Citation
Psychological construction of sexual satisfaction among high and low sex guilt male and female university students

Material Information

Title:
Psychological construction of sexual satisfaction among high and low sex guilt male and female university students
Creator:
Suchman, Alma Rodgers, 1939-
Copyright Date:
1988
Language:
English

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Alma Rodgers Suchman. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
19691935 ( OCLC )
AFJ8515 ( LTUF )
0024127167 ( ALEPH )

Downloads

This item has the following downloads:


Full Text
PSYCHOLOGICAL CONSTRUCTION OF
SEXUAL SATISFACTION AMONG HIGH AND LOW
SEX GUILT MALE AND FEMALE UNIVERSITY STUDENTS
By
ALMA RODGERS SUCHMAN

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

1 988




COPYRIGHT 1988
by
ALMA RODGERS SUCHMAN




ACKNOWLEDGEMENTS

I would like to take this opportunity to express my sincere appreciation to my former "boss" and committee chairperson, Dr. Joe Wittmer, who has always come through with the right balance of information, support, and concern throughout my doctoral program. My heartfelt gratitude goes to my committee members, Dr. Franz Epting, who has been tireless in his willingness to provide information, time, and resources to make this dissertation possible; Dr. Harry Grater, who inspired me to take that "first step" to begin graduate school and has always been there for each step along the way; Dr. Phyllis Meek, who never failed to "lift my spirits" during her many roles as instructor, supervisor, and committee member; and Dr. Jaquie Resnick, who generously entrusted me with professional developmental opportunities and gave steadfast support to my research ideas.
A special thank you is due to Mr. Steve Wiggins, Dr. Greg Neimeyer, and Ms. April Metzler for their participation, consultation, and interest in the data analyses of this study. my friend, Ms. Joan Davidsen, gave constant encouragement as well as invaluable assistance with the organization of the data.

iii




I would also like to give special acknowledgement to my husband, David Suchman, for his love, support, and constant faith in me. David, my family, and friends have been great sources of comfort and encouragement throughout this endeavor.




TABLE OF CONTENTS

PAGE
ACKNOWLEDGEMENTS ................ ... ...... ........i iii
ABSTRACT ...... *........ o...... *...... .. ....... vii
CHAPTER
ONE INTRODUCTION .............. ... o. ....... ... 1
Need for the Study ..... ..... ..... ... ....... 8
Purpose of the Study ........ o ... .......... ... 8
Definition of Terms ............. o ........ .. 9
Organization of the Remainder of the Study... 10
TWO REVIEW OF THE LITERATUREo .................... 12
Sexual Satisfaction. ......... .... ........... 12
Guilt ...................................o... 27
Personal Construct Theory.................... 31
Summary ...... ....... ........ ...... .o....... 35
THREE METHOD ....... .... .... .... .... .... .... .. 37
Population and Sample., ..................... 37
Data Collection. ..... ...........o......... .... 38
Hypotheses ........ .... ... ....... ..... .. 41
Instrumentation .... ........ ......... ........ 43
Data Analyses.......... ..... ........ ...... 53
FOUR RESULTS............... .... ....... ... ... .. 55
Hypotheses Related to Structure ............. 56
Hypotheses Related to Content ..............o. 69
Summary ..... ........ ... ..... ..... o......... 74
FIVE SUMMARY, DISCUSSION, LIMITATIONS AND
IMPLICATIONS FOR FUTURE RESEARCH ............. 75
Summary.. .. ..o.. .. o o .. . ... 75
Discussion.... ....... [[[o .. ..... o .......... 76
Limitations of the Study ................... 84
Implications for Future Research....o......... 85




PAGE
APPENDIX
A INFORMED CONSENT .................... 0 ...... 89
B REVISED MG INVENTORY ................ 94
C REVISED MG INVENTORY INSTRUCTIONS-* ...... 98
D PROCEDURES FOR ADMINISTERING THE REP-TEST .... 100 E RATING SCALE, ..... o ...... o ................. 105
F SEXUAL EXPERIENCES ......... o .... o ............ 107
G SORTS ......................... ......... o ... 110
REFERENCES ............................ ill
BIOGRAPHICAL SKETCH .................................. 121




Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy
PSYCHOLOGICAL CONSTRUCTION OF
SEXUAL SATISFACTION AMONG HIGH AND LOW
SEX-GUILT MALE AND FEMALE UNIVERSITY STUDENTS By
Alma Rodgers Suchman
April 1988
Chairman: Dr. Joe Wittmer
Major Department: Counselor Education
The purpose of this study was to assess differences on the bases of gender and level of sex-guilt in cognitive responses made by male and female university students to describe their satisfaction in personal sexual experiences. Also investigated was the nature of constructs they used to represent concepts of sexual satisfaction. Fourteen null hypotheses, related to the organization and content of sexual satisfaction constructs, were formulated in accord with Kelly's personal construct theory.
The 80 students who participated in this study were selected from among 297 undergraduate students in an introductory psychology course subject pool at the University of Florida during the fall semester 1987. Criteria for selection included that they be (a) white, (b) from ages 18 through 25, (c) heterosexual, and (d) currently or have been engaged in a sexual relationship.

vii




The participants were assigned to high or low
sex-guilt groups based on scores from the Mosher Sex-Guilt Scale. Students then completed a 10 X 10 Sexual Satisfaction Repertory Grid which yielded sexual satisfaction constructs and ratings.
No significant differences (p < .05) were found in the structural properties of sexual satisfaction constructs rated by male and female high and low sex-guilt university students. However, male students in this study did demonstrate somewhat more complexity in their use of personal constructs of sexual satisfaction. The female university students tended to rate their sexual experiences in a more extreme manner and more negatively than did male students.
Two content hypotheses were analyzed through use of the chi-square statistic. Significant differences by gender were found in the types of constructs generated for sexual satisfaction. Male students used constructs connotative of status, appearance, and egoism more frequently than did female students. Female students made more references to feelings of tenderness and self-references than did male students.
The results of this study provide information about
the cognitive structures and content related to constructs of sexual satisfaction. Male and female students were found to be more similar than different in their manner

viii




of construing sexual experiences. There were significant differences, however, in the content of the constructs generated by male and female students to describe these experiences. Sex-guilt did not appear to be a factor in the structure or content of personal constructs used by the university students participating in this study.




CHAPTER ONE
INTRODUCTION
ours is a culture that makes figuring out how to
be physically sexual so complicated that many
people never get to the question of how they
feel about it. (Gagnon, 1977, p. 212)
Few subjects generate as much interest as does human sexuality. The literature of the history of Western civilization is replete with examples of sexual repression and sexual liberation. These reports illuminate many of the myths and misinformation associated with human sexual behavior. Eventually however, the restrictive and repressive influences of the Puritan and Victorian eras gave way to the permissive, "freer" sexual attitudes of the 20th century (Harmatz & Novak, 1983). Indeed, the early 1960s included a sexual revolution (Barrett, 1972; Conley & O'Rourke, 1973). Many factors contributed to the change: (a) availability of birth control pills, (b) protest movements, (c) the new feminism, and
(d) greater openness in discussions and displays of sex (Masters, Johnson, & Kolodny, 1986).
Groundwork for research in human sexual functioning had been laid in the late 19th and early 20th century by the works of Viennese physician Sigmund Freud, German




psychiatrist Richard von Krafft-Ebing, British psychologist Havelock Ellis, and others. Unfortunately their written works were based primarily on pathological populations. It was not until the middle of the 1900s that substantial research involving "normal" people took place. Alfred Kinsey (1953), in an effort to gather information on human sexual behavior, interviewed thousands of men and women across the United States. He was severely criticized for this investigation of sexual behavior (Harmatz & Novak, 1983). His research, however, paved the way for other more direct approaches to investigations of human sexual functioning. William H. Masters and Virginia E. Johnson (1966) focused on the anatomy and physiology of sexual response. Their direct observation and measurement of sexual behavior received shocked cries of outrage from the public and mixed reviews from professionals (Masters, Johnson, & Kolodny, 1986).
Until recently, the vast majority of the research on human sexual functioning has been focused upon the physiology of sexuality; the psychological significance of human sexuality has been almost totally neglected (Eysenck, 1971). Masters and Johnson (1966) suggested that sexual adequacy is based on appropriate interactions between biophysical and psychosocial factors. It follows that simply alleviating a physiological sexual dysfunction is not necessarily synonymous with helping a person or




couple attain sexual satisfaction. Kolodny (1981) found that some people who materially improved their sexual satisfaction did not change a great deal from a viewpoint of their physical sexual functioning.
Direct behavioral intervention methods in the
treatment of sexual dysfunctions such as erectile failure or premature ejaculation in men and orgasmic dysfunction in women have been effectively used by Masters and Johnson (1970), LoPiccolo and Lobitz (1972), and Kaplan (1974). There have been few attempts, however, to understand the sexual relationship of normal, non-dysfunctional couples who complain of sexual dissatisfaction. Estimates of dissatisfaction with marital sex life range from one third to one half of all couples (Hite, 1976; Kaplan, 1974).
Noted sexologists Kaplan (1974); Comfort (1972);
Masters, Johnson, and Kolodny (1986); and LoPiccolo and LoPiccolo (1978) have cautioned about the complexity of human sexual response and the limitations of a mechanistic attitude toward treatment of sexual problems. In Sexual Turning Points, Lorna and Philip Sarrel (1984) recognized that, "even though ours is supposedly a nation obsessed with sex, there aren't many opportunities to really find out the details of the deepest meanings of sex in others' lives" (p. 4). This lack of opportunity prevents men and women from knowing the commonality of anxieties and experiences. Segal (1984) provided such an opportunity




for 2,436 college students in five universities by having them write "sexual autobiographies." These autobiographies revealed that 83% of the males and 78% of the females were sexually active. Among those who had experienced intercourse, the average age of first coitus was 16.9 years for male and 18.3 years for female students. These results correspond to a recent national poll of 1,000 teenagers in which it was found that 57% of all American teens were sexually active by the age of 17 ("Teens Speak," 1987).
Despite tremendous advances in the understanding of sexual behavior and openness of discussion about sex, and a great deal of sexual activity, sexual partners still complain of a lack of feeling and passion. "So much sex and so little meaning or even fun in it!" exclaimed Rollo May (1969, p. 40). Masters, Johnson, and Kolodny (1986) concurred by stating, "There is little question that personal dissatisfaction with sex is commonplace in our society today. Half of all American marriages are troubled by some form of sexual distress ranging from disinterest and boredom to outright sexual dysfunction" (p. 440). Sexual partners frequently have little insight into the dynamics which contribute to this sense of alienation, loneliness, and depersonalization.
Variables related to enhancement or diminishment of sexual response are found in recent research on sexuality




(Heiman, 1980). Variables studied thus far include
anxiety, expectancy, genital response feedback, sex role, frequency of intercourse, and guilt. Unfortunately, however, sexual satisfaction often has been measured by a single question about how satisfied a person is with his or her sex life. Attempts to define the meaning of sexual satisfaction in operational terms have been few.
There is no single comprehensive theory regarding the etiology of sexual satisfaction. It is generally agreed, however, that the majority of sexual concerns are created or influenced by psychological rather than biological factors (Kaplan, 1974; Koch, 1983; May 1969). Areas of psychological concern are illuminated by an understanding of the way in which an individual construes satisfying and unsatisfying sexual experiences. How do sexually active individuals perceive their sexual experiences? Are there gender differences in the meaning of sexual satisfaction? How does the personality disposition of sex-guilt affect the way in which an individual construes sexual satisfaction? These questions were addressed in this study.
In order to investigate these questions, it was
necessary to adopt a theoretical framework from which to operationalize the concept of sexual satisfaction. One theory particularly well-suited to this purpose was the personal construct theory (Kelly, 1955). Personal




construct theorists have suggested that each individual develops a personal view of sexual satisfaction based upon his or her understanding and experience.
The way in which a person interprets particular
events is known as a (personal) construct. Kelly (1963) posited that reality is subject to as many ways as we ourselves might invent. Thus, a person's constructs are only limited by the range in which the application would be useful. Kelly rejected the notions of "man-the-biological-organism" or "man-the-lucky-guy," and instead embraced the perspective "man-the-scientist." This person is every person, each performing experiments in life events to improve the prediction and control of certain human phenomena. Particularly useful and important from the clinician's perspective is the implication that "man can enslave himself with his own ideas and then win his freedom again by reconstruing his life" (Kelly, 1963, p. 21). This suggests that the mind is dynamic, an individual formulates a theoretical framework of his/her universe, and the person concocts and re-concocts the design to better anticipate events. The person's view of the event is subject to modification and elaboration, a construing of events defined as the individual's personal construct system (Fransella & Bannister, 1977).




Kelly (1955) was interested in the content of the
person's construct system and analyzed the interrelations between the constructs to determine the meaningfulness, positivity or negativity, integration, and intricacy of the constructs. The technique for analyzing construct content, and for which Kelly is well known, is the repertory grid. Some construct systems may be cognitively complex, thus enabling the person to differentiate to a greater extent in the perception of events or behaviors. Other construct systems may reflect a constricted, less developed pattern. An individual's personal construct system of sexually satisfying experiences is particularly germane to this study.
There is evidence to suggest that cognitive processes in sexual situations are constricted by guilt (Mosher & Cross, 1971). Indeed, sex-guilt may influence the way in which sexual satisfaction is perceived (Abramson, Mosher, Abramson, & Woychowski, 1977; Janda, Magri, & Barnhart, 1977). Ways in which sex-guilt mediates sexual behavior are expanded upon in Chapter Two of this study and are accompanied by greater explanation of personal construct theory. In general, however, at present there are no studies concerning the mediating effects of sex-guilt on the personal constructs of sexual satisfaction.




Need for the Study
An understanding of the enormous complexity of human sexual response is essential for the planning of sexual education programs and clinical interventions. An erroneous current assumption is that factual knowledge about the physiological aspects of sex and reproduction alleviates the infinite variety of psychological sexual problems confronting males and females. Farley and Davis (1980) pointed to the absence of research that identifies the psychological sources of sexual satisfaction, even though it is reported as a significant factor in the happiness and discord in relationships. The development of operational definitions of sexual satisfaction other than physiologic is a necessary step in understanding human sexual behavior.
Purpose of the Study
The purpose of this study was to investigate the ways in which high and low sex-guilt male and female university students understand and organize the meaning of sexual satisfaction in their personal sexual experiences. The manners in which sex-guilt and/or gender mediates the structure and content of the constructs of sexual satisfaction also were examined. More specifically, the following questions were addressed:




1. What are the differences in cognitive differentiation
of sexual constructs among high and low sex-guilt male
and female students?
2. What are the differences in level of integration of
sexual satisfaction constructs among high and low
sex-guilt male and female students?
3. What are the differences in rating extremity of sexual
satisfaction constructs among high and low sex-guilt
male and female students?
4. What are the differences in overall positive and
negative perceptions of sexual experiences among high
and low sex-guilt male and female students?
5. What are the differences in the terms used to describe
the meaning of sexual satisfaction among high and low
sex-guilt male and female students?
Definition of Terms
The following definitions are used throughout this dissertation.
Cognitive differentiation is the process through
which an individual construes events in a multidimensional way. A more cognitively complex person has more dimensions for perceiving events than a less cognitively complex person.
Construct is a term used in personal construct theory to denote the unique way each person has of analyzing




events that occur in his or her life. The term was originally used by Kelly (1955) to denote a process of comparing and contrasting events.
Construct categories are 20 semantically meaningful groups defined by Landfield (1971).
Extremity refers to the degree or amount of scale
polarization of ratings found in the subjects' ratings on a 13-point scale.
Integration refers to the extent of the interrelationship among constructs used by an individual in his/her construct system.
Sex-guilt is a term defined by Mosher and Cross (1971) as a personality disposition to a generalized expectancy for self-mediated punishment for violating or anticipating violating standards of proper sexual conduct. High or low sex-guilt are assignments to levels of guilt as determined by subjects' responses to the Revised Mosher Sex-Guilt Inventory.
Sexual satisfaction is defined by the constructs
useful to males and females for understanding the meaning of a sexual experience as a satisfying one.
Organization of the Remainder of the Study
In Chapter Two is described the literature related to the research problem including research related to sexual satisfaction, sex-guilt, personal construct theory, and




the repertory grid technique. Chapter Three contains descriptions of the sampling procedures, instrumentation, data collection procedures, research hypotheses, and data analyses. The results of the study are presented in Chapter Four. Finally, in Chapter Five is presented a summary, discussion, and interpretation of the results, the limitations of the study, and recommendations for future research.




CHAPTER TWO
REVIEW OF THE LITERATURE
This chapter includes further delineation of the
literature related to the research problem. Those studies specifically related to the issue of sexual satisfaction are described in greater detail. An overview of the personality disposition of sex-guilt and studies illustrating its mediating effect on behavior are included as well as a brief overview of personal construct theory and an elaboration of repertory grid use.
Sexual Satisfaction
Much research on human sexuality has focused on the physiological problems of sexual dysfunction. Yet, there is evidence of a great deal of sexual dissatisfaction in the larger, nonclinical population (Barbach, 1984; Hite, 1976, 1982; Sarrel & Sarrel, 1984). This salient, recurring theme in human sexual behavior has received scant attention. Rollo May (1969) saw this excessive concern with the technical performance of sex as symptomatic of a culture devoid of a personal meaning of sexual behavior. May stated that "it is not surprising then, in this preoccupation with techniques that the




questions typically asked about an act of love-making are not, was there passion or meaning or pleasure in the act? but, How well did I perform?" (p. 44).
Masters and Johnson (1970) and Zilbergeld (1978) also emphasized the extent to which performance/goal-oriented sexual behavior detracts from sexual pleasure. At the 71st annual meeting of the International Academy of Sex Research, in Haifa, Israel, Gunter Schmidt (1982) lamented the current dominance of biological and clinical aspects of sex research and presented an argument for interdisciplinary thinking and approaches to sex research. This, however, is hardly a new theme. Krafft-Ebing in Psychopathia Sexualis (1924) asserted that medicine is "a science which ever seeks to trace all psychological manifestations to their anatomical and physiological sources" (p. vi).
This concern about the lack of an integration of
biological and psychological approaches to sex research arises again and again in the literature. Kaplan (1974) suggested that "despite tremendous recent advances, sexuality still remains a mystery in many respects" (p. 3). In the last several years there have been numerous efforts to further an understanding of sexual behavior. The research reported here is focused on efforts to understand sexual satisfaction.




