XXVIII, (October , 1964), 4Tfo-42i>.

61

CHAPTER VI

SUMMARY AND CONCLUSIONS

The primary purpose of this experiment was to determine the effect

of anxiety, hos t i l i t y , and depression on responses to the Rotter

Incomplete Sentences Blank.

The subjects were th i r ty-e ight students, sixteen females and

twenty-two males, who were enrolled in General Psychology I . In

addit ion,al l subjects were enrolled in one of the two General

Psychology I sections which met from 12:00 - 1:00 PM on Monday,

Wednesday, and Friday and had completed a minimum of two of the three

tests used in computing the baseline and the f ina l test .

The Rotter Incomplete Sentences Blank and the Multiple Affect

Adjective Check List were administered four times with an interval

of one week elapsing between each administration. The f i r s t three

administrations were averaged and used as a baseline. The fourth

administration occurred after the experimental group was given an

examination threat.

The hypotheses were as follows:

Hypothesis 1 stated that the experimental group would show an

increase in anxiety, hos t i l i t y , and depression relat ive to thei r

baselines. This hypothesis was accepted with respect to anxiety

and hos t i l i t y , but rejected with respect to depression. In addition,

as hypothesized hos t i l i t y showed the greatest increase. This increase

was signi f icant ly greater than the increase on the Depression Scale,

but not s igni f icant ly greater than the increase on the Anxiety Scale.

Hypothesis 2 predicted that the control group would not d i f fe r

62

63

signif icant ly from i t s baselines. This hypothesis was accepted.

Hypothesis 3 stated that , in the experimental group, the total

adjustment score of the Rotter Incomplete Sentences Blank on the

examination threat day would be s ignif icant ly greater than the total

adjustment score of i t s baseline. This hypothesis was rejected.

Hypothesis 4 predicted that in the control group, the total

adjustment score of the Rotter Incomplete Sentences Blank on the

examination threat day would not be s ignif icant ly d i f fe ren t than the

total adjustment score of i t s ' baseline. This hypothesis was accepted.

Hypotheses, resul ts obtained, and their implications with regard

to therapy, and the amount of training necessary for the scoring of

Rotter Incomplete Sentences Blank protocols were discussed.

This study concluded that the evidence appeared to favor the

s t ab i l i t y of the Rotter Incomplete Sentences Blank and gave fur ther

support to two of the Multiple Affect Adjective Check List scales.

APPENDIX A 64

INCOMPLETE SENTENCES BLANK - COLLEGE FORM

Name Sex Age Marital Status

School Class Date

Complete these sentences to express your real feelings. Try to do every one.

Be sure to make a complete sentence.

1. I like

2. The happiest time

3. I want to know ....

4. Back home

5. I regret

6. At bedtime

7. Boys

8. The best

9. What annoys me ..

10. People

11. A mother

12. I feel

13. My greatest fear ...

14. In high school

15. I can't

(?) (TURN PAGE OVER AND CONTINUE)

Copyright 1950. All rights reserved as stated in the test manual and Catalog. The Psychological Corporation, 304 East 45th Street, New York, N. Y. 10017

Printed in U.S.A. 65-152 AS JULIAN B. ROTTER, AUTHOR

16. Sports

17. When I was a child .

18. My nerves

19. Other people

20. I suffer

21. I failed :

22. Reading

23. My mind

24. The future

25. I need

26. Marriage

27. I am best when

28. Sometimes

29. What pains me

30. I hate

31. This school

32. I am very

33. The only trouble

34. I wish

35. My father

36. I secretly

37. I

38. Dancing

39. My greatest worry is

40. Most girls

APPENDIX B V 65

TODAYfORM

By Marvin Zuckerman

Bernard Lubin

Name Age Sex.

Date Highest grade completed in school.

DIRECTIONS: On this sheet you will find words which describe different

kinds of moods and feelings. Mark an El in the boxes beside the words

which describe how you feel now - today. Some of the words may sound

alike, but we want you to check all the words that describe your feelings.

Work rapidly.

