This article was downloaded by: [University of Dayton] On: 29 December 2014, At: 08:16 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Personality Assessment Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/hjpa20

Gender Bias on the Thematic Apperception Test Frances T. Worchel , Lana L. Aaron & Deanne F. Yates Published online: 22 Jun 2011.

To cite this article: Frances T. Worchel , Lana L. Aaron & Deanne F. Yates (1990) Gender Bias on the Thematic Apperception Test, Journal of Personality Assessment, 55:3-4, 593-602, DOI: 10.1080/00223891.1990.9674093 To link to this article: http://dx.doi.org/10.1080/00223891.1990.9674093

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

JOURNAL OF PERSONALITY ASSESSMENT, 1990, 55(3&4),593-602 Copyright o 1990, Lawrence Erlbaum Associates, Inc.

Gender Bias on the Thematic Apperception Test Frances T. Worchel, Lana L. Aaron, and Deanna F. Yates Downloaded by [University of Dayton] at 08:16 29 December 2014

Texas A&M University

Thirty male and 30 female college students generated stories to a set of nine Thematic Apperception Test (TAT) cards with either male or female stimulus characters. The Fine scoring system was factor analyzed to produce seven factor clusters plus three outcome variables. Multiple analysis of variance (MANOVA) revealed a main effect for type of card. The female TAT cards elicited significantly more responses on the General Concerns scale than did the male TAT cards. A main effect for subject gender showed the females gave more responses than males on the Interpersonal Relations scale. No differences were found on the outcome variables.

T h e Thematic Apperception Test (TAT) remains one of the most important techniques used for clinical and research purposes ( K a ~ l a n& Saccuzzo, 1989) despite its questionable psychometric properties (Worchel & Dupree, 1990). O n e area that has been widely yet inconclusively researched involves gender differences o n the TAT. Murstein (1963, 1965) found that women gave more positive stories with more positive endings. May (1975) found sex linked patterns in "deprivation/enhancement" themes. Pollack and Gilligan (1982) evaluated intimacy and concluded that men were more likely t o fear intimacy, as manilfested in more frequent themes of violence o n TAT cards with interpersonal1 picture cues. McAdams, Lester, Brand, McNamara, and Lensky (1988) ex:panded o n the Pollack and Gilligan study, confirming that women had a greater intimacy motivation than men, but failed to confirm the increase i n violent themes by men. In a review of 25 years of research regarding affiliation o n the TAT, Stewart a n d Chester (1982) concluded that sex differences were inconclusive. T h e majority of studies they reviewed involved men producing stories to TAT pictures of men and women producing stories t o TAT pictures of women, thus

Downloaded by [University of Dayton] at 08:16 29 December 2014

594

WORCHEL, AARON, YATES

confounding subject gender and type of card. There may be some gender differences on the TAT due to the fact that males and females actually differ on some personality traits. However, gender bias may be operational to the extent that the alternate sets of cards elicit differential responding. Bias denotes prejudical judgment against some group based on confounded data (Palmer, Olivarez, Willson, & Fordyce, 1989). If the TAT were to result in systematic errors in evaluating either males or females, then gender bias might be operational. For example, it might be speculated that the male cards include more content suggestive of aggression (e.g., a gun and a man being grabbed from behind). Thus, males might systematically receive higher scores for aggression (or some other variable) due to the nature of the stimulus cards. Bias such as this would be problematic on two levels. One, research findings based on the TAT might be used to perpetuate stereotypic notions about personality differences between males and females. Two, individual evaluations might lead to inaccurate decisions or treatment recommendations. Our study was designed to evaluate whether gender bias is introduced due to the male and female TAT stimulus cards. In order not to confound subject gender and gender of the stimulus cards, both of these variables were used as factors. The following questions were posited:

1. Do male and female TAT cards elicit differential responses? 2. Is there an interaction between subject gender and stimulus card?

METHOD Subjects The subjects were 60 undergraduate students, 30 males and 30 females, solicited from introductory courses in psychology. Subjects received extra class credit for volunteering to participate in the study.

Procedure The design of the study was a 2 x 2 factorial. Factors were Gender of Subject (male and female) and Type of Card (male cards and female cards). Subjects were stratified by gender and then randomly assigned to the type of card condition. Subjects were individually administered nine TAT cards chosen from Murray's (1943) original test. All cards with parallel male/female forms were utilized. Subjects were given 30 min to generate stories for the nine cards. Responses were tape recorded. Debriefing included full disclosure of the purpose of the study. In sequence, the pictures presented were:

GENDER BIAS ON THE TAT

59!5

FEMALE CARDS

Downloaded by [University of Dayton] at 08:16 29 December 2014

A woman shielding her face as she enters a doorway. A woman turned around in her chair looking at a man. A woman reading to a young girl as the girl looks off into the distance. A woman sitting in a chair and gazing off into the distance. A woman in the foreground standing behind a tree with a woman running in the background. The faces of a younger woman and an older woman. A woman walking up a long flight of stairs. 17GF A woman leaning over a bridge as a group of men unload a boat in thLe water below her. 18GF A woman holding another woman around the neck at the bottom of a flight of stairs.

