THE AMERICAN JOURNAL OF PSYCHOANALYSIS 35:163-168 (1975)
HOMOSEXUALITY AND PSYCHOPATHOLOGY IN NONPATIENT MALES* Thomas R. Clark A primary assumption running through the clinical literature is that homosexuality is always an indication of emotional maladjustment, personality disorder, aberrant self-esteem, and/or other serious psychopathology. ~'2 Despite recent evidence to the contrary from the biological and social sciences3 and also from some behavioral researchers, 4,s this homosexuality-psychopathology equation still rules the day among both clinical researchers and practitioners. However, Churchill 6 points out that most studies up until now have dealt in stereotypes and have involved only mentally disturbed subjects in therapy or under incarceration. These studies have inappropriately generalized the psychopathology found in these individuals to the whole homosexual population. He states that what is needed now, in order to get a more representative appraisal and profile of the homosexual population, are studies using empirical psychological tests with a large sample of nonpatient subjects, including not only those who are exclusively homosexual and heterosexual, but also those who express varying degrees of homosexual behavior and preference along with heterosexual desires (this latter group not having been considered in the plethora of past research). The purpose of the present study was to further experimentally investigate the assumed relationship between homosexuality and psychopathology. It used a reliable and valid psychological measure of personality adjustment and focused on collegeeducated, nonpatient males expressing varying degrees of homosexual behavior and preference. The hypotheses tested were: (I) Groups of nonpatient subjects involved in varying degrees of homosexual preference (Groups I - 6 ) will not differ significantly i n levels of psychopathology from a group of matched heterosexuals (Group 0), as revealed by total group mean scores across a battery of eight personality adjustment and psychopathology scales on a standard psychological measure; (2) These same groups *This study is the first to use nonpatient homosexual subjects expressingvarying degrees of homosexual behavior and preference,a significant methodological departure from past research which did not consider either of these two important variables. Thomas R. Clark, Ph.D., University of Windsor, Ontario, and Wayne Country Psychiatric Hospital, Detroit. 163
164
THOMAS R. CLARK
also will not differ significantly in pattern of individual psychopathology scores on the same personality measure, that is, the pattern of group mean: scores upon indi~?idual measures across all groups will not be significantly different; (3) There will be no significant prediction of psychopathology from level of homosexuality, that is, homosexuality per se will not be a criterion predictor of psychopathology, and there will be no statistically significant degree of relationship between level of homosexual behavior and preference (as indicated by group position on the Kinsey Heterosexual-Homosexual Rating Scale) and any of the types of psychopathology emphasized in the clinical literature (as measured by psychological test scores on eight measures of emotional adjustment and disturbance). Methods
The subjects that participated in the entire experiment were 140 college-educated, socially functioning, nonpatient adult males from various parts of the United States and Canada, chosen on the basis of admitted homosexual or heterosexual preference. The subjects' were from 20 to 39 years old, with a mean age of 29.2 years and a median age of 30 years. Several subjects had been awarded the M.A. and/or Ph.D. or M.D. degrees; others were still college undergraduates or had recently graduated with baccalaureate degrees. The subjects' academic and occupational backgrounds were extremely heterogeneous, with approximately twenty academic majors represented. Occupations included medicine and other health professions (31), law (10), education (16), business (29), professional athletics (17), the ministry (4), the armed forces (5), and the acting profession (28). All subjects were either employed or continuing their education and were functioning adequately in society; none were in psychotherapy or legal difficulty because of homosexual behavior. Since, as Churchill 6 points out, the homosexuality of most males is not an all-ornothing proposition (a variable overlooked in all previous studies), and the only valid approach to researching homosexuality is to distinguish degrees of homosexual orientation on the