• v . \ ;

; „ » j J ; '•• / >;

r- ;l~ r'i i J? Li ; i i ' i i ; ^ i I ' i : •'; BOX 72S4, SAN OIE<?Q, CALIFORNIA:

. « : J '!• : ' V •!'i • t ? ' / - ' ; 0 -\ j ffpPWu%tT. ©' -tSSS^tOUB'^TiQN^^lW/tl^ytTftlAL^-^TlNS sglitvrcfc;/;

'' * ; : ' ; •' * * } <\ ' "-it; \ i ' ; •' •? ' ^ ^ ,< '' -Mir ilSHf# jMESWffiD, i' • i ' ~ •

• r ;•: t)4i>.i"«•?i•'»•> • ? * : >;'-- : . • •

w>* ^ s ^? i 41\"? ; s " > ^ ' * - v > < > > \ ^ } >'; ,v r . i ' ' ; - <'

1 • active 45 • fit 89 0 peaceful

2 • adventurous 46 • forlorn 90 0 pleased

3 0 affectionate 47 0 frank 91 0 pleasant

4 • afraid 48 0 free 92 0 polite

5 • agitated 49 0 friendly 93 0 powerful

6 • agreeable 50 0 frightened 94 0 quiet

7 • aggressive 51 0 furious 95 0 reckless

8 • alive 52 Ogay 96 0 rejected

9 • alone 53 0 gentle 97 0 rough

10 • amiable 54 0 glad 98 0 sad

11 • amused 55 0 gloomy 99 0 safe

12 • angry 56 0 good 100 0 satisfied

13 • annoyed 57 0 good-natured 101 0 secure

14 • awful 58 0 grim 102 0 shaky

15 • bashful 59 0 happy 103 0 shy

16 • bitter 60 0 healthy 104 0 soothed

17 • blue 61 0 hopeless 105 0 steady

18 • bored 62 0 hostile 106 0 stubborn

19 • calm 63 0 impatient 107 0 stormy

20 • cautious 64 0 incensed 108 0 strong

21 • cheerful 65 0 indignant 109 0 suffering

22 • clean 66 0 inspired 110 0 sullen

23 • complaining 67 0 interested 111 0 sunk

24 • contented 68 0 irritated 112 0 sympathetic

25 • contrary 69 0 jealous 113 0 tame

26 • cool 70 0 joyful 114 0 tender

27 • cooperative 71 0 kindly 115 0 tense

28 • critical 72 0 lonely 116 0 terrible

29 • cross 73 0 lost 117 0 terrified

30 • cruel 74 0 loving 118 0 thoughtful

31 • daring 75 0 low 119 0 timid

32 • desperate 76 0 lucky 120 0 tormented

33 Q destroyed 77 0 mad 121 0 understanding

34 • devoted 78 0 mean 122 0 unhappy

35 • disagreeable 79 0 meek 123 0 unsociable

36 • discontented 80 0 merry 124 0 upset

37 • discouraged 81 0 mild 125 0 vexed

38 • disgusted 82 0 miserable 126 0 warm

39 • displeased 83 0 nervous 127 0 whole

40 • energetic 84 0 obliging 128 0 wild

41 • enraged 85 0 offended 129 0 willful

42 • enthusiastic 86 0 outraged 130 0 wilted

43 • fearful 87 0 panicky 131 0 worrying r»o r—I x* J. inn I—I

BIBLIOGRAPHY

Books

1. Rotter, Julian B., "Word Association and Sentence Completion Methods," in H. H. Anderson and G. L. Anderson (Eds.), An_ Introduction to Projective Techniques, New York, Prentice -..Hall, 1951, 279-311.

2. Rotter, J. B. and Janet E. Rafferty, Manual: The Rotter Incomplete Sentences Blank, New York: Psychological Corporation, 1950.

3. Spence, Janet T., Benton J. Underwood, Carl P. Duncan, and John W. Cotton, Elementary Stat is t ics , New York, Appleton - Century - Crofts, 1968.

4. Sundberg, Norman D. and Leona E. Tyler, Clinical Psychology, New York, Appleton - Century - Crofts, 1962.

5. Wheeler, D. R., "Imaginal Productivity Tests," in H. A. Murray, Exploration in Personality, New York, Oxford University Press, 1938, 545-550; 680-684.

6. Zuckerman, M. and B. Lubin, Manual: The Multiple Affect Adjective Check L is t , San Diego: Educational and Industrial Testing Service, 1965.

Articles

1. Arnold, Frank C. and Verne A. Walter, "The Relationship Between a Self and Other - Reference Sentence Completion Test," Journal of Counseling Psychology, IV, (Spring, 1957), 65-70.

2. Bendig, A. W. and Gail Bruder, "The Effect of Repeated Testing on Anxiety Scale Scores," Journal of Consulting Psychology, XXVI, (August, 1962) 392.

3. Berg, Wilbert Arthur, "Determining Val idi ty of the Incomplete Sentences Blank through Appraisal of Qualitative Interpretations, Dissertation Abstracts, XXII, (January, 1952), 504-505.