Investigations of sexual satisfaction are often
correlated with sexual performance. Waterman and Chiauzzi (1982) conducted a study which examined the role of orgasm in male and female sexual enjoyment. Although several studies have been reported concluding that orgasm was not necessarily an indicator of sexual satisfaction (Perlman & Abramson, 1982), Waterman and Chiauzzi were concerned about the absence of research that included males as subjects. They concluded in their study of 42 university couples that there was no significant difference between the sexes regarding the relationship between both orgasm and sexual enjoyment. However, their findings suggested that orgasm consistency was significantly related to sexual satisfaction in females but not in males. This finding was attributed to the restricted range of male orgasm consistency among the male subjects and was not considered to be evidence that orgasm consistency was more important a factor for females than males. Both males' and females' pleasure ratings were higher when sexual activity occurred without orgasm. Instrument limitations and/or possibly the lack of pressure to achieve orgasm were two of several reasons proffered by the authors for this result. The latter reason stated does suggest that goal-oriented sexual behavior detracts from sexual pleasure.




Another study of a random sample of 249 college
students (130 males and 119 females) provided evidence of gender differences in motives for intercourse. The results are in accord with Bardwick's (1971) findings that commitment and love are the primary motivations for a female's participation in sexual intercourse. However, Bardwick's research did not include a male sample. Carroll, Volk, and Hyde (1985) found that males' motives more often included pleasure, fun, and physical reasons while females' motives were predominantly related to love, commitment, and emotion.
Hoon and Hoon (1978) examined responses of 370 women to questions related to demographic and cognitive measures and forms of sexual behavior and expression. They were interested in ratings which would discriminate between respondents who reported high and low levels of orgasm consistency and high and low satisfaction with sexual responsivity. A discriminant analysis showed that women who were most satisfied with their sexual responsivity reported frequent intercourse and orgasm consistency, and enjoyed erotic activities and sensate pleasures.
Denney, Field, and Quadagno (1984) were also
interested in assertions by Hite (1976, 1982) of differing needs in males and females for sexual satisfaction and compared differences in attitudes toward foreplay, intercourse, and afterplay of 39 male and 49 female




students from a large midwestern university. The students filled out a multiple-choice questionnaire about their own sexual behavior. Included was a question about which behaviors were most satisfying and which most dissatisfying. There were significant sex differences in the students' needs and desires. Women reported more enjoyment in foreplay and afterplay than in intercourse, while men reported intercourse as the most enjoyable aspect of sexual activity. This report did not include a definition of what constituted before or afterplay.
Perlman and Abramson (1982) designed a study to
assess the relationship between sexual satisfaction and nine theoretically relevant variables taken from Helen Kaplan's (1974) multicausal model of sexual functioning. Social desirability bias and gender differences were examined as well. Fifty-seven male and 91 female married or cohabiting UCLA students and other adults completed self-report measures of attitudes toward one's body, attitudes toward sex, sexual knowledge, experience, anxiety and pleasure evaluation, communication, and life stress.
Data were analyzed by computations of correlations
and factor analyses. The single most important correlate of sexual satisfaction (r = .78, p < .001) was found to be how pleasurable sex was evaluated by the respondent. Other significant variables included marital happiness




(r = .46, p < .001) and the absence of sexual anxiety (r = .24, p < .003). Three factors emerged from the factor analysis: a Sexual Satisfaction factor with significant loadings on satisfaction, pleasure, and sexual activity; an Age or Duration or Relationship factor; and a Sexual Frequency factor.
Koch (1983) investigated the etiology of sexual concerns by determining the relationship between the sex-related attitudes and beliefs expressed by college students and each of the 10 highest ranked sexual concerns experienced by these students. The study sample, 899 white, non-married students, voluntarily completed Koch's investigation of sexual satisfaction. Included in this investigation were background questions and two instruments developed for the study. The Sexual Concerns Inventory consisted of 54 potential sexual concerns which could be situational, relational, or physical. The respondent indicated the frequency and impact of each concern during the relationship with his or her last sexual partner. The Sexual Attitudes and Beliefs Scale consisted of 29 Likert-scaled items which measured the respondent's attitudes and beliefs toward body image, sexual self-understanding, sex-roles, sexual self-control, guilt, communication, intimacy, masturbation, rejection, and sexual performance. The 10 highest-ranked sexual concerns were (a) lack of privacy, (b) lack of time, (c)




ejaculating too quickly, (d) commitment, (e) difficulty reaching orgasm, (f) wanting more foreplay, (g) frequency of sexual relations, (h) being turned off by certain behaviors, (i) affection, and (j) inability to reach orgasm. Every one of the 10 highest-ranked sexual concerns, except for lack of time, was found to be significantly related to overall sex-related attitudes and beliefs.
These studies have focused on the attitudes,
physiological responses, and behavior related to sexual satisfaction. An effort to understand the personal meaning of sexual experience was undertaken by Grater and Downing (1981), Rodriguez (1981), Bernstein (1982), and Garrison (1983). Garrison examined the effects of gender and sex-role orientation (using the Bern Sex-Role Inventory) on the meaning of the sexual experience. His investigation was built on the results of the previous research which had established a meaning of sexual experience questionnaire (MOSE III). This questionnaire categorizes subject's responses on the following subscales: Affiliation, Inadequate/Undesirable, Achievement, Morality, and Erotic/Dominance. Garrison analyzed data from a total of 269 students, 117 females and 152 males. Few gender differences were reported. Male and female differences occurred on Inadequate/ Undesirable and Morality subscales with men rating




themselves greater on both factors. Sex-role differences emerged on all but one subscale--Inadequate/Undesirable. Garrison concluded that sex-role orientation influenced the meaning of sexual experience more significantly than did gender alone.
Sex role identity was the focus of a study by Baucom and Aiken (1984) who explored the relationships among masculinity, femininity, marital satisfaction, and response to behavioral marital therapy. Fifty-four married nonclinic couples and 54 maritally distressed clinic couples participated. The results indicated that femininity and masculinity were significantly correlated with marital satisfaction. Androgyny was a less frequent sex role identity for either husband or wife in the clinic group. Conversely, in nonclinic couples, there were more adrogynous husbands and wives than any other sex role type.
Another role definition was examined by Frank, Anderson, and Rubinstein (1979). This was an investigation of discrepancies between marital role ideals and marital role behaviors and the relationship of these role tensions to sexual satisfaction. Eighty nonpatient, 50 marital therapy, and 50 sex therapy couples completed a self-report questionnaire. Distressed couples tended to adhere to marital role assignments that were not consistent with ideal role assignments. They were also




more likely to disagree in their perception of who carried out various roles within the marriage. Higher levels of role strain were found to correlate with sexual dissatisfaction. Sexual satisfaction was defined in global terms.
Traupmann, Hatfield, and Wexler (1983) interviewed 189 college men and women to determine whether or not equity considerations were important in sexual relations. The investigators expected subjects who reported equitable treatment to have more satisfying sexual relationships than couples in inequitable relationships. Sexual satisfaction was measured by (a) rating on a scale of 8 (extremely satisfied) through 1 (extremely dissatisfied) and (b) respondents' reports as to how satisfied they felt immediately after a sexual encounter. The investigators hypothesized that subjects who reported equitable relationships (a) should be more content and (b) should have more satisfactory sexual relationships than those who feel either overbenefited or underbenefited. There was strong support for the first hypothesis and some, much weaker support, for the second hypothesis.
Farley and Davis (1980) examined the relationship between sexual satisfaction and the role of similarity/ dissimilarity of personality in 102 married couples. The personality dimensions of interest were extraversionintroversion, neuroticism, and psychoticism (described by




the authors as aggressive, cold, cruel, antisocial, and bizarre characteristics). These were measured by the PEN (psychoticism, extraversion-introversion, neuroticism inventory). Sexual satisfaction was an overall global self-rating. The authors reported results which indicated that females derived the greatest sexual satisfaction when spouses had almost identical personalities in terms of extraversion-introversion and neuroticism. For males there was no relationship between sexual satisfaction and similarity or dissimilarity on those dimensions, but males indicated greater sexual satisfaction when their spouses were identical to them in psychoticism compared to when spouses were opposite to themselves on this personality dimension. Farley and Davis speculated that females might respond sexually to the more normal aspects of personality while males might respond sexually to less healthy aspects
of personality.
Personality was also of interest to Heath (1978) who investigated the relationship between personality and sexual compatibility in 60 married men in their early 30s who participated in this longitudinal study. Increased sexual compatibility was significantly related to increased psychological maturity, interpersonal maturity, fulfillment of various adult roles, and to marital happiness.




The literature reveals a variety of approaches to
understanding sexual satisfaction. These self-report data have obvious relevance for the assessment of sexual functioning and behavior. While it is known that response bias such as social desirability may affect a subject's responses, research indicates results are meaningful (Eysenck, 1972).
A review of the literature reveals several scales
which have been constructed to measure sexual experience. Some of the scales are cumulative in style in the Guttman (1950) sense. These scales are cumulative and have a high degree of unidimensionality. This means that a person who has experienced 'more advanced"' sexual behaviors in a list of sexual behaviors (e.g., sexual intercourse) will more than likely have experienced the "less advanced" behaviors (e.g., nude body caresses), as well. Podell and Perkins (1957), Brady and Levitt (1965), Bentler (1968a, 1968b) and Zuckerman (1973) have developed Guttman-type scales which assess a limited range of sexual behaviors. None of these measures include a measure of sexual satisfaction, however.
Several instruments do include an assessment of sexual satisfaction. A few of these have employed a single-item global measurement that required a yes or no response or possibly the single-item was rated on a degree of satisfaction (Heiman, 1980). The major disadvantages




to this approach include restricted variance and response set which is the tendency for people to respond higher when only one question is asked.
A somewhat more complex approach, developed by Snyder and Berg (1983), is a sexual dissatisfaction scale comprised of 29 true-false items ranging broadly in content to assess frequency, variety, and quality of sexual activities. This is 1 of 11 scales in a comprehensive 280-item Marital Satisfaction Inventory.
Another questionnaire used to measure sexual
satisfaction is Eysenck's Inventory of Attitudes to Sex (Eysenck, 1976). This inventory consisted of 98 questions requiring "yes" or "n" responses. A factor analysis of the items resulted in 13 factors. Nine items loaded on a factor which was labeled Sexual Satisfaction. The highest loading for both men and women were for questions indicating "I am satisfied with my sex life," "I have not been deprived sexually," "My love life has not been disappointing." The items appear to have face validity, but add little understanding of the meaning of sexual satisfaction.
A multidimensional assessment of sexual functioning
and satisfaction in couples was developed by LoPiccolo and Steger (1974). The Sexual Interaction Inventory (SII) presents 17 heterosexual behaviors adapted from Bentler's (1968a, 1968b) Guttman scaling of sexual behavior. Each




behavior requires responses to six questions about each behavior rated on a 6-point scale. This results in 102 items. Test-retest reliability over a 2-week interval ranged from +0.53 through +0.90, coefficients that were all significant at the .05 level or better. Statistically significant interval consistency coefficients were also reported. This scale is useful for couples, but is time-consuming to complete. The scoring procedure is complicated and assessment of sexual satisfaction requires assumptions that responses demonstrating agreement between frequency of occurrence and rating of the sexual behavior constitute satisfaction with the sexual experience.
Walter Hudson (1982) has developed nine short form measurement scales to assess a variety of personal and social problems. Each scale contains 25 items scored to range from 0 through 100 where the low score is indicative of an absence of the problem being measured and a high score presents a more severe problem. one of the scales labeled the Index of Sexual Satisfaction (ISS) was designed to measure the level of sexual discord or dissatisfaction perceived by the subject with respect to the sexual relationship with a partner. A few of the 25 items rated from 1 (rarely or none of the time) through 5 (most of all of the time) are "I feel that my partner enjoys our sex life," "My sex life is exciting," "Sex is fun for my partner and me." The author reported a scale




reliability coefficient of .90 and describes content, concurrent, discriminant and construct validity as good.
Another comprehensive inventory of sexual functioning is the Derogatis Sexual Functioning Inventory (DSFI). This inventory includes 10 areas of sexual functioning composed of 258 items. The sexual satisfaction subtest consists of 10 statements; e.g., "Usually, I am satisfied with my sexual partner" and "I am not very interested in sex." These are to be responded to by checking either true or false. A Global Sexual Satisfaction Index (GSSI) is included which simply requests a rating that best describes the individual's present sexual relationship. The rating scale ranges from 1 (could not be worse) through 8 (could not be better). Reliability for the various subtests are reported to be good (Derogatis, 1975, 1978) with both internal consistency and test-retest coefficients relatively high over a 2-week retest interval.
The Pinney Sexual Satisfaction Inventory, developed by Pinney, Gerrard, and Denney (1987), was included in an investigation by its authors of the sexual concerns that were thought to be particularly relevant to women. The inventory consists of rating 36 items on a 7-point Likert scale from 1 (strongly agree) through 7 (strongly disagree). These items were selected from a 51-item pool purported to be relevant to sexual concerns of women. Two




hundred and seventy-five college-aged women who attended a large midwest public university completed the inventory and other assessments of sexual behavior, experiences, and attitudes. Two factors emerged, one labeled General Sexual Satisfaction and the other labeled Satisfaction with Partner. The first factor included 14 items such as, "I feel that nothing is lacking in my sex life," "I am satisfied with the frequency with which I have orgasms," and "I am satisfied with the amount of foreplay involved in my lovemaking." The second factor contains items such as, "I wish I were less inhibited when I make love," "I wish my partner(s) would make me feel more attractive," and "I wish my partner(s) initiated sex more." The largest portion of variance reported in the total PSSI score was found to be that of relationship commitment. This is consistent with other research (Carroll, Volk, & Hyde, 1985).
Even though these scales have the limitations of
being either too simplistic or too complex, developed for only one gender or exclusively for couple assessment, they have clinical and research utility and also contribute to an understanding of the nature of human sexual behavior.
Many psychological theories point to other factors which contribute to sexual development. One of these factors, described in the literature and pertinent to the execution of this investigation, is guilt.




Guilt
Previous researchers have emphasized the inhibitory factor of sex-guilt on sexual behavior. Guilt is conceived as an internalized, transituational condition which is viewed as an avoident or inhibitory type of motivation (Abramson, Mosher, Abramson, & Woychowski, 1977; Janda & O'Grady, 1976; Langston, 1973; Mosher, 1965). Guilt has figured prominently in a number of personality theories (Freud, 1927; Kelly, 1955; Lowen, 1969; Piaget, 1948). There is much evidence to support the concept of the inhibitory effect of sex-guilt. This disposition has been found to mediate an individual's form of sexual expression (Langston, 1973; Mosher, 1973; Mosher & Cross, 1971; Wanlass, Killman, Bella, & Tarnowski, 1983), frequency of intercourse (Love, Sloan, and Schmidt, 1976; Mosher, 1973) and the number of sex partners (Mosher, 1973). Too, sex-guilt has been implicated as a causal factor in the development of sexual dysfunction (Kaplan, 1974; Masters & Johnson, 1970).
Several researchers have addressed problems of sexual behavior by examining personality variables that are potential mediators of sexual activity. Gerrard (1987) compared the relationship between the personality disposition of sex-guilt with reports of sexual activity and contraceptive use of 93 sophomore female college




students in the 1983-84 academic year with comparable samples in 1973-74 and 1978-79 academic years. The hypothesized negative relationship between sexual intercourse and sex-guilt and the hypothesized negative relation between effective contraceptions and sex-guilt were supported. Sexually inactive subjects had significantly higher sex-guilt scores than did sexually active subjects (p <.0001). Additionally sexually active subjects who reported using ineffective contraceptive methods (e.g., rhythm and withdrawl or no birth control methods) had significantly higher sex-guilt scores than did sexually active subjects using effective birth control methods (e.g., the pill).
Mosher and Cross (1971) identified a personality
disposition that they called sex-guilt which manifested behaviorally in resistance to sexual temptation, inhibited sexual behavior, and a disruption of cognitive processes in sex-related situations. To measure sex-guilt as a personality disposition, Mosher (1961, 1966, 1968) developed both male and female versions of forced-choice inventories to measure sex-guilt, hostility guilt, and morality conscience.
Mosher (1965) distinguished between two motives for the inhibition of behavior. One, he defined as fear of external punishment for breaking societal standards. The second, Mosher defined as guilt which has developed as a




result of a past reinforcement history which favored the internalization of moral standards. He stated, Guilt may be defined as a generalized expectancy
for self-mediated punishment (i.e., negative
reinforcement) for violating, anticipating the
violation of, or failure to attain internalized standards of proper behavior. The standards of
proper behavior are seen as encompassing both the internalized prohibitions ("should not's")
and the internalized positively valued
ideal-goals ("ought-to's") which are related to
the individual's feeling of self-worth.
(p. 162)
Janda, Magri, and Barnhart (1977) were interested in clarifying the distinctions made between affective guilt (external controls) and dispositional guilt (internalized controls). They examined responses of 96 female subjects to Mosher's dispositional guilt scale and a Perceived Guilt Index-Trait State which measures affective guilt. They reported results which suggested that dispositional and affective guilt were independent constructs.
Later Janda and O'Grady (1980) devised a sexual
anxiety inventory (SAI) and posited distinctions between anxiety and guilt. The authors suggested that there was a crucial difference between Mosher's conceptualization of guilt and the conceptualization of anxiety. They contended that guilty individuals are concerned with what they think of themselves, whereas sexually anxious individuals are concerned with what others will think of them. Ninety-five male and 135 female undergraduate university students participated. A factor analysis of




items on the SAI and Mosher's Sex-Guilt subscale indicated that items from the two scales tended to load on different factors. A regression analysis demonstrated that both scales reached the .05 level of significance in predicting sexual experience.
During the past decade changes in attitudes about sexual expression have had an impact on the research of human sexual behavior. This has been reflected in college samples which are more sexually experienced and exhibit fewer sexually restrictive attitudes (Abramson & Handschumacher, 1978). Scores on the Mosher sex-guilt scale were also affected by this change in sexual behavior. Mosher (1987) thus developed a Revised Mosher Guilt Inventory in response to drops in the range of guilt score means, truncated during the past 25 years (Mosher & O'Grady, 1979). The present inventory has a limited comparison format (7-point Likert format) arranged in pairs of responses to sentence completion stems. The revised inventory consists of 114 items labeled (a) Sex-Guilt--50 items, (b) Hostility-Guilt--42 items, and
(c) Guilty-Conscience--22 items. The limited comparison format was selected to increase the range of responses and to eliminate problems of a forced-choice format. This revised version was updated and tested on 408 male and female university students and is suitable for both sexes.