basis of frequency and preference of homosexual activity, the first step of this study made use of the seven-point Heterosexual-Homosexual Rating Scale devised by Kinsey et. al., 7 to better describe the gradual transitional continuum of sexuality between exclusive heterosexuality and exclusive homosexuality. In this rating scale, 0 represents individuals with a history of exclusive heterosexuality (0 percent homosexual) and 6 represents individuals with a history of exclusive homosexuality (100 percent homosexual). One through 5 represent all those individuals with a history of varying combinations of heterosexual and homosexual experience and mental preference, as follows: those rated 1 are 1 - 1 9 oercent homosexual; those rated 2 are 2 0 - 3 9 percent homosexual; those rated 3 are 4 0 - 5 9 percent homosexual; those rated 4 are 6 0 - 7 9 percent homosexual; and those rated 5 are 8 0 - 9 9 percent homosexual. These ratings take into account not only the individual's overt behavioral experiences but also his psychosexual reactions and mental preferences or fantasies. An individual's position on this scale is determined not by the absolute numerical amount of overt ex-
H O M O S E X U A L I T Y A N D P S Y C H O P A T H O L O G Y IN N O N P A T I E N T MALES
165
perience or psychic reaction alone, but rather by the percentage ratio (relation) of the heterosexual to the homosexual in his history. The Kinsey Scale was used in the present research because past studies have involved only those subjects at the upper extreme of the continuum (exclusive homosexuals rated 6), and the serious neglect of those ratable from 1 through 5 has tended to obscure the full significance of homosexuality in the life of the adult human male. The second step of the present study utilized the Tennessee (Department of Mental Health) Self-Concept Scale (TSCS), developed by Fitts, 8 and ;its Clinical and Research Form answer sheet. The eight TSCS empirical psychopathology measures used in the study were self-criticism, defensiveness, self-concept, general emotional maladjustment, neurosis, psychosis, personality-character disorder, and overall personality integration. Over 200 subjects were interviewed by the experimenter during an eighteen-month period. The homosexual subjects were contacted through persons in the homosexual subculture; the heterosexuals came from a similar college and professional population in the United States and Canada. However, only 140 subjects actually participated in the final experimental task, giving an n = 20 for each group (the largest that could be used). The purpose of the interview sessions was to determine the extent of each subject's sexual preferences from both an overt and psychic standpoint; on the basis of the interview findings, each subject's percentage ratio of heterosexual-to-homosexual behavior and preference was calculated. Each subject was then assigned to the appropriate one of seven matched group ratings on the Kinsey Heterosexual-Homosexual Rating Scale. This produced seven matched groups, each with an n = 20. Each group varied from the others only in the percentage of homosexuality expressed by its members; all subjects were otherwise matched for age, educational background, and lack of clinical or patient history of significant psychological disturbance. All subjects within a group thus showed the same approximate balance between the heterosexual and homosexual elements in their history. Group 0 had only those subjects in it who expressed no homosexual tendencies at all; they formed the exclusively heterosexual comparison group for this experiment. Groups 1 through 5 were experimental groups composed of subjects espousing varying degrees of both heterosexuality and homosexuality, with the percentage of homosexual activity and preference of the group members increasing (and heterosexual behavior decreasing) as the group's numerical designation increased. Experimental Group 6 contained only those who were completely homosexual in behavior and preference and who responded to no heter0sexual stimuli. All seven groups were then administered the TSCS, and the Clinical and Research Form answer sheets were sent away for computer scoring. In order to test Hypotheses 1 and 2, mean performances of the seven groups on the eight scales of the TSCS were analyzed for significant differences using a two-way analysis of variance (ANOVA), 8 X 7 design, for a two-factor experiment with repeated measures on one factor. 9 Additionally, to analyze the data obtained in a manner consistent with Hypothesis 3, a multiple regression analysis was utilized.