4. B ier i , James, Edward Blacharsky, and J. William Reid, "Predictive Behavior and Personal Adjustment," Journal of Consulting Psychology, XIX, (October, 1955), 351-356.

66

67

5. Bourne, P. G., W. M. Coli, and W. E. Date!, "Affect Levels of Ten Special Forces Soldiers Under Threat of Attack." Psychological Reports, XXII, (Apr i l , 1968), 363-366.

6. Bourne, P. G., W. M. Coli, and W. E. Date!, "Anxiety Levels of Six Helicopter Ambulance Medics in a Combat Zone," Psychological Reports, XIX, (December, 1966), 821-822.

7. Cameron, N., "Reasoning, regression, and communication in schizophrenia," Psychological Monograph, L, (whole #221, 1938), 1-34.

8. Churchil l, Ruth and V. J. Crandall, "The Rel iab i l i ty and Val id i ty of the Rotter Incomplete Sentences Test," Journal of Consulting Psychology, XIX, (October, 1955), 345-350.

9. Crenshaw, David, Suzanne Bohn, Marlene R. Hoffman, John M. Matheus, and Stefan G. Offenbach, "The Use of Projective Methods in Research: 1947-1965," Journal of Projective Techniques and Personality Assessment, XXXII, TFebruary, 1968), 3-9.

10. Datel, W. E. and E. 0. Engle, "Affect Levels in Another Platoon of Basic Trainees," Psychological Reports, XIX, (October, 1966), 407-412.

11. Datel, W. E., E. 0. Engle, and M. A. Barba, "Affect Levels in a Company of Basic Trainees," Psychological Reports, XIX, (December, 1966), 903-909.

12. Datel, W. E., C. F. Gieseking, E. 0. Engle, and M. J. Dougher, "Affect Levels in a Platoon of Basic Trainees," Psychological Reports, XVIII, (February, 1966), 271-285.

13. Denenberg, Victor H., "The Relationship Between a Measure of Kinesthesis and Two Indices of Adjustment," Journal of General Psychology, LXII, (January, 1960), 43-52.

14. Fisher, Rhoda Lee, "The Effect of a Disturbing Situation Upon the Stabi l i ty of Various Projective Tests," Psychological Monograph (#467, 1958), 1-23.

15. Fiske, Donald W. and Charles Van Buskirk, "The Stab i l i ty of Interpretations of Sentence Completion Tests," Journal of Consulting Psychology, XXIII, (Apr i l , 1959), 177-180.

16. Fitzgerald, Barnard Joseph, "The Relationship of Two Projective Measures of a Sociometric Measure of Dependent Behavior," Dissertation Abstracts, XX, (December, 1959), 2380-2381.

68

17. Folkins, C. H., K. D. Lawson, E. M. Opton, J r . , arid R. S. Lazarus, "Desensitizatiori and the Experimental Reduction of Threat," Journal of Abnormal Psychology, LXXIII, (Apr i l , 1968), 100-113.

18. Gardner, James M., "The Adjustment of Drug Addicts as Measured by the Sentence Completion Test," Journal of Projective Techniques and Personality Assessment, XXXI, (June, 1967), 28-29.

19. Geer, J. H. and A. Turteltaub, "Fear Reduction Following Observation of a Model," Journal of Personality and Social Psychology, VI, (July, 1967)7 327-331.

20. Getter, Herbert and Stephan D. Weiss, "The Rotter Incomplete Sentences Blank Adjustment Score as an Indicator of Somatic Complaint Frequency," Journal of Projective Techniques and Personality Assessment, XXXII,"TJune, 1968), 266.

21. Giddon, Donald B., "Individual Differences in Average Affect Adjective Check List (AACL) Scores: Their Relation to Psychological and Physiological Variables," Psychological Reports, XIV, (Apr i l , 1964), 541-542.

22. Goldberg, Ph i l l i p , "The Current Status of Sentence Completion Methods," Journal of Projective Techniques and Personalit.y Assessment, XXXII(June, 1968), 215-221.

23. Goldberg, Ph i l l ip and Milton J. Stark, "Johnson or Goldwater?: Some Personality and Attitude Correlates of Pol i t ical Choice," Psychological Reports, XVII, (October, 1951), 627-631.

24. Goldstein, H., "Placebo, Psychotherapy, and Change in Anxiety, Mood and Adjustment," Dissertation Abstracts, XXVI, (September, October, 1965), 1775-1776.