Most of the literature thus far presented in this
review has reflected efforts to understand the complexity of sexual satisfaction as experienced by young adults. The articles reviewed illustrate a range of methods employed to accomplish this. An example of another approach is to gain access to the psychological meaning of sexual satisfaction by investigating the way in which an individual construes these experiences. Kelly (1955) suggested that the sharing of personal experiences requires attention to the manner in which a person organizes his or her understanding of an event. The psychology of personal constructs offers a method of gaining access to the individual's inner world. The conceptual framework of George Kelly's personal construct theory was particularly suited for this investigation.
Personal Construct Theory
A recent review of Kelly's ideas was presented by
Jankowicz (1987) in which he noted Kelly's maxim that "if you want to know what people think, why not ask them? They might just tell you" (p. 481).
Kelly (1955) took the position that an individual's interpretation of the world is explained in terms of that person's organized system of personal constructs. He proposed, as the fundamental postulate of his theory that "la person's processes are psychologically channelized by




the ways in which he anticipates events" (Kelly, 1955, p. 46). Bannister and Mair (1968) suggested that this philosophical position led Kelly to assume that "events themselves do not imply their own meanings or classifications, but that events can be appreciated, appear meaningful and be classified only in so far as a person has erected constructions to subsume them" (p. 12).
Eating (1984) presented Kelly's 11 corollaries to the fundamental postulate by defining them in terms of the four functions they serve in the system. The following two corollaries represent the nature of people:
1. construction corollary: "A person anticipates
events by construing their [sic] replications"
(Kelly, 1955, p. 50).
2. dichotomy corollary: "A person's construction
system is composed of a finite number of
dichotomous constructs" (Kelly, 1955, p. 59).
The corollaries which describe the organizational "system" of constructs that represent the personality structure of the individual are
1. organization corollary: "Each person
characteristically evolves, for his convenience
in anticipating events, a construction system
embracing ordinal relationships between
constructs" (Kelly, 1955, p. 56).




2. fragmentation corollary: "A person may
successively employ a variety of construction
subsystems which are inferentially incompatible
with each other" (Kelly, 1955, p. 83).
3. range corollary: "A construct is convenient for
the anticipation of a finite range of events
only" (Kelly, 1955, p. 68).
4. modulation corollary: "The variation in a
person's construction system is limited by the
permeability of the constructs within whose
ranges of convenience the variants lie" (Kelly,
1955, p. 77).
This concern with the structural properties of the constructs is important to the analysis of the system. The following two motivational corollaries are
1. choice corollary: "A person chooses for himself
that alternative in a dichotomized construct
through which he anticipates the greater
possibility for extension and definition of his
system" (Kelly, 1955, p. 64).
2. experience corollary: "A person's construction
system varies as he successively construes the
replication of events" (Kelly, 1955, p. 72). The final three corollaries deal with the nature of relationships that exist among people:




1. individuality corollary: "Persons differ from
each other in their construction of events"
(Kelly, 1955, p. 55).
2. commonality corollary: "To the extent that one
person employs a construction of experience which
is similar to that employed by another, his
processes are psychologically similar to those of
the other person" (Kelly, 1955, p. 90).
3. sociality corollary: "To the extent that one
person construes the construction processes of another, he may play a role in a social process
involving the other person" (Kelly,, 1955,, p. 95).
Personal constructs are found to carry more
information and to provide more psychologically meaningful information than that employed in standard ready-made lists of traits or constructs (Lemon & Warren, 1974). Bem (1974) also noted that self-perception did not demand the same kinds of inferential leaps as the constructs supplied by others.
Kelly believed that we are all "scientists" who have hypotheses (expectations) from our theories (personal construing). We subject these hypotheses to experimental tests and modify the theory accordingly (Fransella & Bannister, 1977).
Kelly was interested in assessing the structural as well as the content of a person's construct system and




devised the repertory grid as a method for exploring the personal construct system. Kelly (1963) defined a construct as the way an individual construes an event as similar or different from another. He posited that we never affirm anything without simultaneously denying something. His essential premise was that all constructs are bipolar in nature. Fransella and Bannister (1977) noted that this bipolarity makes the designing of grids possible.
For the purposes of this study only those aspects of grid analysis pertinent to the research questions are included in Chapter Three. Kelly's (1955) conceptual structures which were of interest for this study are (a) cognitive differentiation, (b) integration, (c) extremity (meaningfulness), and (d) negativity and positivity. Procedures related to the content analysis of the constructs are also presented in Chapter Three.
Summary
In this chapter the researcher has reviewed selective materials illustrating methods and inventories that have addressed the measurement of sexual satisfaction. Some of the literature revealed limited and at times meaningless methods of assessing sexual satisfaction by either resorting to "yes" or "no" responses or including statements to be rated, e.g., "I am satisfied with my sex




life." Neither approach gives much additional information to the meaning of sexual satisfaction. Grater and Downing (1981), Bernstein (1982), Rodriguez (1981), and Garrison (1983) more substantially investigated the meaning of the sexual experience by assessing responses to semantically relevant descriptors of the sexual experience.
This study was undertaken in an attempt to enlarge on the understanding of the meaning of sexual satisfaction by analyzing university students' personal construct systems and by determining the mediating effect of guilt, if any, on the cognitive processes. Details of the methodology of this study are presented in Chapter Three.




CHAPTER THREE
METHOD
The researcher investigated the meaning of sexual
satisfaction in the sexual experiences of male and female university students. Further, the researcher explored the way in which sex-guilt mediated the structure and content of the constructs of sexual satisfaction.
This chapter is divided into several sections. In the first section, the population of interest, sample selection procedures, and the study's sample are discussed. In the second section the data collection is described. The null hypotheses are presented in the third section, followed in the fourth section by a presentation of the instruments used in this study and the variables of interest. In the final section the data analyses are presented.
Population and Sample
The population for this study included 297
undergraduate students attending a major southeastern land-grant university. The students were assigned from the psychology department subject pool during the second week of fall semester 1987. Each was asked to complete a




pretest consisting of the Revised Mosher Guilt Inventory (Mosher, 1987) subscale on sex-guilt and a brief demographic information questionnaire.
On the basis of responses to the demographic
information section and scores on the sex-guilt scale, 83 females and 81 males met screening criteria. To be included in this study, a student was required to be singlet from 18 through 25 years of age, have been or presently be involved in heterosexual sexual experiences, represent the white ethnic group, and agree to be contacted for participation in the study. The investigator contacted students by telephone to confirm their willingness to participate in the study and to schedule time for data collection. Students were drawn from the extreme ranges of ranked scores on the sex-guilt scale. The investigator continued to call until 80 students, 20 high and 20 low sex-guilt males and 20 high and 20 low sex-guilt females, were obtained for this study.
Data Collection
The demographic section and the Revised Mosher Sex-Guilt scale was administered to the psychology department subject pool to screen for students to participate in the research. Each student was given a packet containing an "informed consent" form, the Revised




Mosher Sex-Guilt scale, and the demographic information questionnaire (Appendices A, B, and C). It took approximately 20 minutes for each student to complete the screening. The materials were collected and scored to determine designations of high or low sex-guilt. Students were rank-ordered according to the scores on the sex-guilt scale. On the basis of these scores, a median split was performed with the upper half of the split consisting of the high sex-guilt group and the lower half of the group consisting of the low sex-guilt group used in this study. This was performed for each gender. Mosher (personal communication, April 8, 1987) discussed assignments to high or low sex-guilt levels by use of a median split in the scores from the Mosher Sex-Guilt scale and attested to its utility in previous research. Eighty students, 20 high sex-guilt males, 20 high sex-guilt females, 20 low sex-guilt males, and 20 low sex-guilt females who were white, single, heterosexual, and had been or were currently involved in a sexual relationship were contacted to participate in this study.
The Sexual Satisfaction Repertory Grid was
administered to the selected students in small groups of 3 through 5 depending upon scheduling and availability. Participants reported to a room outfitted with three-sided individual cubicles. It provided maximum privacy for the students while they responded to the items on the grid.




Each student was told that she or he would be
participating in a study focusing on personal sexual experiences. The students were apprised of the confidential nature of the research and of how the information would be used. They were given the option to terminate the testing or to refuse to respond to a question at any time. Each student was assured of confidentiality and had complete control over the length and tone of disclosure. The 80 selected students were instructed to read and sign the informed consent form and were given a packet containing the repertory grid (Appendices E, F, and G). The Sexual Satisfaction Repertory Grid took approximately 60 minutes to complete. The task was structured in the sense that the investigator directed the students through the first steps of the grid procedure to insure understanding of the process. The procedures concerning the grid employed by the investigator are explained in Appendix D.
All students participating in the study were given a subject number code with which to be identified while maintaining anonymity. All participants also received experimental undergraduate credit for their participation in this study.




Hypotheses
This study was designed to compare the structure and content of the sexual satisfaction constructs of high and low sex-guilt male and female university students. The null hypotheses (tested at the .05 level of significance) follow:
1. There is no significant difference between male and female university students' cognitive differentiation
of sexual satisfaction constructs.
2. There is no significant difference between high and
low sex-guilt students' cognitive differentiation of
sexual satisfaction constructs.
3. There is no significant interaction by gender and
sex-guilt of the students' cognitive differentiation
of sexual constructs.
4. There is no significant difference between male and
female university students' level of integration of
sexual satisfaction constructs.
5. There is no significant difference between high and
low sex-guilt students' level of integration of
sexual satisfaction constructs.
6. There is no significant interaction by gender and
sex-guilt of the students' level of integration of
sexual satisfaction constructs.




7. There is no significant difference between male and female university students' extremity of ratings of
sexual satisfaction constructs.
8. There is no significant difference between high sex-guilt and low sex-guilt students' extremity of
ratings of sexual satisfaction constructs.
9. There is no significant interaction by gender and sex-guilt of students' extremity of ratings of sexual
satisfaction constructs.
10. There is no significant difference between male and
female university students' overall negative and
positive perceptions of sexual experiences.
11. There is no significant difference between high
sex-guilt and low sex-guilt students' overall
negative and positive perceptions of sexual
experiences.
12. There is no significant interaction by gender and
sex-guilt of students' overall negative and positive
perception of sexual experiences.
13. Male and female university students do not differ
significantly in the content used to describe the
meaning of sexual satisfaction.
14. High and low sex-guilt individuals do not differ
significantly in the content used to describe the
meaning of sexual satisfaction.




Instrumentation
The psychology department subject pool population was asked to complete a paper-and-pencil demographic information form and the Revised Mosher Sex-Guilt scale. The Sex-Guilt scale and demographic information were used to screen for students to be included in the sample. The sample was requested to complete the Sexual Satisfaction Repertory Grid.
Demographic Information
The demographic information requested was brief and salient to this investigation. The participants in the study sample were screened based on their responses to questions concerning age, ethnic origin, gender, and relationship status. The form was coded to insure confidentiality.
Mosher Guilt Inventories
The Mosher Guilt Inventories (1966, 1968, 1987) were developed to measure three aspects of guilt: Sex-Guilt
(SG), Hostility-Guilt (HG), and Morality-Conscience (MC). Mosher (1961) first developed the Mosher Incomplete Sentences Test (MIST) in a doctoral dissertation (Mosher, 1961) for research that would be acceptable for clinical use (Izard, 1979). Multitrait-multimethod matrices have been provided as evidence of the discriminant validity of the guilt subscales (Mosher, 1966, 1968). The construct validity of the measure of sex-guilt is supported by data




from several investigations (Galbraith, Hahn, & Leiberman, 1968; Lamb, 1968; Mosher, 1965). Additionally, Mosher (1979) reviewed approximately 100 studies appearing prior to or during 1977 that consistently supported the construct validity of the inventory as a valid measure of guilt as a personality disposition (Green & Mosher, 1985; Kelley, 1985; Mosher & Vonderheide, 1985).
Both the male and female versions of the guilt inventories developed by Mosher used a forced-choice format of sentence completion, true-false reponses to measure sex-guilt, hostility guilt, and morality conscience. During the past 25 years the range of guilt scores has been truncated as means have dropped (Mosher & O'Grady, 1979). Mosher submitted the non-overlapping items in both male and female versions of the true-false (233 items) and the forced-choice (151 items) inventories to a sample of 187 male and 221 female university undergraduates for an updated item analysis. To insure discriminant validity among the subscales, 90% of the items had a correlation with its own subscale that was significantly different from the other subscale totals. Because several Morality-Conscience items were too highly correlated with Sex-Guilt, they were dropped from the item pool. The Morality-Conscience subscale was renamed Guilt-Conscience to more adequately reflect the retained items. Mosher (personal communication, March 25, 1987)




suggested that subscales could be omitted or given separately.
Only the subscale Sex-Guilt of the Revised Mosher Guilt Inventory was completed by the students in this study (Appendix C). The revised subscale, now in a limited comparison format, was selected to increase the range of responses available to the students. The 50 items in the subscale are arranged in pairs of responses to the same sentence completion stem in a 7-point Likert format. Students responded to items by rating their responses from: 0, NOT AT ALL TRUE OF (FOR) ME, through 6, EXTREMELY TRUE OF (FOR) ME. The inventory was completed in approximately 10 minutes. Scores were summed for the subscale by reversing the nonguilty alternatives. Scores ranged from 24 low sex-guilt through 210 high sex-guilt. Repertory Grid
The Sexual Satisfaction Repertory Grid is a modified version of Kelly's Role Construct Repertory Grid (Rep Grid). The Rep Grid was originally devised as a measure of different aspects of the structure and content of individual's personal cognitive systems (Kelly, 1955). A grid contains three components: "elements," which define the material upon which the grid will be based; "constructs," which are the ways that the subject is grouping and differentiating among the elements; and a




"linking mechanism" which can show how each element is being assessed regarding each construct (Shaw, 1981).
"Repertory grid technique is basically a method of quantifying and statistically analyzing relationships between the categories used by a subject in performing a complex sorting task" (Adams-Webber, 1979, p. 20). The sorting for this study required that the student consider 3 sexual experiences (elements) from his or her list of 10 past experiences and state how 2 of them were similar in some important way that distinguished them from the third. For example, a subject might have said that 2 of the sexual experiences involved "mutual sexual desire" and the other sexual experience in the sort evoked "reluctant participation." This dimension (reluctance vs. mutual desire) is a personal construct. Personal constructs such as these were elicited from each subject, rather than being provided by the investigator, since this method has proven more meaningful in previous research (Fransella & Bannister, 1977; Lemon & Warren, 1974). This comparing and contrasting continued until 10 bipolar descriptions (constructs) of satisfying and dissatisfying sexual experiences resulted for each student. The grid then consisted of 10 elements which were rated across the 10 bipolar constructs on a 13-point scale (6 0 6).
The rationale, reliability, and validity of repertory grid testing have been described by Slater (1976),




Fransella and Bannister (1977), Pedersen (1958), and Fjeld and Landfield (1961).
Slater (1976) suggested that conventional statistical methods for assessing reliability of psychometric techniques are not appropriate for grids. Instead, it is preferred to assess the stability of the grid indices. It has been commonly found that the indices change very little unless some intervention occurs which results in substantial behavior change.
Fransella and Bannister (1977) concurred with Slater, noting the problems of concepts of reliability, where the grid was concerned simply because there was no such thing as "the" grid. These researchers illustrated this point by citing variations in the form, content, and analysis a grid may take. They further described grid methods which have yielded coefficients of reliability falling largely within the range of 0.60 to 0.80. Epting (1972) and Fjeld and Landfield (1961) investigated the stability of elicited constructs over an interval of time, using the same elements, and found correlations of 0.80 existing between the first and second sets of elicited constructs. It has also been found that subjects being studied would reproduce the same elements when asked to supply them after a 1-week interval with agreement in reproduction of elements from 72% to 77% (Fjeld & Landfield, 1961; Pedersen, 1958).




The repertory grid has no specific content, unlike a questionnaire, and its validity can be discussed only in the sense that one can question whether or not it indicates patterns and relationships in the data generated (Fransella & Bannister, 1977). An assortment of studies have shown the repertory grid technique to be useful in the clinical setting (Ryle & Breen, 1972; Wijesinghe & Wood, 1976; Wright, 1970). Grids have shown efficacy in professional training (Zaken-Greenberg & Neimeyer, 1986). The contention, then, is that the grid is a format for data and as such it is important to investigate the validity of the technique in terms of how effectively it can operationally provide means for testing hypotheses. "A grid may be defined as any form of sorting task which allows for the assessment of relationships between constructs and which yields these primary data in matrix form" (Bannister & Mair, 1968, p. 136). A grid may be constructed to result in many different data matrixes, e.g., 5 X 5, 10 X 15 (Landfield & Cannell, 1987). In this study, a 10 X 10 grid resulted forming a matrix of 100 ratings. Structural analyses of the grid were performed resulting in scores for differentiation, integration, extremity, and negativity or positivity.
Differentiation. Bieri et al. (1966) were the first to operationalize the concept of differentiation of constructs and defined cognitive complexity as "the




capacity to construe social behavior in a multidimensional way. A more cognitively complex person has available a more differentiated system of dimensions for perceiving others' behavior than does a less cognitively complex individual" (p. 185).
Landfield (1971) considered cognitive complexity "as essentially a measure of how many cognitions are interrelated" (p. 58). His construct organization scoring resembles that of Bieri's but Landfield uses the term functionally independent construction (FIC) rather than complexity. Landfield (1977) developed a computer program (ELTORP II) for calculating the number of functionally independent constructions (FIC). Higher FIC scores reflect a capacity to differentiate personal constructs into more independent dimensions as applied to the different elements.
The FIC score reflects information about conceptual differentiation. All construct relationships for all columns and rows are computed for a total FIC score. Hypotheses 1 through 3 in this study refer to variances in this score. This score is derived from the ratings on the 13-point scale (6 to 0 to 6). The overlap or match between row ratings is used to find independent dimensions and clusters of the descriptive dimension. The ratings across the 13-point scale are reduced to sidedness and mid-point ratings. Thus, all left side ratings become




1; right side ratings become 2, and center ratings are designated 0. If all rows are strung together in some way, the FIC score would be only 1 point. If none of the rows show relatedness, the FIC score would be 10 points, or 10 independent constructions. For the purposes of this study only the FIC score for constructs was useful for testing the hypotheses. FIC scores can also be derived for column (element) ratings and a total FIC score for combined construct and element ratings.
Integration. Integration reflects Kelly's (1955) premise that an individual organizes constructs into hierarchies thus building a system of ordinal relationships between constructs to better anticipate events. Landfield (1977) suggested that the ability to form hierarchies could be defined by the variation in the extremity of one's ratings. This variation is called an integration score. The degree of statistical association between constructs reflects the degree of integration of the conceptual structures (Adams-Webber, 1979).
The integration score measures within construct
differentiation and indicates how an individual uses the full range of scoring on the rating scale. The integration level for a particular descriptive construct, for example, dull vs. exciting regarding a sexual experience, is derived by computing how the student has rated all the other sexual experiences on this construct.