MALE CARDS A person sitting on the floor leaning over the edge of a couch. O n the floor beside the person is a revolver. A woman standing with her back to a younger man. The faces of an older man and a younger man. A young man standing in the foreground while behind him a surgical operation is occurring. Four men lying together on the grass. A young man lying on a couch with an older man standing above hirn. A little boy sitting on the doorstep of a log cabin. A man clinging to a rope. A man being grabbed from behind.

TAT Scoring The Fine scoring system (Shneidman, 1951) was chosen based on the following criteria: (a) a broad range of content areas, (b) content areas that miglht traditionally be ~erceivedas either masculine (e.g., physical hostility) or ferninine (e.g., submission) in nature, (c) a reasonable balance between difficulty in scoring and amount of information obtained, and (d) sufficient structure to allow for high interrater reliability. The Fine system contains 33 affective categories, 3 outcome categories, and 12 interpersonal relationship categories. Based on an evaluation of several other scoring techniques, the following content categories were added to the Fine system: jealousy, need to achieve, ambivalence, fear, and nonhostile death. Table 1 lists the content categories utilized.

TABLE 1 Content Categories for the Fine TAT Scoring System

Downloaded by [University of Dayton] at 08:16 29 December 2014

Name Affection, verbala Affection, Anger Anxiety Compulsiona Conflict Crime Depression Effort Escape Excitementa Frustrationa Guilta Hostility: a. Verbala b. Physicala c. Death" Management of: 1. Externalized" 2. Internalized" 3. Repressed" 4. Displaced" 5. Punished" Inferioritya Loneliness Paina Path indicators: a. Tabooed activitya b. Bizarre verbala c. Rejection of card" Pleasure Possessivenessa Sexuality Suicide Superiority Wishful thinking Jealousyb Need to achieveb ~mbivalence~ Fearb Death-nonhostilityb Outcomes: a. Favorable b. Unfavorable c. Indeterminate

GENDER BIAS ON THE TAT

597

Downloaded by [University of Dayton] at 08:16 29 December 2014

TABLE 1(Confinued) Moving Towards: a. Affection-verbala b. Affection-physical" c. Acceptance d. Dominancea e. Submissiona Moving against: a. Verbal b. Physical c. Death Moving away from: a. Separation b. Indifference c. Rejection No relationshipa

-

"Items deleted from the analysis. 'kerns added to the scoring system.

Stories were scored for the presence or absence of the content categories. Stories were coded by three scorers blind to the experimental condition of the subjects. In cases of disagreement between Scorer 1 and Scorer 2, Scorer 3 was used as a tiebreaker. In the 3% of cases in which there was disagreement between a11 three scorers, the category was discussed until a consensus was reached. The following are examples of categories scored for three different responses. She's all excited because he asked her to go out and he hasn't asked her to do something in a long time because they've been so busy working and so she decides to go out and they had a really good time and she was real happy. Scored for: Excitement, acceptance, pleasure. So they got in this big fight and started wrestling around and stuff and she started screaming things at him and all of a sudden he swung and hit her in the face. Scored for: Conflict, verbal hostility, physical hostility, managementexternalized, moving against -verbal, moving against -physical. He is climbing, he's got a firm grip on the rope and so he is showing his determination during the performance. Scored for: Effort. Frequency data for the 53 TAT categories were examined; categories with frequencies of less than 10 (out of a possible 60) were dropped from further

598

WORCHEL, AARON, YATES

Downloaded by [University of Dayton] at 08:16 29 December 2014

analysis. Categories dropped were affection (4 items), dominance, submission, compulsion, excitement, frustration, guilt, inferiority, pain, path indicators (3 items), and possessiveness. In addition, 3 hostility categories, 5 management of anger categories, and 1 relationship category were dropped due to redundancy. In order to reduce the number of dependent variables, a factor analysis was applied to the remaining 25 TAT categories, excluding the 3 outcome categories. A principal components solution was used, with the minimum eigenvalue to retain factors set at 1. Alternate factor structures were analyzed with an eight-factor solution emerging as the best solution, which accounted for 65% of the variance. In deciding which categories to retain or eliminate, the following method was used:

1. Categories with factor loadings of less than .4O were dropped (pleasure, superiority, and indifference). 2. Several categories loaded highly on more than one factor. Categories were maintained on both factors if the factor clusters appeared to tap different constructs. 3. Factors were eliminated if all categories on that factor with loadings above .40 were duplicated on other factors. Thus, one factor was eliminated. The final seven-factor solution is presented in Table 2. A brief description of the factors, with a hypothesis regarding the underlying content commonality, follows:

1. General Concerns-two items, sexuality and acceptance, which denote a 2. 3.

4. 5.