166
THOMAS R. CLARK
Results
Hypothesis 1 would be established if a nonsignificant F for main effect A (between groups main effect or homosexuality level) could be obtained from the ANOVA, while Hypothesis 2 would be confirmed if no significant A X B interaction (group level of homosexuality X mean score on TSCS measures) could be obtained. The statistical analysis revealed no significant main or interaction effects (F
HOMOSEXUALITY AND PSYCHOPATHOLOGY IN NONPATIENT MALES* Thomas R. Clark A primary assumption running through the clinical literature is that homosexuality is always an indication of emotional maladjustment, personality disorder, aberrant self-esteem, and/or other serious psychopathology. ~'2 Despite recent evidence to the contrary from the biological and social sciences3 and also from some behavioral researchers, 4,s this homosexuality-psychopathology equation still rules the day among both clinical researchers and practitioners. However, Churchill 6 points out that most studies up until now have dealt in stereotypes and have involved only mentally disturbed subjects in therapy or under incarceration. These studies have inappropriately generalized the psychopathology found in these individuals to the whole homosexual population. He states that what is needed now, in order to get a more representative appraisal and profile of the homosexual population, are studies using empirical psychological tests with a large sample of nonpatient subjects, including not only those who are exclusively homosexual and heterosexual, but also those who express varying degrees of homosexual behavior and preference along with heterosexual desires (this latter group not having been considered in the plethora of past research). The purpose of the present study was to further experimentally investigate the assumed relationship between homosexuality and psychopathology. It used a reliable and valid psychological measure of personality adjustment and focused on collegeeducated, nonpatient males expressing varying degrees of homosexual behavior and preference. The hypotheses tested were: (I) Groups of nonpatient subjects involved in varying degrees of homosexual preference (Groups I - 6 ) will not differ significantly i n levels of psychopathology from a group of matched heterosexuals (Group 0), as revealed by total group mean scores across a battery of eight personality adjustment and psychopathology scales on a standard psychological measure; (2) These same groups *This study is the first to use nonpatient homosexual subjects expressingvarying degrees of homosexual behavior and preference,a significant methodological departure from past research which did not consider either of these two important variables. Thomas R. Clark, Ph.D., University of Windsor, Ontario, and Wayne Country Psychiatric Hospital, Detroit. 163
164
THOMAS R. CLARK
also will not differ significantly in pattern of individual psychopathology scores on the same personality measure, that is, the pattern of group mean: scores upon indi~?idual measures across all groups will not be significantly different; (3) There will be no significant prediction of psychopathology from level of homosexuality, that is, homosexuality per se will not be a criterion predictor of psychopathology, and there will be no statistically significant degree of relationship between level of homosexual behavior and preference (as indicated by group position on the Kinsey Heterosexual-Homosexual Rating Scale) and any of the types of psychopathology emphasized in the clinical literature (as measured by psychological test scores on eight measures of emotional adjustment and disturbance). Methods
The subjects that participated in the entire experiment were 140 college-educated, socially functioning, nonpatient adult males from various parts of the United States and Canada, chosen on the basis of admitted homosexual or heterosexual preference. The subjects' were from 20 to 39 years old, with a mean age of 29.2 years and a median age of 30 years. Several subjects had been awarded the M.A. and/or Ph.D. or M.D. degrees; others were still college undergraduates or had recently graduated with baccalaureate degrees. The subjects' academic and occupational backgrounds were extremely heterogeneous, with approximately twenty academic majors represented. Occupations included medicine and other health professions (31), law (10), education (16), business (29), professional athletics (17), the ministry (4), the armed forces (5), and the acting profession (28). All subjects were either employed or continuing their education and were functioning adequately in society; none were in psychotherapy or legal difficulty because of homosexual behavior. Since, as Churchill 6 points out, the homosexuality of most males is not an all-ornothing proposition (a variable overlooked in all previous studies), and the only valid approach to researching homosexuality is to distinguish degrees of homosexual orientation on the basis of frequency and preference