To obtain the integration level for the sexual satisfaction constructs used by the student, sidedness is disregarded so the scale is reduced to 7 points, ranging from 0 through 6. The theoretical number of ratings for each scale point 1 through 6 is then equal on the construct scale, since each point occurs twice. Since 0 occurs only once on the 13-point scale, its expected occurrence will always be half that of the other numbers. The analysis for every row consists of subtracting the actual occurrences of the scale point from the expected occurrence for each of the 7 scale points from 0 through
6. This number is then squared and divided by the theoretical expectation for each of the scale points from O through 6. This integration score was used in the analysis of hypotheses 4 through 6.
Extremity. Personal construct language has been
shown to have more relevance and extremity of ratings than those methods which provide descriptors (Bonarius, 1968; Isaacson & Landfield, 1965).
Extremity (meaningfulness) refers to the degree of
rating scale polarization or extremeness of ratings. The events which are described in more extreme ways are considered more meaningful to the subject than are events signaled by more neutral ratings. A high degree of polarization is reflected in ratings of only 6 in the scale (6 0 6). A low extremity rating is shown by the




subject who most frequently rates on scale points of 0 and 1. Meaningfulness has been examined by many investigators including Landfield (1977), Adams-Webber (1979)f Adams-Webber and Mancuso (1983).
Extremity is calculated by computing the absolute values of all ratings along all constructs. This extremity score could range from 0 (mid-point) through 6 (all extreme ratings). This score is used in the analyses of hypotheses 7 through 9.
Positivity or Negativity. The grid matrix used in this study provided measures of how positively or negatively students viewed each of the elicited constructs. Self valence scores were computed for each of the constructs by calculating the mean of the ratings along each construct across all of the experiences. Scores could range from -6 through +6. A percentile score was derived for each subject. In the analysis of hypotheses 9 through 12, an arc sine transform was used to normalize these data.
Grid Constructs. Murphy and Neimeyer (1987) designed a computer program (Auto-Rep) for the content analysis of constructs. This program contains 8 dictionary files and the main Auto-Rep source code. The scoring of constructs was accomplished by means of a computer algorithm which searches the dictionary files for corresponding content codes. Each dictionary entry bears one or more content




codes pertaining to a set of categories connoting semantic meaning.
The 20 categories are (1) Social Interaction, (2)
Forcefulness, (3) Organization, (4) Self-Sufficiency, (5) Status, (6) Factual Description, (7) Intellective, (8) Self-reference, (9) Imagination, (10) Alternatives, (11) Sexual, (12) Morality, (13) External Appearance, (14) Emotional Arousal, (15) Egoism, (16) Tenderness, (17) Time orientation, (18) Involvement, (19) Extreme Qualifiers, and (20) Humor. Landfield (1971) provided an elaborated definition of each category and a step-by-step presentation of their selection. Once codes were derived for each of the constructs, the construct was organized by content category according to its meaning. The elicited constructs not found in the the dictionary files were entered into the files in the categories delineated by the programmed format.
Data Analyses
The analyses of data generated by the repertory grid provided structural measures and content categories. Mosher's sex-guilt measure provided the independent variable of high or low sex-guilt for each subject. The data were analyzed using both parametric and nonparametric statistics. Specifically, hypotheses 1 through 12 were analyzed using a 2 X 2 multivariate analysis of variance




54
(MANOVA). Hypotheses 13 and 14 were analyzed using the chi-square statistic, a non-parametric approach.




CHAPTER FOUR
RESULTS
The major question of this investigation focused on
the manner in which high and low sex-guilt male and female university students understand and organize the meaning of sexual satisfaction in their personal sexual experiences. High and low sex-guilt categories were assessed with the Revised Mosher Guilt Inventory Sex-Guilt subscale. The content and structure of the participant's personal construct system were elicited by the Sexual Satisfaction Repertory Grid.
Fourteen null hypotheses related to the organization of sexual satisfaction constructs and the content categories used for describing these experiences were tested at the .05 level of significance. The results of testing did not permit the rejection of hypotheses 1 through 12, indicating that there were no significant differences between high and low sex-gui lt male and female university students participating in this study on the 5 dependent variables related to organization of constructs. Hypotheses 13 and 14, referring to categorical data, were analyzed using the chi-square statistic. Findings of significant differences (p < .001) led to the rejection of




hypothesis 13, but no significant differences emerged for hypothesis 14. The results of testing for critical values for these 14 hypotheses are sequentially presented in this chapter.
Hypotheses Related to Structure
A 2 X 2 multivariate analysis of variance (MANOVA)
was used to examine the structural properties of the four dependent variables taken from the repertory grid: differentiation (FIC), integration, extremity, and negativity/positivity scores. The results were not found to be statistically significant by the Wilks' Lambda criterion with values related to the effect of guilt being F = 1.11, df = 5/72, p < .36; the effect of gender, F = 1.25, df = 5/72, p < .29; and the interaction of gender and guilt was F = 0.10, df = 5/72, p < .99.
It was stated in hypotheses 1 through 3 that there
were no main effects or interactions related to gender or sex-guilt in the complexity (differentiation) of sexual satisfaction constructs used to describe sexual experiences among the students in this study. Although the overall MANOVA did not reach significance, a univariate analysis revealed an interesting effect for gender with an F value of 3.77 reaching the .05 level. Univariate sex-guilt F values can be seen in Table 1. Table 2 displays the means for differentiation scores of




high and low sex-guilt male and female students participating in this study. Males had more functionally independent constructs (FIC) than did females. This reflects a greater degree of differentiation in the use of sexual satisfaction constructs. High sex-guilt male students had the highest mean scores of the four groups. However critical values for hypothesis 1 did not reach significance in the multivariate analysis of variance. The testing of hypothesis 2 also failed to reach significance (p < .08). The interaction of differentiation differences occurring due to guilt and gender was not supported and the results of testing did not permit the rejection of null hypotheses 1, 2,, or 3.
Critical values for hypotheses 4 through 6 were
indicative of no differences in the level of integration of the use of sexual satisfaction constructs due to gender, guilt, or the combination of the two. As can be seen in Tables 3 and 4, data reveal similarities in the group responses. High sex-guilt students had higher mean integration scores, but these differences did not reach significance. Thus, null hypotheses 4, 5, and 6 were not rejected at the .05 alpha level.
Null hypotheses were also stated for extremity
ratings of constructs. These were posited for gender, sex-guilt, and the interaction of gender and sex-guilt in hypotheses 7 through 9. The investigator failed to reject




hypotheses 7. 8r and 9 due to the critical values reported in Table 5. However, hypothesis 7, related to gender differences in extreme ratings, may be worthy of note due to the univariate critical value reported (p < .09). The means, reported in Table 6, were higher for female than male students, reflecting the use of more extreme ratings of sexual satisfaction descriptions of sexual experiences by the females in this study.
Hypotheses 10 through 12 stated in general that no
differences existed in the positive and negative responses that high and low sex-guilt male and female university students use when rating their sexual experiences. The investigator failed to reject these hypotheses. Data are presented in Tables 7 through 10. Mean scores for these positive and negative ratings are presented in Table 8 and 10. Inspection of the means table reveals that females in this study gave higher negative ratings of sexual experiences than did males. The analyses of variance for positive and negative ratings are presented in Table 7 and
9. Although low sex-guilt students participating in this study had higher positive rating means than did high sex-guilt individuals, these were not significant at the .05 level.




Table 1
Analysis of Variance Summary Table for Differentiation (FIC)
ANOVA
Source df SS MS F p>F
Guilt 1 14.45 14.45 3.02 .08
Gender 1 18.05 18.05 3.77 .05
Guilt*Gender 1 0.80 .80 0.17 .68
Error 76 363.50




Table 2
Mean Differentiation (FIC) Scores for High (HSG) and Low (LSG) Sex-Guilt Males and Females
Subjects HSG LSG Group X
Males (n=40) x 4.00 3.35 3.67
(2.79) (2.03) (2.43)
Females (n=40) 3.25 2.20 2.72
(2.38) (1.23) (1.94)
Group x 3.62 2.77
(2.58) (1.76)
Note. Standard deviations are given in parentheses.




Table 3 Analysis of Variance Summary Table for Integration of Constructs

Source Guilt Gender Guilt*Gender Error

A-NOVA
SS
7033.125 138.601
5.408 453573.98

MS
7033.125 138.601 5.408 5968.07

F
1.18
0.02 0.00

R>F
.28 .87
.97




Table 4
Mean Integration Scores for High (HSG) and Low (LSG) Sex Guilt Males and Females
Subjects HSG LSG Group X
Males (n=40) 161.59 142.32 151.95
(69.39) (91.15) (80.55)
Females (n=40) 158.44 140.20 149.32
(69.36) (77.04) (72.94)
Group x 160.01 141.26
(68.50) (83.30)
Note. Standard deviations are given in parentheses.




Table 5 Analysis of Variance Summary Table for Extremity Ratings of Constructs

Source Guilt Gender Guilt*Gender Error

ANOVA
SS
2784.80 11956.05 627.20 315605.50

MS
2784.80 11956.05 627.20

F
0.67 2.88
0.15

P>F
.41 .09 .69




Table 6
Mean Extremity Scores for High (HSG) and Low (LSG) Sex

Guilt Males and Females

Subjects HSG LSG Group X
Males (n=40) 396.00 402.20 399.10
(60.42) (70.33) (64.79)
Females (n=40) 414.85 432.25 423.55
(71.58) (53.74) (63.09)
Group x 405.42 417.22
(66.08) (63.62)
Note. Standard deviations are given in parentheses.




Table 7
Analysis of Variance Summary Table for Negative Ratings of Sexual Experiences
ANOVA
Source df SS MS F 2>F
Guilt 1 23.112 23.112 0.35 .55
Gender 1 248.51 248.51 3.81 .05
Guilt*Gender 1 3.612 3.612 0.06 .81
Error 76 4958.15 65.23




Table 8
Mean Negativity Scores for High (HSG) and Low (LSG) Sex Guilt Males and Females
Subjects HSG LSG Group X
Males (n=40) 35.7 37.20 36.45
(8.5) (8.09) (8.25)
Females (n=40) 39.65 40.30 39.97
(6.52) (8.9) (7.72)
Group x 37.67 38.75
(7.7) (8.55)
Note. Standard deviations are given in parentheses.




Table 9
Analysis of Variance Summary Table for Positive Ratings of

Sexual Experiences

ANOVA
Source df SS MS F >)F
Guilt 1 189.112 189.112 2.12 .14
Gender 1 6.612 6.612 0.07 .78
Guilt*Gender 1 7.812 7.812 0.09 .76
Error 76 6789.850 89.340




Table 10
Mean Positivity Scores for High (HSG) and Low (LSG) Sex Guilt Males and Females
Subjects HSG LSG Group X
Males (n=40) 50.8 53.30 52.07
(10.15) (10.21) (10.13)
Females (n=40) 49.65 53.35 51.50
(10.77) (5.81) (8.74)
Group x 50.25 53.32
(10.35) (8.20)
Note. Standard deviations are given in parentheses.




Hypotheses Related to Content
Hypothesis 13 stated that no differences would exist in the constructs used by male and female university students to describe their sexually satisfying experiences. Also, as stated in hypothesis 14, there would be no differences in the constructs used by high and low sex-guilt students.
Hypothesis 13 was rejected; the chi-square statistic for gender differences in content used to describe sexual satisfaction was significant. The large chi-aquare value was indicative of differences in the way male and female university students in this study construed sexual satisfaction: chi-square (16, N = 1600) = 73.45, p < .001. The category frequencies for males and females are presented in Figure 1.
The analysis of category distribution as a function
of high or low sex-guilt resulted in the failure to reject hypothesis 14: (chi-square (16, N = 1600) = 22.75, p < .14) but since the critical chi-square value is 26.30 at the .05 level of significance, the finding of a critical value of 22.75 seems worthy of attention. The frequency of category occurrence used by high and low sex-guilt individuals is displayed in Figure 2.
Each of the categories with high enough frequencies to meet chi-square assumptions were then tested for significance. Significant gender differences were found




0C
Self Sufficiency
Status Factual Description
Self Referrence
Alternatives
Sexual
Morality External Appearance Emotional Arousal Egoism
Tenderness
m
Time Orientation
Involvement
m
Extreme Qualifiers
Forcefulness
Environment
m

FREQUENCY
C,. C) C) C)
0 0 (I 0
0 0 0 0 0 II III

m
J ,, wl
.b.,,

'" '':"" ""' " ............... .v~.: +::<:t<+.tt+::+::+xx~tt.:+:+:tt~t ~ ................ .. .i~ii.
.1
mo

m

Ela
1-1 X
(D a




Self Sufficiency
Status
Factual Description
Self Referrence
Alternatives
Sexual
Morality

FREQUENCY
. J NJ Li
0 U 0 UL 0
0 0 0 0 0
1 1

I
-J
-..
-mN
m+

External Appearance
Emotional Arousal
Egoism
Tenderness
at
Time Orientation
Involvement Extreme Qualifiers
Forcefulness
n
Environment

NJ
It- I L3
0

| |

I




for frequency of sexual satisfaction constructs in the following categories:
1. Status
2. Factual Description
3. Self-Reference
4. Alternatives
5. External Experience
6. Egoism
7. Tenderness
Significant sex-guilt differences were found in two categories:
1. Self-Reference
2. Environment
The gender effect was significant at the .05 level in 7 of the 17 categories. The guilt effect was significant in 2 of the 17 categories (see Table 11). Because of infrequency of assignment, 5 categories were eliminated. Those categories were as follows:
1. Social Interaction (n = 0) 2. Organization (n = 0)
3. Intellective (n = 2)
4. Imagination (n = 7)
The category "Environment" was added due to the frequency of occurrence (n = 38) and the absence of a suitable category for subject reference such ass, "took place at the




Table 11
Results of Independent Samples Chi-Square Test

Category df = 1 Gender Guilt

Forcefulness Self-Sufficiency Status
Factual Description Self-Reference Alternatives Sexual
Morality External Appearance Emotional Arousal Egoism
Tenderness Time Orientation Involvement Extreme Qualifiers Environment

chi-square= 1.28 chi-square= 0.47 chi-square= 5.40** chi-square= 5.53** chi-square= 7.53*** chi-square=27.59****

chi-square= 2.31

chi-square= chi-square= chi-square= chi-square= chi-square= chi-square= chi-square= chi-square= chi-square=

0.21
9.80****
0.03
7.75***
3.87*
0.01 0.26 0.27 0.94

chi-square=0.82 chi-square=2.38 chi-square=0.06 chi-square=2.46 chi-square=5.50** chi-square=0.16 chi-square=0.008 chi-square=0.86 chi-square=0.20 chi-square=0.51 chi-square=0.31 chi-square=0.06 chi-square=1.0 chi-square=0.06 chi-square=0.27 chi-square=6.73***

*P<.05. ***.02.

***g<.01.




beach," "in the hot tub," "great atmosphere," and "places other than the bedroom."
Table 11 lists the chi-square critical values for all categories for both gender and guilt. The direction of the findings that reach levels of significance at .05 are illustrated in Figures 1 and 2.
Summary
The data for the 12 structural null hypotheses were analyzed by using a 2 X 2 multivariate analysis of variance (MANOVA) procedure. The chi-square test for independent samples was used for the two categorical hypotheses. Main effects and interactions were examined for statistical significance at the .05 level. Overall, the results of the structural hypotheses multivariate analysis led to a failure to reject the contention that there were no differences in the personal construct paradigms of high and low sex-guilt male and female university students. Though not significant, there were notable gender differences in complexity of construct use (FIC), extremity ratings, and negative ratings of sexual experiences. Significant gender differences were found in constructs used to describe sexual experiences. There were minimal differences in the constructs used by high and low sex-guilt students participating in this study.




CHAPTER FIVE
SUMMARY, DISCUSSION, LIMITATIONS, AND
IMPLICATIONS FOR FUTURE RESEARCH Summary
The primary purpose of the present investigation was to examine the effects of gender and sex-guilt upon the cognitive responses that male and female university students generate to describe sexual satisfaction in their personal sexual experiences. of equal importance was the exploration of the nature of constructs which individuals in this study would elect to represent concepts of sexual satisfaction. Hypotheses for these effects were formulated from a personal construct theory perspective.
Eighty students were selected from a prescreeening of 297 students available from the introductory psychology subject pool during the fall semester 1987. It was required that participants were from the white ethnic group, aged 18 through 25, heterosexual, and have had, or be currently engaged in a sexual relationship. They were assigned to high or low sex-guilt groups based on scores derived from the Mosher Sex-Guilt scale. The 80 university students who participated in this study then completed a 10 X 10 Sexual Satisfaction Repertory Grid.




No significant differences were found in the
structural properties of sexual satisfaction constructs rated by male and female high and low sex-guilt university students. However, there were observable gender differences in this study with male students showing more complexity in their use of personal constructs of sexual satisfaction. Female university students tended to rate their sexual experiences more extremely and more negatively than did male university students.
The content hypotheses were analyzed using the
chi-square statistic. Significant gender differences were found in the types of constructs the male and female university students generated for sexual satisfaction. These results are discussed in this chapter. Also included in this chapter are limitations and implications of the study and recommendations for future research.
Discussion
Hypotheses 1 through 12 refer to the cognitive structures of sexual satisfaction. The cognitive structures of interest to this investigation were
(a) differentiation (FIC), (b) integration, (c) extremity, and (d) negativity and positivity.
It was indicated in hypotheses 1 through 3.
respectively, that no differences in the complexity of cognitive differentiation of sexual satisfaction would




emerge between the male students' ratings and the female students' ratings. In addition, there would be no significant differences due to level of sex-guilt, nor would differences occur in combination of these independent variables. The multivariate analysis of variance did not reach significance at the .05 level. However, the analysis of variance for univariates seemed evidential of differences in cognitive complexity of sexual satisfaction. It would appear from the results of testing that male students in this study were somewhat more differentiated or cognitively complex in sexual satisfaction constructs than were the female students.
Due to the exploratory nature of this research, care must be taken to limit the conclusions drawn from the data. Indeed, there is a paucity of reported personal construct research relative to gender differences in cognitive structure, offering little previous research bearing directly on this issue. Recently, Neimeyer and Metzler (1987) reported significant differences between males and females along measures of cognitive structure. Similarly, males were found to be much more differentiated (p < .003), but, their study measured vocational constructs. Females were reported to demonstrate higher levels of integration (p < .05). No conclusions were drawn from these observed gender differences but directions for future research were identified.