6. 7.

positive interrelatedness. The remainder of the items suggest psychological distress-an emotional irritant requiring action. Withdrawal-remorse regarding loss of relatedness due to someone's death. Aggression-death, wrongdoing, and aggression. Anxiety-a sense of urgency or upset regarding wrongdoing. Rejection-similar to the withdrawal factor, with the loss of relatedness due to being rejected as opposed to someone's death. Interpersonal Relations-longing for interpersonal relationships. Suicide-an action oriented suicidal ideation.

RESULTS A MANOVA was used with gender and type of card as independent variables; dependent variables were the seven TAT-derived factors (General Concerns,

GENDER BIAS ON THE TAT TABLE 2 Factor Loadings of Items on the TAT Scoring System

599

-

Factor

Downloaded by [University of Dayton] at 08:16 29 December 2014

Item Nonhostile death Hostile death Crime Physical hostility Conflict Effort Escape Verbal hostility Sexuality Ambivalence Fear Acceptance Separation Rejection Depression Wishful thinking Anger Anxiety Need to achieve Loneliness Suicide

General Interpersonal Concerns Withdrawal Aggression Anxiety Rejection Relations Suicide .13 -.I3 -.I2 .23 .59 .61 .52 .49 .43 .54 .52 .53 .64 .18 .12 .30 .37 .33 .12 .07 .18

Withdrawal, Aggression, Anxiety, Interpersonal Relations, Rejection, and Suicide) and the three outcome items (favorable, unfavorable, and indeterminate). As seen in Table 3, there was a statisticallysignificant main effect for type of card on General Concerns, F(l, 59) = 11.34, p < .001. Subjects responded more often to categories on the General Concerns scale when the female TAT carcls were administered than when the male TAT cards were administered. There was also a statistically significant main effect for gender on Interpersonal Relations, F(l, 59) = 4.59, p < .03. Females responded more often to categories on the Interpersonal Relations scale than did males. In addition, two statistically significant interactions emerged. There was a Gender x Type of Card interaction for Withdrawal, F(1,59) = 7.19, p < .009. Males responded more often to categories on the Withdrawal factor when the female TAT cards were administered; females responded more often to categories on the Withdrawal factor when the male TAT cards were administerecl.

600

WORCHEL, AARON, YATES TABLE 3

Mean Comparisons Across Groups on TAT Scores Female Cards Females

Downloaded by [University of Dayton] at 08:16 29 December 2014

Items General Concerns Withdrawal Aggression Anxiety Rejection Interpersonal Relations Suicide Favorable Outcome Unfavorable Outcome Indeterminate Outcome

Male Cards Males

Females

Males

M

SD

M

SD

M

SD

M

19.26 4.60 2.46 3.76 4.46

7.12 2.79 2.23 1.70 1.72

15.68 5.93 2.62 4.37 5.18

4.06 1.38 2.24 1.31 1.04

13.20 5.93 2.33 4.13 5.26

4.34 1.79 1.87 1.68 1.83

13.13 4.66 1.66 3.06 4.93

6.60 1.40

1.80 .73

5.68 1.31

1.30 1.30

6.00 1.40

1.46 .50

5.33 1.26

1.04" .70

2.73

2.18

3.00

1.96

2.80

2.27

1.33

1.34

.06

.82

1.18

1.22

1.06

1.03

1.26

1.48

5.66

2.38

4.81

2.71

5.06

2.31

6.33

2.49

SD 3.41" 1.~3~ 1.95 1.79~ 1.22

"Main effect for type of card. b ~ e n d e rx Type of Card interaction. 'Main effect for gender.

There was also a Gender x Type of Card interaction for Anxiety, F(l, 57) = 6.79, p < .01. Males responded more often to items on the Anxiety factor when the female TAT cards were administered; females responded more often to items on the Anxiety factor when the male TAT cards were administered. There were no significant main effects or interactions for the three outcome categories, favorable, unfavorable, and indeterminate.