25. Guertin, Wilson H., "An Analysis of Gross Errors on a Sentence Completion Test," Journal of Clinical Psychology, XV (October, 1959), 415-416.

26. Harrow, M., J. Colbert, T. Detre, and R. Bakeman, "Symptomatology and Subjective Depression in Current Depressive States," Archives of General Psychiatry, XIV, (February, 1966), 203-212.

27. Hankoff, L. D., L. Rudorfer, and H. M. Paley, "A Reference Study of Ataraxics: A Two Week Double Blind Outpatient Evaluation," Journal of New Drugs, I I , (May, June, 1962), 173-178.

28. Hankoff, L. D., L. Rudorfer, and H. M. Paley, "The Psychiatric Use of Pyrbenziondole: A Double Blind Out-Patients Study," Journal of New Drugs, I I , (May, June, 1962), 167-172.

69

29. Hayes, Carol V., "The Measurement of Anxiety in Sophomore Nursing Students Using Zuckerman's AACL," Nursing Research, XV, (Summer, 1966), 262-267.

30. Herron, E. Wayne, Lewis Bernstein, and Harold Rosen, "Psychometric Analysis of the Multiple Affect Adjective Check L is t : MAACL Today," American Psychologist, XXII, (July, 1967), 591.

31. Holzberg, J . , A. Teicher, and J. L. Taylor, "Contributions of Clinical Psychology to Mi l i tary Neuropsychiatry in an Army Psychiatric Hospital," Journal of Clinical Psychology, I I I , (January, 1947), 84-95.

32. Hutt, M. L. , "The Use of Projective Methods of Personality Measurements in Army Medical Instal lat ions," Journal of Clinical Psychology, I , (Apr i l , 1945), 134-140.

33. Johnson, D. T. "Effects of Interview Stress on Measures of State and Trait Anxiety," Journal of Abnormal Psychology, LXXIII, (June, 1968), 245-251.

34. JOhnson, D. T. and C. D. Spiel bergen, "The Effects of Relaxation Training and the Passage of Time in Measures of State and Trait Anxiety," Journal of Clinical Psychology, XXIV, (January, 1968), 20-23.

35. Kennedy, Wallace A., Ted Cot t re l l , and A1 Smith, "Norms of Gifted Adolescents on the Rotter Incomplete Sentences Blank," Journal of Clinical Psychology, XIX, (July, 1963), 314-316.

36. Knapp, R. R., W. S. Zimmerman, W. S. Roscoe, and W. B. Michael, "The Effects of College Entrance Examinations on Effects of Anxiety, Depression, and Host i l i t y , " Educational and Psychological Measurement, XXVII, (Winter, 1967), 1121-1126.

37. Lev i t t , E. E., A. DenBreeijen, and H. Persky, "The Induction of Clinical Anxiety by Means of a Standardized Hypnotic Technique," American Journal of Clinical Hypnosis, I I , (Apr i l , 1960), 206-214.

38. Lieberman, L. R., "Effects Upon Anxiety, Depression, and Host i l i ty of Postponement of an Examination," Perceptual and Motor Sk i l l s , XXIII, (December, 1966), 1051-1054,

39. Lorges, I . and E. L. Thorndike, "The Value of the Responses in a Completion Test as Indicators of Personal Trai ts ," Journal of Applied Psychology, XXV, (Apr i l , 1941), 191-199.

40. Lubin, B., "A Modified Version of the Self - Disclosure Inventory," Psychological Reports, XVII, (October, 1965), 498.

70

41. Lubin, Bernard and Marvin Zuckerman, "Affective and Perceptual -Cognitive Pattern in Sensit ivi ty Training Groups," Psychological Reports, XXI, (October, 1967), 365-376.

42. Mi l l s , David H., "The Research Use of Projective Techniques: A Seventeen Year Survey," Journal of Projective Techniques and Personality Assessment, XXIX, (December, 1965), 513-515.

43. Morton, Robert B., "An Experiment in Brief Psychotherapy," Psycholoqical Monographs, LXIX, (January, 1955), 1-17.

44. Neuringer, Charles and Perry 0. Orwick, "The Measurement of Anxiety on the Sentence Completion Test," Journal of General Psychology, LXVIII, (Apr i l , 1968), 197-207.

45. Renner, Edward, Brendan A. Maher, and Donald T. Campbell, "The Val idi ty of a Method for Scoring Sentence Completion Responses for Anxiety, Dependency, and Host i l i t y , " Journal of Applied Psychology, XXXXVI, (August, 1962), 285-290.