Relatedly, males in this study had higher
differentiation mean scores than did females. In fact, the high sex-guilt males were the most differentiated of the four groups. One possible explanation of this is Bieri's (1955) "vigilance" hypothesis. Bieri suggested that subjects differentiated more between negative, disliked persons, or unpleasant situations than they did between positive events or close associates. He posited a tendency for individuals to be more guarded when confronted with negative or unknown parameters, suggesting that cognitive differentiation serves an adaptive function to help the individual anticipate outcomes. Harrell (1970) offered an alternative interpretation of vigilancece, proposing that social demands create the expectation that one would like a particular person or situation. If that is not the case, she or he is behooved to have a good reason for this aberration and therefore must refine perceptions of disliked persons or situations. The "vigilance" hypothesis offers a plausible explanation for the higher differentiation scores reported for high sex-guilt male students. Considering the definition of sex-guilt as a generalized expectancy for self-mediated punishment for violating or for anticipating violating standards of proper sexual conduct (Mosher & Cross, 1971), it would follow that this group would be the most differentiated in attending to sexual stimuli. It would




seem congruent that this group would be the most cautious and complex in attending to perceived sexual threats. These interpretations may be reasonable, but the results may be explained in other ways as well.
Hypotheses 4 through 6 in the present study indicated that no significant differences would occur among integration scores due to gender, sex-guilt level, or an interaction of the two. Integration scores reflect the degree of statistical association between constructs suggesting that the dimensions are organized into an interrelated system of perceptions. Individuals with a well integrated system of constructs understand nuances of application of the independent constructs in his or her repertoire. Unlike the differentiation and integration research just cited (Neimeyer & Metzler, 1987), no differences were found in this study among the groups. This suggests that there was a comparable facility of application of sexual satisfaction constructs among the high and low sex-guilt male and female university students in this study regardless of the degree of complexity of those constructs.
Hypotheses 7 through 9 were tested for gender and
sex-guilt main effects and interactions for differences in the use of extreme ratings to describe sexually satisfying or dissatisfying experiences. No significant differences were found in extreme ratings used by male and female




university students. Females in this study, however tended to rate sexual experiences more extremely than did males. Higher extremity scores suggest more meaningfulness of the construct to the subject. There is considerable evidence that "elicited" constructs, the procedure performed in this study, produces more extreme ratings by subjects (Adams-Webber, 1979; Landfield, 1971) than do ratings across "provided" constructs. This suggests, according to personal construct literature, more personal meaningfulness to the ratings of ones personal constructs than constructs supplied from another source. This does not answer the question of whether females generally tend to use more extreme ratings than do males or if this is an isolated occurrence. Of additional interest would be gender comparisons in extreme ratings across varying domains. Possibly the extremity of ratings differences in this instance attest to stronger feelings of investment in the constructs associated with the females' sexual experiences.
The final structural null hypotheses of this study,
10 through 12, were tested for differences in positive and negative ratings of the sexual satisfaction constructs used to describe sexual experiences. The hypotheses were that no differences would exist in the ratings used by male and female university students, high and low sex-guilt students, nor high and low sex-guilt male and




female students. Critical values for hypotheses 10 through 12 did not reach .05 levels of significance, therefore, the researcher failed to reject these hypotheses. It is evident from the univariate analysis
(P<.05) that there were some gender differences related to negative ratings. Females in this study tended to rate sexual experiences more negatively than did males. It would be too simplistic to ascribe this observance to any one aspect of the female sexual response. There are many psychosocial and physiological factors which could account for these rating differences. Kaplan (1974) discussed the increasing prevalence of complaints about lack of sexual interest and negative feelings about sexual behavior. This has become so common that it has been classified as inhibited sexual desire (ISD). Males also exhibit ISD but not as often as females. Although this may be a plausible explanation for the more frequent negative ratings used by females in this study, it is only one possible explanation.
Surprisingly, sex-guilt in this investigation, did not have a significant influence upon any of the five dependent variables related to cognitive structure of sexual satisfaction. This is not in keeping with the review of the literature related to the impact of sex-guilt on sexual attitudes and behavior. This is discussed in more detail later in this chapter.




Null hypotheses 13 and 14 concerned the content of the repertory grid. The hypotheses were tested for differences in the ways university students (by gender and level of sex-guilt) construed sexual experiences.
Gender differences in the use of constructs to
describe sexual satisfaction were found to be significant in seven categories (see Table 11). Male students in this study made significantly more references to status, factual descriptions, external appearance, and statements denoting self importance (egoism) that did female students. Females made significantly more self references, used multiple descriptions for an experience (alternatives) and referred to feelings of love, gentleness, and kindness denoting tenderness more often than did males.
Guilt had an effect in two categories in this study--self reference and environment. In both categories, low sex-guilt individuals were more likely to describe sexual satisfaction in terms relating directly to him or herself and to make statements about surroundings than were high sex-guilt individuals.
Gender differences were found to occur in the ways male and female students described sexually satisfying experiences. These differences occurred in directions which are in accord with the literature. Kaplan (1974), Masters, Johnson, & Kolodny (1986), Hite (1982). Carroll,




Volk, and Hyde (1985) described male sexual response as much less sensitive to psychological influences and relatively impervious to the quality of the relationship, especially when they are young. On the other hand, females have been found to focus more on the psychological involvement as a major component of sexual pleasure.
It is important to take note of the large frequencies in categories in which no significant differences were found in this study but were representative of responses which reflect gender agreement on constructs for sexual satisfaction.
The types of responses having to do with the status category were statements such as, "embarrassed to be seen in public with him," "proud, boastful," "accomplishment," and "not proud to be out with her." Factual descriptive responses were observable or measurable such as "naked," 11coarse skin," "clothed," "drunk," or "abstainer." External appearance constructs included "ugly," physically dreadful," "attractive," and "they were the hottest girls." Egoism constructs were "I conquered these chicks," "convenience," "with a bore,," "the girl flipped out over me," and "waste of my time." Self-reference had to do with direct self statements such as "I was naive," "I was happy," "I felt bad about myself," and "I lacked desire." constructs that conveyed several different meanings were placed in the alternative category. Those




included "unsatisfied, ashamed"; "ashamed, noncommitted"; "good time, quick, fun"; and "happy, new, and exciting." Tenderness terms included "gentle," "caressing," "calm," ''caring,'' and ''sweet.''
It may be concluded that male and female university
students in this study were more similar than different in their manner of construing sexual experiences. However, there were significant gender differences in the constructs generated by the university students to describe these experiences. Sex-guilt did not appear to have an effect on the responses of the university students participating in this study.
Limitations of the Study
One apparent limitation of this study was the size of the grid used. Although it was selected because of administrative constraints in consideration of the subjects' time and the sensitive nature of the investigation, a larger, more complex grid may have been a better choice. The resulting 10 X 10 grid was a relatively small representation of the'possible field of sexual satisfaction constructs observed in this study.
Another limitation was the homogeneity of the sample and possible restricted range of sexual experiences from which the data were drawn. The data also are applicable to a white college-aged population only. Whether the data




are representative of other populations can be empirically determined in future research. However, generalizability of the results is restricted.
A final limitation was the use of the revised Mosher Sex-Guilt scale. Although the scale was revised to accommodate a recurring change in the range of guilt scores in recent years (Mosher, 1987), the revised version may not have been in use a long enough time to provide substantial evidence of its utility. The absence of an effect due to the level of sex-guilt may reflect more of a problem of instrumentation than an accurate estimate of the role the disposition of sex-guilt plays in a person' s
construct system.
Implications for Future Research
The study presented here was exploratory in nature and represents a beginning step in understanding the psychological constructions of sexual satisfaction among university students. The results of this study provide information about the cognitive structures related to sexual satisfaction by operationalizing the term in the form of a repertory grid.
The testing of the null hypotheses by using a multivariate analysis of variance failed to yield significant differences at the .05 level for effects of gender and guilt. Therefore, it seems questionable to




interpret univariate tests, but there were interesting gender differences in differentiation, extremity of ratings, and negative ratings of sexual experiences reported in this study.
Future researchers may wish to address these factors and examine other possible differences between male and female university students' understandings of sexually satisfying experiences. If males are more differentiated on this dimension, what might that say about self-disclosure? There is some evidence which corroborates the occurrence noted in this study of gender differences in extremity ratings (Neimeyer, Brown, Metzler, Hagans, & Tanguy, 1988). The notion of female negativity toward sexual experiences presents important implications for counselors and should be investigated further and in more detail.
It is also important to extend and refine the rating system for content categories. This study revealed differences in the way male and female university students described sexually satisfying experiences. Similarities of content were also noted. Would these content categories and cognitive structures be similar with differing age groups, non-student groups, ethnic groups, or socioeconomic groups? There appear to be numerous research possibilities using refinements of the methodological approach initiated in this study. This




87
would result, hopefully, in even more meaningful and worthwhile contributions to the understanding of human sexuality.




APPENDIX A
INFORMED CONSENT




INFORMED CONSENT
PART I
This Inventory is the first part of a research project focused on sexual attitudes and experiences. If you are willing to participate in Part II, a pencil-and- paper questionnaire, write your name in the space provided on the answer sheet. Include your telephone number in the space labeled Social Security Number.
Your name is requested solely for the purpose of contacting you to schedule the time for Part II participation. When all subjects have been contacted, the names and telephone numbers will be destroyed to provide confidentiality.
If you have any questions or concerns regarding this topic, please contact me at 475-1050.
Alma R. Suchman
Principal Investigator
I have read the above passage and understand that my participation in this research is voluntary. I may be included in Part II of this research if I include my name and telephone number. This is only for the purpose of scheduling the research time and all identification with




90
responses will be destroyed when all subjects have been contacted.
Signed Date




INFORMED CONSENT
PART II
You have volunteered to participate in research which will involve paper and pencil responses to an instrument that focuses on sexual experiences. This will require approximately 75 minutes of your time. Your participation and identification with any response is strictly confidential. Should this subject cause embarrassment, you are free to discontinue the testing, refuse to respond to any question, and/or limit the scope of disclosure.
Participation in this research is voluntary and no monetary compensation is being offered to you for participation.
If you have any questions or concerns regarding this topic and your experience as a subject, please contact me at 475-1050.
Alma R. Suchman
Principal Investigator
I have read the above passage and understand that I am free to withdraw my consent at any time. I also




Full Text

PAGE 1

PSYCHOLOGICAL CONSTRUCTION OF SEXUAL SATISFACTION AMONG HIGH AND LOW SEX GUILT MALE AND FEMALE UNIVERSITY STUDENTS By ALMA RODGERS SUCHMAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1988 5

PAGE 2

COPYRIGHT 1988 by ALMA RODGERS SUCHMAN

PAGE 3

ACKNOWLEDGEMENTS I would like to take this opportunity to express my sincere appreciation to my former "boss" and committee chairperson, Dr. Joe Wittmer, who has always come through with the right balance of information, support, and concern throughout my doctoral program. My heartfelt gratitude goes to my committee members, Dr. Franz Epting, who has been tireless in his willingness to provide information, time, and resources to make this dissertation possible; Dr. Harry Grater, who inspired me to take that "first step" to begin graduate school and has always been there for each step along the way; Dr. Phyllis Meek, who never failed to "lift my spirits" during her many roles as instructor, supervisor, and committee member; and Dr. Jaquie Resnick, who generously entrusted me with professional developmental opportunities and gave steadfast support to my research ideas. A special thank you is due to Mr. Steve Wiggins, Dr. Greg Neimeyer, and Ms. April Metzler for their participation, consultation, and interest in the data analyses of this study. My friend, Ms. Joan Davidsen, gave constant encouragement as well as invaluable assistance with the organization of the data. iii

PAGE 4

I would also like to give special acknowledgement to my husband, David Suchman, for his love, support, and constant faith in me. David, my family, and friends have been great sources of comfort and encouragement throughout this endeavor.

PAGE 5

TABLE OF CONTENTS PAGE ACKNOWLEDGEMENTS iii ABSTRACT vii CHAPTER ONE INTRODUCTION 1 Need for the Study 8 Purpose of the Study 8 Definition of Terms 9 Organization of the Remainder of the Study... 10 TWO REVIEW OF THE LITERATURE 12 Sexual Satisfaction 12 Guilt 27 Personal Construct Theory 31 Summary 35 THREE METHOD 37 Population and Sample 37 Data Collection 38 Hypotheses 41 Instrumentation 43 Data Analyses 53 FOUR RESULTS 55 Hypotheses Related to Structure 56 Hypotheses Related to Content 69 Summary 74 FIVE SUMMARY, DISCUSSION, LIMITATIONS AND IMPLICATIONS FOR FUTURE RESEARCH 75 Summary 75 Discussion 76 Limitations of the Study 84 Implications for Future Research 85

PAGE 6

PAGE APPENDIX A INFORMED CONSENT 8 9 B REVISED MG INVENTORY 94 C REVISED MG INVENTORY INSTRUCTIONS 98 D PROCEDURES FOR ADMINISTERING THE REP-TEST 100 E RATING SCALE 105 F SEXUAL EXPERIENCES 107 G SORTS 110 REFERENCES -j 1 1 BIOGRAPHICAL SKETCH 121

PAGE 7

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PSYCHOLOGICAL CONSTRUCTION OF SEXUAL SATISFACTION AMONG HIGH AND LOW SEX-GUILT MALE AND FEMALE UNIVERSITY STUDENTS By Alma Rodgers Suchman April 1988 Chairman: Dr. Joe Wittmer Major Department: Counselor Education The purpose of this study was to assess differences on the bases of gender and level of sex-guilt in cognitive responses made by male and female university students to describe their satisfaction in personal sexual experiences. Also investigated was the nature of constructs they used to represent concepts of sexual satisfaction. Fourteen null hypotheses, related to the organization and content of sexual satisfaction constructs, were formulated in accord with Kelly's personal construct theory. The 80 students who participated in this study were selected from among 297 undergraduate students in an introductory psychology course subject pool at the University of Florida during the fall semester 1987. Criteria for selection included that they be (a) white, (b) from ages 18 through 25, (c) heterosexual, and (d) currently or have been engaged in a sexual relationship.

PAGE 8

The participants were assigned to high or low sex-guilt groups based on scores from the Mosher Sex-Guilt Scale. Students then completed a 10 X 10 Sexual Satisfaction Repertory Grid which yielded sexual satisfaction constructs and ratings. No significant differences (p < .05) were found in the structural properties of sexual satisfaction constructs rated by male and female high and low sex-guilt university students. However, male students in this study did demonstrate somewhat more complexity in their use of personal constructs of sexual satisfaction. The female university students tended to rate their sexual experiences in a more extreme manner and more negatively than did male students. Two content hypotheses were analyzed through use of the chi-square statistic. Significant differences by gender were found in the types of constructs generated for sexual satisfaction. Male students used constructs connotative of status, appearance, and egoism more frequently than did female students. Female students made more references to feelings of tenderness and self -references than did male students. The results of this study provide information about the cognitive structures and content related to constructs of sexual satisfaction. Male and female students were found to be more similar than different in their manner

PAGE 9

of construing sexual experiences. There were significant differences, however, in the content of the constructs generated by male and female students to describe these experiences. Sex-guilt did not appear to be a factor in the structure or content of personal constructs used by the university students participating in this study.

PAGE 10

CHAPTER ONE INTRODUCTION Ours is a culture that makes figuring out how to be physically sexual so complicated that many people never get to the question of how they feel about it. (Gagnon, 1977, p. 212) Few subjects generate as much interest as does human sexuality. The literature of the history of Western civilization is replete with examples of sexual repression and sexual liberation. These reports illuminate many of the myths and misinformation associated with human sexual behavior. Eventually however, the restrictive and repressive influences of the Puritan and Victorian eras gave way to the permissive, "freer" sexual attitudes of the 20th century (Harmatz & Novak, 1983). Indeed, the early 1960s included a sexual revolution (Barrett, 1972; Conley & O'Rourke, 1973). Many factors contributed to the change: (a) availability of birth control pills, (b) protest movements, (c) the new feminism, and (d) greater openness in discussions and displays of sex (Masters, Johnson, & Kolodny, 1986). Groundwork for research in human sexual functioning had been laid in the late 1 9th and early 20th century by the works of Viennese physician Sigmund Freud, German

PAGE 11

psychiatrist Richard von Kraf ft-Ebing, British psychologist Havelock Ellis, and others. Unfortunately their written works were based primarily on pathological populations. It was not until the middle of the 1900s that substantial research involving "normal" people took place. Alfred Kinsey (1953), in an effort to gather information on human sexual behavior, interviewed thousands of men and women across the United States. He was severely criticized for this investigation of sexual behavior (Harmatz & Novak, 1983). His research, however, paved the way for other more direct approaches to investigations of human sexual functioning. William H. Masters and Virginia E. Johnson (1966) focused on the anatomy and physiology of sexual response. Their direct observation and measurement of sexual behavior received shocked cries of outrage from the public and mixed reviews from professionals (Masters, Johnson, & Kolodny, 1986). Until recently, the vast majority of the research on human sexual functioning has been focused upon the physiology of sexuality; the psychological significance of human sexuality has been almost totally neglected (Eysenck, 1971). Masters and Johnson (1966) suggested that sexual adequacy is based on appropriate interactions between biophysical and psychosocial factors. It follows that simply alleviating a physiological sexual dysfunction is not necessarily synonymous with helping a person or

PAGE 12

couple attain sexual satisfaction. Kolodny (1981) found that some people who materially improved their sexual satisfaction did not change a great deal from a viewpoint of their physical sexual functioning. Direct behavioral intervention methods in the treatment of sexual dysfunctions such as erectile failure or premature ejaculation in men and orgasmic dysfunction in women have been effectively used by Masters and Johnson (1970), LoPiccolo and Lobitz (1972), and Kaplan (1974). There have been few attempts, however, to understand the sexual relationship of normal, non-dysfunctional couples who complain of sexual dissatisfaction. Estimates of dissatisfaction with marital sex life range from one third to one half of all couples (Hite, 1976; Kaplan, 1974). Noted sexologists Kaplan (1974); Comfort (1972); Masters, Johnson, and Kolodny (1986); and LoPiccolo and LoPiccolo (1978) have cautioned about the complexity of human sexual response and the limitations of a mechanistic attitude toward treatment of sexual problems. In Sexual Turning Points. Lorna and Philip Sarrel (1984) recognized that, "even though ours is supposedly a nation obsessed with sex, there aren't many opportunities to really find out the details of the deepest meanings of sex in others' lives" (p. 4). This lack of opportunity prevents men and women from knowing the commonality of anxieties and experiences. Segal (1984) provided such an opportunity