DISCUSSION This study questioned whether male and female TAT cards elicit differential responses. It was found that there was an effect for type of card on one of the factor analyzed clusters. This factor, General Concerns, contained a relatively larger number of categories: conflict, effort, escape, verbal hostility, sexuality, ambivalence, fear, acceptance, and separation. Subjects responded more frequently to these categories when the female TAT cards were administered. Given that a standard administration involves showing the female cards to females, it appears that females would be evaluated as responding more strongly in all of these content areas. With the exception of the acceptance and sexuality categories, General Concerns appears to tap into a construct involving emotional distress. The content categories conflict, effort, escape, verbal hostiIity, ambivalence, and fear are suggestive of a psychological irritant requiring coping

Downloaded by [University of Dayton] at 08:16 29 December 2014

GENDER BIAS ON THE TAT

60'1

or resolution. These types of contents might change the tenor of an individual1 evaluation to indicate greater pathology. This particular cluster of categories was the first and largest factor extracted in the factor solution, accounting for the largest percent of variance in this scoring system-17%. The Fine system contains categories similar to several other scoring systems (e.g., Eron, 1950; Morgan & Murray, 1935). Thus, one could generalize in suggesting that other scoring systems may also generate some gender bias. Although the degree of confounding due to gender is relatively small, the risk of invalid testing leads us to conclude that a standard set of stimulus cards should be considered. The interactions were consistent in finding that on two factor scales, Withdrawal and Anxiety, males responded more often to the female cards whereas females responded more often to the male cards. Because this type of administration would rarely be utilized, this finding has little clinical significance. In addition, this study revealed a gender difference not attributable to the type of card administered. Women responded more often to items on the Interpersonal Relations factor scale than did men. This scale loads positively on the categories loneliness and acceptance. The acceptance category is a measure of positive, nurturant interactions between two or more individuals and iis somewhat similar to intimacy as measured by McAdams et al. (1988). Thus, this finding of a gender difference supports McAdams et ale'sresults. They failed to confirm Pollack and Gilligan's (1982) results, which indicate more violent themes by men. Our study shows no gender difference for the Aggression scale, which positively loads on the categories hostile death, crime, physical hostility, and conflict. Thus, we are again consistent with McAdams et a1.s' findings concerning gender differences. There are three limitations to our study. First, a card by card analysis would be desirable to determine the specific cards that elicit differential responses. However, this analysis was not possible with 60 subjects. Thus, we are only able to suggest a nonspecific sex bias across the two groups of cards. Second, the results of this study were based on a specific scoring system. This scoring system was chosen based on its representation of a broad number of content categories and similarity to other techniques; however, replication with alternate systems would be desirable. Third, an older, noncollege student population may have been less influenced by the Women's Movement and might have been more likely to show differences along the lines of more traditional sex-role stereotypes.

ACKNOWLEDGMENT We thank Erik Olsen for his help in data collection.

REFERENCES Eron, L. D. (1950).A normative study of the Thematic Apperception Test. Psychological Monographs, 64(Whole No. 315).

Downloaded by [University of Dayton] at 08:16 29 December 2014

602

WORCHEL, AARON, YATES

Kaplan, R. M., & Saccuzzo, D. P. (1989). Psychologrcal tesnng: Principles, apphcations, and issues (2nd ed.). Pacific Grove, CA: Brooks/Cole. May, R. (1975). Further studies on deprivatiodenhancement patterns. Journal of Personality Assessment, 39, 166-122. McAdams, D. P., Lester, R. M., Bqand, P. A., McNamara, W. J., & Lensky, D. B. (1988). Sex and the TAT: Are women more intimate than men? Do men fear intimacy? Journal of Personality Assessment, 52, 397409. Morgan, C. D., & Murray, H. A. (1935). A method for investigating phantasies: The Thematic Apperception Test. Archives of Neurology and Psychiatry, 34, 289-306. Murray, H. A. (1943). Themanc Apperception Test Manual. Cambridge: Harvard University. Murstein, B. L. (1963). Them and research m projective techniques. New York: Wiley. Murstein, B. L. (1965). Handbook of projectwe techniques. New York: Basic Books. Palmer, D. J., Olivarez, A., Willson, V. L., & Fordyce, T. (1989). Ethnicity and language dominance-Influence on the prediction of achievement based on intelligence test scores in nonreferred and referred samples. Learning Disabiltty Quarterly, 12, 261-274. Pollack, S., & Gilligan, C. (1982). Images of violence in Thematic Apperception Test stories. lournal of Personality and Social Psychology, 42, 159-167. Shneidman, E. S. (1951). Thematic test analysis. New York: Grune & Stratton. Stewart, A. J., & Chester, N. L. (1982). Sex differences in human social motives: Achievement, affdiation, and power. In A. Stewart (Ed.), Motwanon and society (pp. 172-218). San Francisco: Jossey-Bass. Worchel, F. F., & Dupree, J. L. (1990). Projective storytelling techniques. In C. Reynolds & R. Kamphatls (Eds.), Handbook of psychological and educational assessment of children (pp. 70-88). New York: Guilford.

Frances F. Worchel Department of Educational Psychology Texas AGrM University College Station, TX 77843 Received October 3, 1989 Revised February 3, 1990

Gender bias on the Thematic Apperception Test.

Thirty male and 30 female college students generated stories to a set of nine Thematic Apperception Test (TAT) cards with either male or female stimul...
463KB Sizes 0 Downloads 0 Views