of homosexual activity, the first step of this study made use of the seven-point Heterosexual-Homosexual Rating Scale devised by Kinsey et. al., 7 to better describe the gradual transitional continuum of sexuality between exclusive heterosexuality and exclusive homosexuality. In this rating scale, 0 represents individuals with a history of exclusive heterosexuality (0 percent homosexual) and 6 represents individuals with a history of exclusive homosexuality (100 percent homosexual). One through 5 represent all those individuals with a history of varying combinations of heterosexual and homosexual experience and mental preference, as follows: those rated 1 are 1 - 1 9 oercent homosexual; those rated 2 are 2 0 - 3 9 percent homosexual; those rated 3 are 4 0 - 5 9 percent homosexual; those rated 4 are 6 0 - 7 9 percent homosexual; and those rated 5 are 8 0 - 9 9 percent homosexual. These ratings take into account not only the individual's overt behavioral experiences but also his psychosexual reactions and mental preferences or fantasies. An individual's position on this scale is determined not by the absolute numerical amount of overt ex-
H O M O S E X U A L I T Y A N D P S Y C H O P A T H O L O G Y IN N O N P A T I E N T MALES
165
perience or psychic reaction alone, but rather by the percentage ratio (relation) of the heterosexual to the homosexual in his history. The Kinsey Scale was used in the present research because past studies have involved only those subjects at the upper extreme of the continuum (exclusive homosexuals rated 6), and the serious neglect of those ratable from 1 through 5 has tended to obscure the full significance of homosexuality in the life of the adult human male. The second step of the present study utilized the Tennessee (Department of Mental Health) Self-Concept Scale (TSCS), developed by Fitts, 8 and ;its Clinical and Research Form answer sheet. The eight TSCS empirical psychopathology measures used in the study were self-criticism, defensiveness, self-concept, general emotional maladjustment, neurosis, psychosis, personality-character disorder, and overall personality integration. Over 200 subjects were interviewed by the experimenter during an eighteen-month period. The homosexual subjects were contacted through persons in the homosexual subculture; the heterosexuals came from a similar college and professional population in the United States and Canada. However, only 140 subjects actually participated in the final experimental task, giving an n = 20 for each group (the largest that could be used). The purpose of the interview sessions was to determine the extent of each subject's sexual preferences from both an overt and psychic standpoint; on the basis of the interview findings, each subject's percentage ratio of heterosexual-to-homosexual behavior and preference was calculated. Each subject was then assigned to the appropriate one of seven matched group ratings on the Kinsey Heterosexual-Homosexual Rating Scale. This produced seven matched groups, each with an n = 20. Each group varied from the others only in the percentage of homosexuality expressed by its members; all subjects were otherwise matched for age, educational background, and lack of clinical or patient history of significant psychological disturbance. All subjects within a group thus showed the same approximate balance between the heterosexual and homosexual elements in their history. Group 0 had only those subjects in it who expressed no homosexual tendencies at all; they formed the exclusively heterosexual comparison group for this experiment. Groups 1 through 5 were experimental groups composed of subjects espousing varying degrees of both heterosexuality and homosexuality, with the percentage of homosexual activity and preference of the group members increasing (and heterosexual behavior decreasing) as the group's numerical designation increased. Experimental Group 6 contained only those who were completely homosexual in behavior and preference and who responded to no heter0sexual stimuli. All seven groups were then administered the TSCS, and the Clinical and Research Form answer sheets were sent away for computer scoring. In order to test Hypotheses 1 and 2, mean performances of the seven groups on the eight scales of the TSCS were analyzed for significant differences using a two-way analysis of variance (ANOVA), 8 X 7 design, for a two-factor experiment with repeated measures on one factor. 9 Additionally, to analyze the data obtained in a manner consistent with Hypothesis 3, a multiple regression analysis was utilized.
166
THOMAS R. CLARK
Results
Hypothesis 1 would be established if a nonsignificant F for main effect A (between groups main effect or homosexuality level) could be obtained from the ANOVA, while Hypothesis 2 would be confirmed if no significant A X B interaction (group level of homosexuality X mean score on TSCS measures) could be obtained. The statistical analysis revealed no significant main or interaction effects (F