PAGE 13

for 2,436 college students in five universities by having them write "sexual autobiographies." These autobiographies revealed that 83% of the males and 78% of the females were sexually active. Among those who had experienced intercourse, the average age of first coitus was 16.9 years for male and 18.3 years for female students. These results correspond to a recent national poll of 1 ,000 teenagers in which it was found that 57% of all American teens were sexually active by the age of 17 ("Teens Speak," 1987). Despite tremendous advances in the understanding of sexual behavior and openness of discussion about sex, and a great deal of sexual activity, sexual partners still complain of a lack of feeling and passion. "So much sex and so little meaning or even fun in it!" exclaimed Rollo May (1969, p. 40). Masters, Johnson, and Kolodny (1986) concurred by stating, "There is little question that personal dissatisfaction with sex is commonplace in our society today. Half of all American marriages are troubled by some form of sexual distress ranging from disinterest and boredom to outright sexual dysfunction" (p. 440). Sexual partners frequently have little insight into the dynamics which contribute to this sense of alienation, loneliness, and depersonalization. Variables related to enhancement or diminishment of sexual response are found in recent research on sexuality

PAGE 14

(Heiman, 1980). Variables studied thus far include anxiety, expectancy, genital response feedback, sex role, frequency of intercourse, and guilt. Unfortunately, however, sexual satisfaction often has been measured by a single question about how satisfied a person is with his or her sex life. Attempts to define the meaning of sexual satisfaction in operational terms have been few. There is no single comprehensive theory regarding the etiology of sexual satisfaction. It is generally agreed, however, that the majority of sexual concerns are created or influenced by psychological rather than biological factors (Kaplan, 1974; Koch, 1983; May 1969). Areas of psychological concern are illuminated by an understanding of the way in which an individual construes satisfying and unsatisfying sexual experiences. How do sexually active individuals perceive their sexual experiences? Are there gender differences in the meaning of sexual satisfaction? How does the personality disposition of sex-guilt affect the way in which an individual construes sexual satisfaction? These questions were addressed in this study. In order to investigate these questions, it was necessary to adopt a theoretical framework from which to operationalize the concept of sexual satisfaction. One theory particularly well-suited to this purpose was the personal construct theory (Kelly, 1955). Personal

PAGE 15

construct theorists have suggested that each individual develops a personal view of sexual satisfaction based upon his or her understanding and experience. The way in which a person interprets particular events is known as a (personal) construct. Kelly (1963) posited that reality is subject to as many ways as we ourselves might invent. Thus, a person's constructs are only limited by the range in which the application would be useful. Kelly rejected the notions of "man-the-biological-organism" or "man-the-lucky-guy, and instead embraced the perspective "man-the-scientist. This person is every person, each performing experiments in life events to improve the prediction and control of certain human phenomena. Particularly useful and important from the clinician's perspective is the implication that "man can enslave himself with his own ideas and then win his freedom again by reconstruing his life" (Kelly, 1963, p. 21). This suggests that the mind is dynamic, an individual formulates a theoretical framework of his/her universe, and the person concocts and re-concocts the design to better anticipate events. The person's view of the event is subject to modification and elaboration, a construing of events defined as the individual's personal construct system (Fransella & Bannister, 1977).

PAGE 16

Kelly (1955) was interested in the content of the person's construct system and analyzed the interrelations between the constructs to determine the meaningfulness, positivity or negativity, integration, and intricacy of the constructs. The technique for analyzing construct content, and for which Kelly is well known, is the repertory grid. Some construct systems may be cognitively complex, thus enabling the person to differentiate to a greater extent in the perception of events or behaviors. Other construct systems may reflect a constricted, less developed pattern. An individual's personal construct system of sexually satisfying experiences is particularly germane to this study. There is evidence to suggest that cognitive processes in sexual situations are constricted by guilt (Mosher & Cross, 1971). Indeed, sex-guilt may influence the way in which sexual satisfaction is perceived (Abramson, Mosher, Abramson, & Woychowski, 1977; Janda, Magri, & Barnhart, 1977). Ways in which sex-guilt mediates sexual behavior are expanded upon in Chapter Two of this study and are accompanied by greater explanation of personal construct theory. In general, however, at present there are no studies concerning the mediating effects of sex-guilt on the personal constructs of sexual satisfaction.

PAGE 17

Need for the Study An understanding of the enormous complexity of human sexual response is essential for the planning of sexual education programs and clinical interventions. An erroneous current assumption is that factual knowledge about the physiological aspects of sex and reproduction alleviates the infinite variety of psychological sexual problems confronting males and females. Farley and Davis (1980) pointed to the absence of research that identifies the psychological sources of sexual satisfaction, even though it is reported as a significant factor in the happiness and discord in relationships. The development of operational definitions of sexual satisfaction other than physiologic is a necessary step in understanding human sexual behavior. Purpose of the Study The purpose of this study was to investigate the ways in which high and low sex-guilt male and female university students understand and organize the meaning of sexual satisfaction in their personal sexual experiences. The manners in which sex-guilt and/or gender mediates the structure and content of the constructs of sexual satisfaction also were examined. More specifically, the following questions were addressed:

PAGE 18

1 What are the differences in cognitive differentiation of sexual constructs among high and low sex-guilt male and female students? 2. What are the differences in level of integration of sexual satisfaction constructs among high and low sex-guilt male and female students? 3. What are the differences in rating extremity of sexual satisfaction constructs among high and low sex-guilt male and female students? 4. What are the differences in overall positive and negative perceptions of sexual experiences among high and low sex-guilt male and female students? 5. What are the differences in the terms used to describe the meaning of sexual satisfaction among high and low sex-guilt male and female students? Definition of Terms The following definitions are used throughout this dissertation. Cognitive differentiation is the process through which an individual construes events in a multidimensional way. A more cognitively complex person has more dimensions for perceiving events than a less cognitively complex person. Construct is a term used in personal construct theory to denote the unique way each person has of analyzing

PAGE 19

10 events that occur in his or her life. The term was originally used by Kelly (1955) to denote a process of comparing and contrasting events. Construc t categories are 20 semantically meaningful groups defined by Landfield (1971). Extremity refers to the degree or amount of scale polarization of ratings found in the subjects' ratings on a 1 3-point scale. Integration refers to the extent of the interrelationship among constructs used by an individual in his/her construct system. Sex-guilt is a term defined by Mosher and Cross (1971) as a personality disposition to a generalized expectancy for self-mediated punishment for violating or anticipating violating standards of proper sexual conduct. High or low sex-guilt are assignments to levels of guilt as determined by subjects' responses to the Revised Mosher Sex-Guilt Inventory. Sexual s atisfaction is defined by the constructs useful to males and females for understanding the meaning of a sexual experience as a satisfying one. Organiz ation of the Remainder of the Study In Chapter Two is described the literature related to the research problem including research related to sexual satisfaction, sex-guilt, personal construct theory, and

PAGE 20

11 the repertory grid technique. Chapter Three contains descriptions of the sampling procedures, instrumentation, data collection procedures, research hypotheses, and data analyses. The results of the study are presented in Chapter Four. Finally, in Chapter Five is presented a summary, discussion, and interpretation of the results, the limitations of the study, and recommendations for future research.

PAGE 21

CHAPTER TWO REVIEW OF THE LITERATURE This chapter includes further delineation of the literature related to the research problem. Those studies specifically related to the issue of sexual satisfaction are described in greater detail. An overview of the personality disposition of sex-guilt and studies illustrating its mediating effect on behavior are included as well as a brief overview of personal construct theory and an elaboration of repertory grid use. Sexual Satisfaction Much research on human sexuality has focused on the physiological problems of sexual dysfunction. Yet, there is evidence of a great deal of sexual dissatisfaction in the larger, nonclinical population (Barbach, 1984; Hite, 1976, 1982; Sarrel & Sarrel, 1984). This salient, recurring theme in human sexual behavior has received scant attention. Rollo May (1969) saw this excessive concern with the technical performance of sex as symptomatic of a culture devoid of a personal meaning of sexual behavior. May stated that "it is not surprising then, in this preoccupation with techniques that the 12

PAGE 22

13 questions typically asked about an act of love-making are not, was there passion or meaning or pleasure in the act? but, How well did I perform?" (p. 44). Masters and Johnson (1970) and Zilbergeld (1978) also emphasized the extent to which performance/goal-oriented sexual behavior detracts from sexual pleasure. At the 71st annual meeting of the International Academy of Sex Research, in Haifa, Israel, Gunter Schmidt (1982) lamented the current dominance of biological and clinical aspects of sex research and presented an argument for interdisciplinary thinking and approaches to sex research. This, however, is hardly a new theme. Krafft-Ebing in Psychopathia Sexualis (1924) asserted that medicine is "a science which ever seeks to trace all psychological manifestations to their anatomical and physiological sources" (p. vi). This concern about the lack of an integration of biological and psychological approaches to sex research arises again and again in the literature. Kaplan (1974) suggested that "despite tremendous recent advances, sexuality still remains a mystery in many respects" (p. 3). In the last several years there have been numerous efforts to further an understanding of sexual behavior. The research reported here is focused on efforts to understand sexual satisfaction.

PAGE 23

14 Investigations of sexual satisfaction are often correlated with sexual performance. Waterman and Chiauzzi (1982) conducted a study which examined the role of orgasm in male and female sexual enjoyment. Although several studies have been reported concluding that orgasm was not necessarily an indicator of sexual satisfaction (Perlman & Abramson, 1982), Waterman and Chiauzzi were concerned about the absence of research that included males as subjects. They concluded in their study of 42 university couples that there was no significant difference between the sexes regarding the relationship between both orgasm and sexual enjoyment. However, their findings suggested that orgasm consistency was significantly related to sexual satisfaction in females but not in males. This finding was attributed to the restricted range of male orgasm consistency among the male subjects and was not considered to be evidence that orgasm consistency was more important a factor for females than males. Both males' and females' pleasure ratings were higher when sexual activity occurred without orgasm. Instrument limitations and/or possibly the lack of pressure to achieve orgasm were two of several reasons proffered by the authors for this result. The latter reason stated does suggest that goal-oriented sexual behavior detracts from sexual pleasure.

PAGE 24

15 Another study of a random sample of 249 college students (130 males and 119 females) provided evidence of gender differences in motives for intercourse. The results are in accord with Bardwick's (1971) findings that commitment and love are the primary motivations for a female's participation in sexual intercourse. However, Bardwick's research did not include a male sample. Carroll, Volk, and Hyde (1985) found that males' motives more often included pleasure, fun, and physical reasons while females' motives were predominantly related to love, commitment, and emotion. Hoon and Hoon (1978) examined responses of 370 women to questions related to demographic and cognitive measures and forms of sexual behavior and expression. They were interested in ratings which would discriminate between respondents who reported high and low levels of orgasm consistency and high and low satisfaction with sexual responsivity. A discriminant analysis showed that women who were most satisfied with their sexual responsivity reported frequent intercourse and orgasm consistency, and enjoyed erotic activities and sensate pleasures. Denney, Field, and Quadagno (1984) were also interested in assertions by Hite (1976, 1982) of differing needs in males and females for sexual satisfaction and compared differences in attitudes toward foreplay, intercourse, and afterplay of 39 male and 49 female

PAGE 25

16 students from a large midwestern university. The students filled out a multiple-choice questionnaire about their own sexual behavior. Included was a question about which behaviors were most satisfying and which most dissatisfying. There were significant sex differences in the students' needs and desires. Women reported more enjoyment in foreplay and afterplay than in intercourse, while men reported intercourse as the most enjoyable aspect of sexual activity. This report did not include a definition of what constituted before or afterplay. Perlman and Abramson (1982) designed a study to assess the relationship between sexual satisfaction and nine theoretically relevant variables taken from Helen Kaplan's (1974) multicausal model of sexual functioning. Social desirability bias and gender differences were examined as well. Fifty-seven male and 91 female married or cohabiting UCLA students and other adults completed self -report measures of attitudes toward one's body, attitudes toward sex, sexual knowledge, experience, anxiety and pleasure evaluation, communication, and life stress. Data were analyzed by computations of correlations and factor analyses. The single most important correlate of sexual satisfaction (r = .78, p < .001) was found to be how pleasurable sex was evaluated by the respondent. Other significant variables included marital happiness

PAGE 26

17 (r = .46, p < .001 ) and the absence of sexual anxiety (r = .24, p < .003). Three factors emerged from the factor analysis: a Sexual Satisfaction factor with significant loadings on satisfaction, pleasure, and sexual activity; an Age or Duration or Relationship factor; and a Sexual Frequency factor. Koch (1983) investigated the etiology of sexual concerns by determining the relationship between the sex-related attitudes and beliefs expressed by college students and each of the 10 highest ranked sexual concerns experienced by these students. The study sample, 899 white, non-married students, voluntarily completed Koch's investigation of sexual satisfaction. Included in this investigation were background questions and two instruments developed for the study. The Sexual Concerns Inventory consisted of 54 potential sexual concerns which could be situational, relational, or physical. The respondent indicated the frequency and impact of each concern during the relationship with his or her last sexual partner. The Sexual Attitudes and Beliefs Scale consisted of 29 Likert-scaled items which measured the respondent's attitudes and beliefs toward body image, sexual self -understanding, sex-roles, sexual self-control, guilt, communication, intimacy, masturbation, rejection, and sexual performance. The 10 highest-ranked sexual concerns were (a) lack of privacy, (b) lack of time, (c)

PAGE 27

18 ejaculating too quickly, (d) commitment, (e) difficulty reaching orgasm, (f) wanting more foreplay, (g) frequency of sexual relations, (h) being turned off by certain behaviors, (i) affection, and (j) inability to reach orgasm. Every one of the 10 highest-ranked sexual concerns, except for lack of time, was found to be significantly related to overall sex-related attitudes and beliefs. These studies have focused on the attitudes, physiological responses, and behavior related to sexual satisfaction. An effort to understand the personal meaning of sexual experience was undertaken by Grater and Downing (1981), Rodriguez (1981), Bernstein (1982), and Garrison (1983). Garrison examined the effects of gender and sex-role orientation (using the Bern Sex-Role Inventory) on the meaning of the sexual experience. His investigation was built on the results of the previous research which had established a meaning of sexual experience questionnaire (MOSE III). This questionnaire categorizes subject's responses on the following subscales: Affiliation, Inadequate/Undesirable, Achievement, Morality, and Erotic/Dominance. Garrison analyzed data from a total of 269 students, 117 females and 152 males. Few gender differences were reported. Male and female differences occurred on Inadequate/ Undesirable and Morality subscales with men rating

PAGE 28

19 themselves greater on both factors. Sex-role differences emerged on all but one subscale — Inadequate/Undesirable. Garrison concluded that sex-role orientation influenced the meaning of sexual experience more significantly than did gender alone. Sex role identity was the focus of a study by Baucom and Aiken (1984) who explored the relationships among masculinity, femininity, marital satisfaction, and response to behavioral marital therapy. Fifty-four married nonclinic couples and 54 maritally distressed clinic couples participated. The results indicated that femininity and masculinity were significantly correlated with marital satisfaction. Androgyny was a less frequent sex role identity for either husband or wife in the clinic group. Conversely, in nonclinic couples, there were more adrogynous husbands and wives than any other sex role type. Another role definition was examined by Frank, Anderson, and Rubinstein (1979). This was an investigation of discrepancies between marital role ideals and marital role behaviors and the relationship of these role tensions to sexual satisfaction. Eighty nonpatient, 50 marital therapy, and 50 sex therapy couples completed a self-report questionnaire. Distressed couples tended to adhere to marital role assignments that were not consistent with ideal role assignments. They were also

PAGE 29

20 more likely to disagree in their perception of who carried out various roles within the marriage. Higher levels of role strain were found to correlate with sexual dissatisfaction. Sexual satisfaction was defined in global terms. Traupmann, Hatfield, and Wexler (1983) interviewed 189 college men and women to determine whether or not equity considerations were important in sexual relations. The investigators expected subjects who reported equitable treatment to have more satisfying sexual relationships than couples in inequitable relationships. Sexual satisfaction was measured by (a) rating on a scale of 8 (extremely satisfied) through 1 (extremely dissatisfied) and (b) respondents' reports as to how satisfied they felt immediately after a sexual encounter. The investigators hypothesized that subjects who reported equitable relationships (a) should be more content and (b) should have more satisfactory sexual relationships than those who feel either overbenef ited or underbenef ited. There was strong support for the first hypothesis and some, much weaker support, for the second hypothesis. Farley and Davis (1980) examined the relationship between sexual satisfaction and the role of similarity/ dissimilarity of personality in 102 married couples. The personality dimensions of interest were extraversionintroversion, neuroticism, and psychoticism (described by

PAGE 30

21 the authors as aggressive, cold, cruel, antisocial, and bizarre characteristics). These were measured by the PEN (psychoticisra, extraversion-introversion, neuroticism inventory). Sexual satisfaction was an overall global self-rating. The authors reported results which indicated that females derived the greatest sexual satisfaction when spouses had almost identical personalities in terms of extraversion-introversion and neuroticism. For males there was no relationship between sexual satisfaction and similarity or dissimilarity on those dimensions, but males indicated greater sexual satisfaction when their spouses were identical to them in psychoticism compared to when spouses were opposite to themselves on this personality dimension. Farley and Davis speculated that females might respond sexually to the more normal aspects of personality while males might respond sexually to less healthy aspects of personality. Personality was also of interest to Heath (1978) who investigated the relationship between personality and sexual compatibility in 60 married men in their early 30s who participated in this longitudinal study. Increased sexual compatibility was significantly related to increased psychological maturity, interpersonal maturity, fulfillment of various adult roles, and to marital happiness.