46. Rhodes, A. R., "Explorations in Personality by the Sentence Completion Method," Journal of Applied Psychology, XXX, (Apr i l , 1946), 169-181.

47. Rotter, Julian B., Janet E. Rafferty, and Antoinette B. Lotsof, "The Val idi ty of the Rotter Incomplete Sentences Blank: High School Norm," Journal of Consulting Psychology, XVIII, (Apr i l , 1954), 105-111.

49. Rotter, Julian B., Janet E. Rafferty and Eva Schachtitz, "Validation of the Rotter Incomplete Sentences Blank for College Screening," Journal of Consulting Psychology, X I I I , (October, 1949), 348-356.

50. Rotter, J. B. and B. WiHerman, "The Incomplete Sentences Test," Journal of Consulting Psychology, XI, (January, 1947), 43-48.

51. Sanford, R. N., "Physique, Personality, and Scholarship," Monogram of Social Research and Child Developments, V I I I , (January, 1943), 1.

52. Shor, J . , "Report on a Verbal Projective Technique," Journal of Clinical Psychology, I I , (July, 1946), 279-282.

53. Siipola, Elsa M., "Incongruence of Sentence Completion Under Time Pressure and Freedom," Journal of Projective Techniques and Personality Assessment, XXXII, "(December, 1968), 562-570.

54. Stephens, Mark W., "The Incomplete Sentences Blank: Sources of Variance in Retest Re l iab i l i t y , " Journal of Clinical Psychology, XVI, (July, 1960), 33 • 333.

71

55. Tendler , A. D., "A Preliminary Report on a Test f o r Emotional In s igh t , " Journal of Applied Psychology, XIV, (Apr i l , 1930), 123-136.

56. Tempone, Vincent J . and Wesley Lamb, "Repression - Sens i t i z a t i on and i t s Relation to Measures of Adjustment and C o n f l i c t , " Journal of Consulting Psychology, XXXI, (Apr i l , 1967), 131-136.

57. Walker, C. Eugene and James D. Linden, "Varying Degrees of Psychological Sophis t ica t ion in the I n t e r p r e t a t i o n of Sentence Completion Data," Journal of Cl inical Psychology, XXIII, (Apr i l , 1967), 229-231.

58. Weaver, T. T . , J r . , "Ef fec t s of Pos i t ive and Negative Persona l i ty Evaluations of the Self Concepts of High School Seniors ," Disse r ta t ion Abs t rac t s , XXVI, (September, October, 1965), 1785-1786.

59. Williams, A. F . , "Social Drinking, Anxiety, and DepEession," Journal of Personal i ty and Social Psychology, I I I , (June, 1966), 689-693.

60. Windle, C. "Further Studies of Test - Retest Ef fec t .on Persona l i ty Ques t iona i res , " Education and Psychological Measurement, XV, (Autumn, 1955), 246-253.

61. Winter, W. D., A. J . F e r r e i r a , and R. Ranson, "Two Measures of Anxiety: A Val ida t ion . Journal of Consulting Psychology, XXVII, (December, 1963), 520-524.

62. Zuckerman, M. "The Development of an Af fec t Adject ive Check L i s t f o r Measurement of Anxiety," Journal of Consulting Psychology, XXIV, (October, 1960), 457-462.

63. Zuckerman, M. and D. V. Biase, "Replicat ion and Further Data on the Affec t Adject ive Check Lis t Measure of Anxiety," Journal of Consulting Psychology, XXVI, (June, 1962), 291.

64. Zuckerman, Marvin, Bernard Lubin, and Sidney Robins, "Validation of the Multiple Affec t Adject ive Check L i s t in Cl inical S i t u a t i o n s , " Journal of Consulting Psychology, XXIX, (December, 1965), 594.

65. Zuckerman, M., E. A. Kolin, L. Pine, and I . Zoob, "Development of a Sensation - Seeking Sca le , " Journal of Consulting Psychology, XXVIII, (December, 1964), 477-482.

66. Zuckerman, M., B. Lubin, L. Vogel, and E. Valer ius , "Measurement of Experimentally Induced A f f e c t s , " Journal of Consulting Psychology, XXVIII, (October, 1964), 418-425.

72

67. Zuckerman, M., H. Persky, and K. E. Link, "The Relation of Mood and Hypnotizability: An Illustration of the Importance of the State vs Trait Distinction," Journal of Consulting Psycholoqy, XXXI, (October, 1967), 464-470.