PAGE 31

22 The literature reveals a variety of approaches to understanding sexual satisfaction. These self-report data have obvious relevance for the assessment of sexual functioning and behavior. While it is known that response bias such as social desirability may affect a subject's responses, research indicates results are meaningful (Eysenck, 1972). A review of the literature reveals several scales which have been constructed to measure sexual experience. Some of the scales are cumulative in style in the Guttman (1950) sense. These scales are cumulative and have a high degree of unidimensionality. This means that a person who has experienced "more advanced" sexual behaviors in a list of sexual behaviors (e.g., sexual intercourse) will more than likely have experienced the "less advanced" behaviors (e.g., nude body caresses), as well. Podell and Perkins (1957), Brady and Levitt (1965), Bentler (1968a, 1968b) and Zuckerman (1973) have developed Guttman-type scales which assess a limited range of sexual behaviors. None of these measures include a measure of sexual satisfaction, however. Several instruments do include an assessment of sexual satisfaction. A few of these have employed a single-item global measurement that required a yes or no response or possibly the single-item was rated on a degree of satisfaction (Heiman, 1980). The major disadvantages

PAGE 32

23 to this approach include restricted variance and response set which is the tendency for people to respond higher when only one question is asked. A somewhat more complex approach, developed by Snyder and Berg (1983), is a sexual dissatisfaction scale comprised of 29 true-false items ranging broadly in content to assess frequency, variety, and quality of sexual activities. This is 1 of 1 1 scales in a comprehensive 280-item Marital Satisfaction Inventory. Another questionnaire used to measure sexual satisfaction is Eysenck's Inventory of Attitudes to Sex (Eysenck, 1976). This inventory consisted of 98 questions requiring "yes" or "no" responses. A factor analysis of the items resulted in 13 factors. Nine items loaded on a factor which was labeled Sexual Satisfaction. The highest loading for both men and women were for questions indicating "I am satisfied with my sex life," "I have not been deprived sexually," "My love life has not been disappointing." The items appear to have face validity, but add little understanding of the meaning of sexual satisfaction. A multidimensional assessment of sexual functioning and satisfaction in couples was developed by LoPiccolo and Steger (1974). The Sexual Interaction Inventory (SII) presents 17 heterosexual behaviors adapted from Bentler's (1968a, 1968b) Guttman scaling of sexual behavior. Each

PAGE 33

24 behavior requires responses to six questions about each behavior rated on a 6-point scale. This results in 102 items. Test-retest reliability over a 2-week interval ranged from +0.53 through +0.90, coefficients that were all significant at the .05 level or better. Statistically significant interval consistency coefficients were also reported. This scale is useful for couples, but is time-consuming to complete. The scoring procedure is complicated and assessment of sexual satisfaction requires assumptions that responses demonstrating agreement between frequency of occurrence and rating of the sexual behavior constitute satisfaction with the sexual experience. Walter Hudson (1982) has developed nine short form measurement scales to assess a variety of personal and social problems. Each scale contains 25 items scored to range from through 100 where the low score is indicative of an absence of the problem being measured and a high score presents a more severe problem. One of the scales labeled the Index of Sexual Satisfaction (ISS) was designed to measure the level of sexual discord or dissatisfaction perceived by the subject with respect to the sexual relationship with a partner. A few of the 25 items rated from 1 (rarely or none of the time) through 5 (most of all of the time) are "I feel that my partner enjoys our sex life," "My sex life is exciting," "Sex is fun for my partner and me." The author reported a scale

PAGE 34

25 reliability coefficient of .90 and describes content, concurrent, discriminant and construct validity as good. Another comprehensive inventory of sexual functioning is the Derogatis Sexual Functioning Inventory (DSFI). This inventory includes 10 areas of sexual functioning composed of 258 items. The sexual satisfaction subtest consists of 10 statements; e.g., "Usually, I am satisfied with my sexual partner" and "I am not very interested in sex." These are to be responded to by checking either true or false. A Global Sexual Satisfaction Index (GSSI) is included which simply requests a rating that best describes the individual's present sexual relationship. The rating scale ranges from 1 (could not be worse) through 8 (could not be better). Reliability for the various subtests are reported to be good (Derogatis, 1975, 1978) with both internal consistency and test-retest coefficients relatively high over a 2-week retest interval. The Pinney Sexual Satisfaction Inventory, developed by Pinney, Gerrard, and Denney (1987), was included in an investigation by its authors of the sexual concerns that were thought to be particularly relevant to women. The inventory consists of rating 3 6 items on a 7-point Likert scale from 1 (strongly agree) through 7 (strongly disagree). These items were selected from a 51 -item pool purported to be relevant to sexual concerns of women. Two

PAGE 35

26 hundred and seventy-five college-aged women who attended a large midwest public university completed the inventory and other assessments of sexual behavior, experiences, and attitudes. Two factors emerged, one labeled General Sexual Satisfaction and the other labeled Satisfaction with Partner. The first factor included 14 items such as, "I feel that nothing is lacking in my sex life," "I am satisfied with the frequency with which I have orgasms," and "I am satisfied with the amount of foreplay involved in my lovemaking." The second factor contains items such as, "I wish I were less inhibited when I make love," "I wish my partner(s) would make me feel more attractive," and "I wish my partner (s) initiated sex more." The largest portion of variance reported in the total PSSI score was found to be that of relationship commitment. This is consistent with other research (Carroll, Volk, & Hyde, 1985). Even though these scales have the limitations of being either too simplistic or too complex, developed for only one gender or exclusively for couple assessment, they have clinical and research utility and also contribute to an understanding of the nature of human sexual behavior. Many psychological theories point to other factors which contribute to sexual development. One of these factors, described in the literature and pertinent to the execution of this investigation, is guilt.

PAGE 36

27 Guilt Previous researchers have emphasized the inhibitory factor of sex-guilt on sexual behavior. Guilt is conceived as an internalized, transituational condition which is viewed as an avoident or inhibitory type of motivation (Abramson, Mosher, Abramson, & Woychowski, 1977; Janda & 0' Grady, 1976; Langston, 1973; Mosher, 1965). Guilt has figured prominently in a number of personality theories (Freud, 1927; Kelly, 1955; Lowen, 1969; Piaget, 1948). There is much evidence to support the concept of the inhibitory effect of sex-guilt. This disposition has been found to mediate an individual's form of sexual expression (Langston, 1973; Mosher, 1973; Mosher & Cross, 1971; Wanlass, Killman, Bella, & Tarnowski, 1983), frequency of intercourse (Love, Sloan, and Schmidt, 1976; Mosher, 1973) and the number of sex partners (Mosher, 1973). Too, sex-guilt has been implicated as a causal factor in the development of sexual dysfunction (Kaplan, 1974; Masters & Johnson, 1970). Several researchers have addressed problems of sexual behavior by examining personality variables that are potential mediators of sexual activity. Gerrard (1987) compared the relationship between the personality disposition of sex-guilt with reports of sexual activity and contraceptive use of 93 sophomore female college

PAGE 37

28 students in the 1983-84 academic year with comparable samples in 1973-74 and 1978-79 academic years. The hypothesized negative relationship between sexual intercourse and sex-guilt and the hypothesized negative relation between effective contraceptions and sex-guilt were supported. Sexually inactive subjects had significantly higher sex-guilt scores than did sexually active subjects (p < .0001). Additionally sexually active subjects who reported using ineffective contraceptive methods (e.g., rhythm and withdrawl or no birth control methods) had significantly higher sex-guilt scores than did sexually active subjects using effective birth control methods (e.g., the pill). Mosher and Cross (1971) identified a personality disposition that they called sex-guilt which manifested behaviorally in resistance to sexual temptation, inhibited sexual behavior, and a disruption of cognitive processes in sex-related situations. To measure sex-guilt as a personality disposition, Mosher (1961, 1966, 1968) developed both male and female versions of forced-choice inventories to measure sex-guilt, hostility guilt, and morality conscience. Mosher (1965) distinguished between two motives for the inhibition of behavior. One, he defined as fear of external punishment for breaking societal standards. The second, Mosher defined as guilt which has developed as a

PAGE 38

29 result of a past reinforcement history which favored the internalization of moral standards. He stated, Guilt may be defined as a generalized expectancy for self-mediated punishment (i.e., negative reinforcement) for violating, anticipating the violation of, or failure to attain internalized standards of proper behavior. The standards of proper behavior are seen as encompassing both the internalized prohibitions ("should not's") and the internalized positively valued ideal-goals ( "ought-to 1 s" ) which are related to the individual's feeling of self-worth, (p. 162) Janda, Magri, and Barnhart (1977) were interested in clarifying the distinctions made between affective guilt (external controls) and dispositional guilt (internalized controls). They examined responses of 96 female subjects to Mosher's dispositional guilt scale and a Perceived Guilt Index-Trait State which measures affective guilt. They reported results which suggested that dispositional and affective guilt were independent constructs. Later Janda and 0' Grady (1980) devised a sexual anxiety inventory (SAI) and posited distinctions between anxiety and guilt. The authors suggested that there was a crucial difference between Mosher's conceptualization of guilt and the conceptualization of anxiety. They contended that guilty individuals are concerned with what they think of themselves, whereas sexually anxious individuals are concerned with what others will think of them. Ninety-five male and 135 female undergraduate university students participated. A factor analysis of

PAGE 39

30 items on the SAI and Mosher's Sex-Guilt subscale indicated that items from the two scales tended to load on different factors. A regression analysis demonstrated that both scales reached the .05 level of significance in predicting sexual experience. During the past decade changes in attitudes about sexual expression have had an impact on the research of human sexual behavior. This has been reflected in college samples which are more sexually experienced and exhibit fewer sexually restrictive attitudes (Abramson & Handschumacher, 1978). Scores on the Mosher sex-guilt scale were also affected by this change in sexual behavior. Mosher (1987) thus developed a Revised Mosher Guilt Inventory in response to drops in the range of guilt score means, truncated during the past 25 years (Mosher & 0' Grady, 1979). The present inventory has a limited comparison format (7-point Likert format) arranged in pairs of responses to sentence completion stems. The revised inventory consists of 1 1 4 items labeled (a) Sex-Guilt--50 items, (b) Hostility-Guilt— 42 items, and (c) Guilty-Conscience--22 items. The limited comparison format was selected to increase the range of responses and to eliminate problems of a forced-choice format. This revised version was updated and tested on 408 male and female university students and is suitable for both sexes.

PAGE 40

31 Most of the literature thus far presented in this review has reflected efforts to understand the complexity of sexual satisfaction as experienced by young adults. The articles reviewed illustrate a range of methods employed to accomplish this. An example of another approach is to gain access to the psychological meaning of sexual satisfaction by investigating the way in which an individual construes these experiences. Kelly (1955) suggested that the sharing of personal experiences requires attention to the manner in which a person organizes his or her understanding of an event. The psychology of personal constructs offers a method of gaining access to the individual's inner world. The conceptual framework of George Kelly's personal construct theory was particularly suited for this investigation. Personal Construct Theory A recent review of Kelly's ideas was presented by Jankowicz (1987) in which he noted Kelly's maxim that "if you want to know what people think, why not ask them? They might just tell you" (p. 481). Kelly (1955) took the position that an individual's interpretation of the world is explained in terms of that person's organized system of personal constructs. He proposed, as the fundamental postulate of his theory that "a person's processes are psychologically channelized by

PAGE 41

32 the ways in which he anticipates events" (Kelly, 1955, p. 46). Bannister and Mair (1968) suggested that this philosophical position led Kelly to assume that "events themselves do not imply their own meanings or classifications, but that events can be appreciated, appear meaningful and be classified only in so far as a person has erected constructions to subsume them" (p. 12). Epting (1984) presented Kelly's 11 corollaries to the fundamental postulate by defining them in terms of the four functions they serve in the system. The following two corollaries represent the nature of people: 1 construction corollary: "A person anticipates events by construing their [ sic ] replications" (Kelly, 1955, p. 50). 2. dichotomy corollary: "A person's construction system is composed of a finite number of dichotomous constructs" (Kelly, 1955, p. 59). The corollaries which describe the organizational "system" of constructs that represent the personality structure of the individual are 1 organization corollary: "Each person characteristically evolves, for his convenience in anticipating events, a construction system embracing ordinal relationships between constructs" (Kelly, 1955, p. 56).

PAGE 42

33 2. fragmentation corollary: "A person may successively employ a variety of construction subsystems which are inferentially incompatible with each other" (Kelly, 1955, p. 83). 3. range corollary: "A construct is convenient for the anticipation of a finite range of events only" (Kelly, 1955, p. 68). 4. modulation corollary: "The variation in a person's construction system is limited by the permeability of the constructs within whose ranges of convenience the variants lie" (Kelly, 1955, p. 77). This concern with the structural properties of the constructs is important to the analysis of the system. The following two motivational corollaries are 1 choice corollary: "A person chooses for himself that alternative in a dichotomized construct through which he anticipates the greater possibility for extension and definition of his system" (Kelly, 1955, p. 64). 2. experience corollary: "A person's construction system varies as he successively construes the replication of events" (Kelly, 1955, p. 72). The final three corollaries deal with the nature of relationships that exist among people:

PAGE 43

34 1 individuality corollary: "Persons differ from each other in their construction of events" (Kelly, 1955, p. 55). 2. commonality corollary: "To the extent that one person employs a construction of experience which is similar to that employed by another, his processes are psychologically similar to those of the other person" (Kelly, 1955, p. 90). 3. sociality corollary: "To the extent that one person construes the construction processes of another, he may play a role in a social process involving the other person" (Kelly, 1955, p. 95). Personal constructs are found to carry more information and to provide more psychologically meaningful information than that employed in standard ready-made lists of traits or constructs (Lemon & Warren, 1974). Bern (1974) also noted that self -perception did not demand the same kinds of inferential leaps as the constructs supplied by others. Kelly believed that we are all "scientists" who have hypotheses (expectations) from our theories (personal construing). We subject these hypotheses to experimental tests and modify the theory accordingly (Fransella & Bannister, 1977). Kelly was interested in assessing the structural as well as the content of a person's construct system and

PAGE 44

35 devised the repertory grid as a method for exploring the personal construct system. Kelly (1963) defined a construct as the way an individual construes an event as similar or different from another. He posited that we never affirm anything without simultaneously denying something. His essential premise was that all constructs are bipolar in nature. Fransella and Bannister (1977) noted that this bipolarity makes the designing of grids possible. For the purposes of this study only those aspects of grid analysis pertinent to the research questions are included in Chapter Three. Kelly's (1955) conceptual structures which were of interest for this study are (a) cognitive differentiation, (b) integration, (c) extremity (meaningfulness) and (d) negativity and positivity. Procedures related to the content analysis of the constructs are also presented in Chapter Three. Summary In this chapter the researcher has reviewed selective materials illustrating methods and inventories that have addressed the measurement of sexual satisfaction. Some of the literature revealed limited and at times meaningless methods of assessing sexual satisfaction by either resorting to "yes" or "no" responses or including statements to be rated, e.g., "I am satisfied with my sex

PAGE 45

36 life." Neither approach gives much additional information to the meaning of sexual satisfaction. Grater and Downing (1981), Bernstein (1982), Rodriguez (1981), and Garrison (1983) more substantially investigated the meaning of the sexual experience by assessing responses to semantically relevant descriptors of the sexual experience. This study was undertaken in an attempt to enlarge on the understanding of the meaning of sexual satisfaction by analyzing university students 1 personal construct systems and by determining the mediating effect of guilt, if any, on the cognitive processes. Details of the methodology of this study are presented in Chapter Three.

PAGE 46

CHAPTER THREE METHOD The researcher investigated the meaning of sexual satisfaction in the sexual experiences of male and female university students. Further, the researcher explored the way in which sex-guilt mediated the structure and content of the constructs of sexual satisfaction. This chapter is divided into several sections. In the first section, the population of interest, sample selection procedures, and the study's sample are discussed. In the second section the data collection is described. The null hypotheses are presented in the third section, followed in the fourth section by a presentation of the instruments used in this study and the variables of interest. In the final section the data analyses are presented. Population and Sample The population for this study included 297 undergraduate students attending a major southeastern land-grant university. The students were assigned from the psychology department subject pool during the second week of fall semester 1987. Each was asked to complete a 37

PAGE 47

38 pretest consisting of the Revised Mosher Guilt Inventory (Mosher, 1987) subscale on sex-guilt and a brief demographic information questionnaire. On the basis of responses to the demographic information section and scores on the sex-guilt scale, 83 females and 81 males met screening criteria. To be included in this study, a student was required to be single, from 18 through 25 years of age, have been or presently be involved in heterosexual sexual experiences, represent the white ethnic group, and agree to be contacted for participation in the study. The investigator contacted students by telephone to confirm their willingness to participate in the study and to schedule time for data collection. Students were drawn from the extreme ranges of ranked scores on the sex-guilt scale. The investigator continued to call until 80 students, 20 high and 20 low sex-guilt males and 20 high and 20 low sex-guilt females, were obtained for this study. Data Collection The demographic section and the Revised Mosher Sex-Guilt scale was administered to the psychology department subject pool to screen for students to participate in the research. Each student was given a packet containing an "informed consent" form, the Revised

PAGE 48

39 Mosher Sex-Guilt scale, and the demographic information questionnaire (Appendices A, B, and C). It took approximately 20 minutes for each student to complete the screening. The materials were collected and scored to determine designations of high or low sex-guilt. Students were rank-ordered according to the scores on the sex-guilt scale. On the basis of these scores, a median split was performed with the upper half of the split consisting of the high sex-guilt group and the lower half of the group consisting of the low sex-guilt group used in this study. This was performed for each gender. Mosher (personal communication, April 8, 1987) discussed assignments to high or low sex-guilt levels by use of a median split in the scores from the Mosher Sex-Guilt scale and attested to its utility in previous research. Eighty students, 20 high sex-guilt males, 20 high sex-guilt females, 20 low sex-guilt males, and 20 low sex-guilt females who were white, single, heterosexual, and had been or were currently involved in a sexual relationship were contacted to participate in this study. The Sexual Satisfaction Repertory Grid was administered to the selected students in small groups of 3 through 5 depending upon scheduling and availability. Participants reported to a room outfitted with three-sided individual cubicles. It provided maximum privacy for the students while they responded to the items on the grid.

PAGE 49

40 Each student was told that she or he would be participating in a study focusing on personal sexual experiences. The students were apprised of the confidential nature of the research and of how the information would be used. They were given the option to terminate the testing or to refuse to respond to a question at any time. Each student was assured of confidentiality and had complete control over the length and tone of disclosure. The 80 selected students were instructed to read and sign the informed consent form and were given a packet containing the repertory grid (Appendices E, F, and G). The Sexual Satisfaction Repertory Grid took approximately 60 minutes to complete. The task was structured in the sense that the investigator directed the students through the first steps of the grid procedure to insure understanding of the process. The procedures concerning the grid employed by the investigator are explained in Appendix D. All students participating in the study were given a subject number code with which to be identified while maintaining anonymity. All participants also received experimental undergraduate credit for their participation in this study.

PAGE 50

41 Hypotheses This study was designed to compare the structure and content of the sexual satisfaction constructs of high and low sex-guilt male and female university students. The null hypotheses (tested at the .05 level of significance) follow: 1 There is no significant difference between male and female university students' cognitive differentiation of sexual satisfaction constructs. 2. There is no significant difference between high and low sex-guilt students' cognitive differentiation of sexual satisfaction constructs. 3. There is no significant interaction by gender and sex-guilt of the students' cognitive differentiation of sexual constructs. 4. There is no significant difference between male and female university students' level of integration of sexual satisfaction constructs. 5. There is no significant difference between high and low sex-guilt students' level of integration of sexual satisfaction constructs. 6. There is no significant interaction by gender and sex-guilt of the students' level of integration of sexual satisfaction constructs.

PAGE 51

42 7. There is no significant difference between male and female university students' extremity of ratings of sexual satisfaction constructs. 8. There is no significant difference between high sex-guilt and low sex-guilt students' extremity of ratings of sexual satisfaction constructs. 9. There is no significant interaction by gender and sex-guilt of students' extremity of ratings of sexual satisfaction constructs. 10. There is no significant difference between male and female university students' overall negative and positive perceptions of sexual experiences. 11. There is no significant difference between high sex-guilt and low sex-guilt students' overall negative and positive perceptions of sexual experiences. 12. There is no significant interaction by gender and sex-guilt of students' overall negative and positive perception of sexual experiences. 13. Male and female university students do not differ significantly in the content used to describe the meaning of sexual satisfaction. 14. High and low sex-guilt individuals do not differ significantly in the content used to describe the meaning of sexual satisfaction.

PAGE 52

43 Instrumentation The psychology department subject pool population was asked to complete a paper-and-pencil demographic information form and the Revised Mosher Sex-Guilt scale. The Sex-Guilt scale and demographic information were used to screen for students to be included in the sample. The sample was requested to complete the Sexual Satisfaction Repertory Grid. Demographic Information The demographic information requested was brief and salient to this investigation. The participants in the study sample were screened based on their responses to questions concerning age, ethnic origin, gender, and relationship status. The form was coded to insure confidentiality. Mosher Guilt Inventories The Mosher Guilt Inventories (1966, 1968, 1987) were developed to measure three aspects of guilt: Sex-Guilt (SG), Hostility-Guilt (HG), and Morality-Conscience (MC). Mosher (1961) first developed the Mosher Incomplete Sentences Test (MIST) in a doctoral dissertation (Mosher, 1961) for research that would be acceptable for clinical use (Izard, 1979). Multitrait-multimethod matrices have been provided as evidence of the discriminant validity of the guilt subscales (Mosher, 1966, 1968). The construct validity of the measure of sex-guilt is supported by data

PAGE 53

44 from several investigations (Galbraith, Hahn, & Leiberman, 1968; Lamb, 1968; Mosher, 1965). Additionally, Mosher (1979) reviewed approximately 100 studies appearing prior to or during 1977 that consistently supported the construct validity of the inventory as a valid measure of guilt as a personality disposition (Green & Mosher, 1985; Kelley, 1985; Mosher & Vonderheide, 1985). Both the male and female versions of the guilt inventories developed by Mosher used a forced-choice format of sentence completion, true-false reponses to measure sex-guilt, hostility guilt, and morality conscience. During the past 25 years the range of guilt scores has been truncated as means have dropped (Mosher & O' Grady, 1979). Mosher submitted the non-overlapping items in both male and female versions of the true-false (233 items) and the forced-choice (151 items) inventories to a sample of 187 male and 221 female university undergraduates for an updated item analysis. To insure discriminant validity among the subscales, 90% of the items had a correlation with its own subscale that was significantly different from the other subscale totals. Because several Morality-Conscience items were too highly correlated with Sex-Guilt, they were dropped from the item pool. The Morality-Conscience subscale was renamed Guilt-Conscience to more adequately reflect the retained items. Mosher (personal communication, March 25, 1987)

PAGE 54

45 suggested that subscales could be omitted or given separately. Only the subscale Sex-Guilt of the Revised Mosher Guilt Inventory was completed by the students in this study (Appendix C). The revised subscale, now in a limited comparison format, was selected to increase the range of responses available to the students. The 50 items in the subscale are arranged in pairs of responses to the same sentence completion stem in a 7-point Likert format. Students responded to items by rating their responses from: 0, NOT AT ALL TRUE OF (FOR) ME, through 6, EXTREMELY TRUE OF (FOR) ME. The inventory was completed in approximately 10 minutes. Scores were summed for the subscale by reversing the nonguilty alternatives. Scores ranged from 24 low sex-guilt through 210 high sex-guilt Repertory Grid The Sexual Satisfaction Repertory Grid is a modified version of Kelly's Role Construct Repertory Grid (Rep Grid). The Rep Grid was originally devised as a measure of different aspects of the structure and content of individual's personal cognitive systems (Kelly, 1955). A grid contains three components: "elements," which define the material upon which the grid will be based; "constructs," which are the ways that the subject is grouping and differentiating among the elements; and a

PAGE 55

46 "linking mechanism" which can show how each element is being assessed regarding each construct (Shaw, 1981). "Repertory grid technique is basically a method of quantifying and statistically analyzing relationships between the categories used by a subject in performing a complex sorting task" (Adams-Webber, 1979, p. 20). The sorting for this study required that the student consider 3 sexual experiences (elements) from his or her list of 10 past experiences and state how 2 of them were similar in some important way that distinguished them from the third. For example, a subject might have said that 2 of the sexual experiences involved "mutual sexual desire" and the other sexual experience in the sort evoked "reluctant participation." This dimension (reluctance vs. mutual desire) is a personal construct. Personal constructs such as these were elicited from each subject, rather than being provided by the investigator, since this method has proven more meaningful in previous research (Fransella & Bannister, 1977; Lemon & Warren, 1974). This comparing and contrasting continued until 10 bipolar descriptions (constructs) of satisfying and dissatisfying sexual experiences resulted for each student. The grid then consisted of 10 elements which were rated across the 10 bipolar constructs on a 1 3-point scale (6-0-6). The rationale, reliability, and validity of repertory grid testing have been described by Slater (1976),

PAGE 56

47 Fransella and Bannister (1977), Pedersen (1958), and Fjeld and Landfield (1961 ). Slater (1976) suggested that conventional statistical methods for assessing reliability of psychometric techniques are not appropriate for grids. Instead, it is preferred to assess the stability of the grid indices. It has been commonly found that the indices change very little unless some intervention occurs which results in substantial behavior change. Fransella and Bannister (1977) concurred with Slater, noting the problems of concepts of reliability, where the grid was concerned simply because there was no such thing as "the" grid. These researchers illustrated this point by citing variations in the form, content, and analysis a grid may take. They further described grid methods which have yielded coefficients of reliability falling largely within the range of 0.60 to 0.80. Epting (1972) and Fjeld and Landfield (1961) investigated the stability of elicited constructs over an interval of time, using the same elements, and found correlations of 0.80 existing between the first and second sets of elicited constructs. It has also been found that subjects being studied would reproduce the same elements when asked to supply them after a 1 -week interval with agreement in reproduction of elements from 72% to 77% (Fjeld & Landfield, 1961; Pedersen, 1958).

PAGE 57

48 The repertory grid has no specific content, unlike a questionnaire, and its validity can be discussed only in the sense that one can question whether or not it indicates patterns and relationships in the data generated (Fransella & Bannister, 1977). An assortment of studies have shown the repertory grid technique to be useful in the clinical setting (Ryle & Breen, 1972; Wijesinghe & Wood, 1976; Wright, 1970). Grids have shown efficacy in professional training (Zaken-Greenberg & Neimeyer, 1986). The contention, then, is that the grid is a format for data and as such it is important to investigate the validity of the technique in terms of how effectively it can operationally provide means for testing hypotheses. "A grid may be defined as any form of sorting task which allows for the assessment of relationships between constructs and which yields these primary data in matrix form" (Bannister & Mair, 1968, p. 136). A grid may be constructed to result in many different data matrixes, e.g., 5X5, 10 X 15 (Landfield & Cannell, 1987). In this study, a 10 X 10 grid resulted forming a matrix of 100 ratings. Structural analyses of the grid were performed resulting in scores for differentiation, integration, extremity, and negativity or positivity. Differentiation Bieri et al. (1966) were the first to operationalize the concept of differentiation of constructs and defined cognitive complexity as "the

PAGE 58

49 capacity to construe social behavior in a multidimensional way. A more cognitively complex person has available a more differentiated system of dimensions for perceiving others' behavior than does a less cognitively complex individual" (p. 185). Landfield (1971) considered cognitive complexity "as essentially a measure of how many cognitions are interrelated" (p. 58). His construct organization scoring resembles that of Bieri's but Landfield uses the term functionally independent construction (FIC) rather than complexity. Landfield (1977) developed a computer program (ELTORP II) for calculating the number of functionally independent constructions (FIC). Higher FIC scores reflect a capacity to differentiate personal constructs into more independent dimensions as applied to the different elements. The FIC score reflects information about conceptual differentiation. All construct relationships for all columns and rows are computed for a total FIC score. Hypotheses 1 through 3 in this study refer to variances in this score. This score is derived from the ratings on the 13-point scale (6 to to 6). The overlap or match between row ratings is used to find independent dimensions and clusters of the descriptive dimension. The ratings across the 13-point scale are reduced to sidedness and mid-point ratings. Thus, all left side ratings become

PAGE 59

50 1; right side ratings become 2, and center ratings are designated 0. If all rows are strung together in some way, the FIC score would be only 1 point. If none of the rows show relatedness, the FIC score would be 10 points, or 10 independent constructions. For the purposes of this study only the FIC score for constructs was useful for testing the hypotheses. FIC scores can also be derived for column (element) ratings and a total FIC score for combined construct and element ratings. Integration. Integration reflects Kelly's (1955) premise that an individual organizes constructs into hierarchies thus building a system of ordinal relationships between constructs to better anticipate events. Landfield (1977) suggested that the ability to form hierarchies could be defined by the variation in the extremity of one's ratings. This variation is called an integration score. The degree of statistical association between constructs reflects the degree of integration of the conceptual structures (Adams -Webber, 1979). The integration score measures within construct differentiation and indicates how an individual uses the full range of scoring on the rating scale. The integration level for a particular descriptive construct, for example, dull vs. exciting regarding a sexual experience, is derived by computing how the student has rated all the other sexual experiences on this construct.

PAGE 60

51 To obtain the integration level for the sexual satisfaction constructs used by the student, sidedness is disregarded so the scale is reduced to 7 points, ranging from through 6. The theoretical number of ratings for each scale point 1 through 6 is then equal on the construct scale, since each point occurs twice. Since occurs only once on the 13 -point scale, its expected occurrence will always be half that of the other numbers. The analysis for every row consists of subtracting the actual occurrences of the scale point from the expected occurrence for each of the 7 scale points from through 6. This number is then squared and divided by the theoretical expectation for each of the scale points from through 6. This integration score was used in the analysis of hypotheses 4 through 6. Extremity. Personal construct language has been shown to have more relevance and extremity of ratings than those methods which provide descriptors (Bonarius, 1968; Isaacson & Landfield, 1965). Extremity (meaningfulness) refers to the degree of rating scale polarization or extremeness of ratings. The events which are described in more extreme ways are considered more meaningful to the subject than are events signaled by more neutral ratings. A high degree of polarization is reflected in ratings of only 6 in the scale (6-0-6). A low extremity rating is shown by the

PAGE 61

52 subject who most frequently rates on scale points of and 1 Meaningfulness has been examined by many investigators including Landfield (1977), Adams-Webber (1979), Adams-Webber and Mancuso (1983). Extremity is calculated by computing the absolute values of all ratings along all constructs. This extremity score could range from (mid-point) through 6 (all extreme ratings). This score is used in the analyses of hypotheses 7 through 9. Positivity or Negativity. The grid matrix used in this study provided measures of how positively or negatively students viewed each of the elicited constructs. Self valence scores were computed for each of the constructs by calculating the mean of the ratings along each construct across all of the experiences. Scores could range from -6 through +6. A percentile score was derived for each subject. In the analysis of hypotheses 9 through 12, an arc sine transform was used to normalize these data. Grid Constructs. Murphy and Neimeyer (1987) designed a computer program (Auto-Rep) for the content analysis of constructs. This program contains 8 dictionary files and the main Auto-Rep source code. The scoring of constructs was accomplished by means of a computer algorithm which searches the dictionary files for corresponding content codes. Each dictionary entry bears one or more content

PAGE 62

53 codes pertaining to a set of categories connoting semantic meaning. The 20 categories are (1) Social Interaction, (2) Forcefulness, (3) Organization, (4) Self-sufficiency, (5) Status, (6) Factual Description, (7) Intellective, (8) Self -reference, (9) Imagination, (10) Alternatives, (11) Sexual, (12) Morality, (13) External Appearance, (14) Emotional Arousal, (15) Egoism, (16) Tenderness, (17) Time Orientation, (18) Involvement, (19) Extreme Qualifiers, and (20) Humor. Landfield (1971) provided an elaborated definition of each category and a step-by-step presentation of their selection. Once codes were derived for each of the constructs, the construct was organized by content category according to its meaning. The elicited constructs not found in the the dictionary files were entered into the files in the categories delineated by the programmed format. Data Analyses The analyses of data generated by the repertory grid provided structural measures and content categories. Mosher's sex-guilt measure provided the independent variable of high or low sex-guilt for each subject. The data were analyzed using both parametric and nonparametric statistics. Specifically, hypotheses 1 through 12 were analyzed using a 2 X 2 multivariate analysis of variance

PAGE 63

54 (MANOVA). Hypotheses 13 and 14 were analyzed using the chi-square statistic, a non-parametric approach.

PAGE 64

CHAPTER FOUR RESULTS The major question of this investigation focused on the manner in which high and low sex-guilt male and female university students understand and organize the meaning of sexual satisfaction in their personal sexual experiences. High and low sex-guilt categories were assessed with the Revised Mosher Guilt Inventory Sex-Guilt subscale. The content and structure of the participant's personal construct system were elicited by the Sexual Satisfaction Repertory Grid. Fourteen null hypotheses related to the organization of sexual satisfaction constructs and the content categories used for describing these experiences were tested at the .05 level of significance. The results of testing did not permit the rejection of hypotheses 1 through 12, indicating that there were no significant differences between high and low sex-guilt male and female university students participating in this study on the 5 dependent variables related to organization of constructs. Hypotheses 13 and 14, referring to categorical data, were analyzed using the chi-square statistic. Findings of significant differences ( £ < .001) led to the rejection of 55

PAGE 65

56 hypothesis 13, but no significant differences emerged for hypothesis 14. The results of testing for critical values for these 1 4 hypotheses are sequentially presented in this chapter. Hypotheses Related to Structure A 2 X 2 multivariate analysis of variance (MANOVA) was used to examine the structural properties of the four dependent variables taken from the repertory grid: differentiation (FIC), integration, extremity, and negativity /positivity scores. The results were not found to be statistically significant by the Wilks' Lambda criterion with values related to the effect of guilt being F = 1.11, df = 5/72, p_ < .36; the effect of gender, F = 1.25, df = 5/72, £ < .29; and the interaction of gender and guilt was F = 0.10, df = 5/72, £ < .99. It was stated in hypotheses 1 through 3 that there were no main effects or interactions related to gender or sex-guilt in the complexity (differentiation) of sexual satisfaction constructs used to describe sexual experiences among the students in this study. Although the overall MANOVA did not reach significance, a univariate analysis revealed an interesting effect for gender with an F value of 3.77 reaching the .05 level. Univariate sex-guilt F values can be seen in Table 1 Table 2 displays the means for differentiation scores of

PAGE 66

57 high and low sex-guilt male and female students participating in this study. Males had more functionally independent constructs (FIC) than did females. This reflects a greater degree of differentiation in the use of sexual satisfaction constructs. High sex-guilt male students had the highest mean scores of the four groups. However critical values for hypothesis 1 did not reach significance in the multivariate analysis of variance. The testing of hypothesis 2 also failed to reach significance (p_ < .08). The interaction of differentiation differences occurring due to guilt and gender was not supported and the results of testing did not permit the rejection of null hypotheses 1 2, or 3. Critical values for hypotheses 4 through 6 were indicative of no differences in the level of integration of the use of sexual satisfaction constructs due to gender, guilt, or the combination of the two. As can be seen in Tables 3 and 4, data reveal similarities in the group responses. High sex-guilt students had higher mean integration scores, but these differences did not reach significance. Thus, null hypotheses 4, 5, and 6 were not rejected at the .05 alpha level. Null hypotheses were also stated for extremity ratings of constructs. These were posited for gender, sex-guilt, and the interaction of gender and sex-guilt in hypotheses 7 through 9. The investigator failed to reject

PAGE 67

58 hypotheses 7 8, and 9 due to the critical values reported in Table 5. However, hypothesis 7 related to gender differences in extreme ratings, may be worthy of note due to the univariate critical value reported (p_ < .09). The means, reported in Table 6, were higher for female than male students, reflecting the use of more extreme ratings of sexual satisfaction descriptions of sexual experiences by the females in this study. Hypotheses 10 through 12 stated in general that no differences existed in the positive and negative responses that high and low sex-guilt male and female university students use when rating their sexual experiences. The investigator failed to reject these hypotheses. Data are presented in Tables 7 through 10. Mean scores for these positive and negative ratings are presented in Table 8 and 10. Inspection of the means table reveals that females in this study gave higher negative ratings of sexual experiences than did males. The analyses of variance for positive and negative ratings are presented in Table 7 and 9. Although low sex-guilt students participating in this study had higher positive rating means than did high sex-guilt individuals, these were not significant at the .05 level.

PAGE 68

59 Table 1

PAGE 69

60 Table 2 Mean Dif ferentiation (FIC) Scores for High (HSG) an d Low (LSG) Sex-Guilt Males and Females Subj ects HSG LSG Males (n=40) x 4.00 3.35 Females (n=40) 3.25 2.20 Group x 3.62 2.77 (2.58) (1.76) Group X 3.67 (2.79) (2.03) (2.43) 2.72 (2.38) (1.23) (1.94) Note. Standard deviations are given in parentheses.

PAGE 70

61 Table 3 Analysis of Variance Summary Table for Integration of Constructs ANOVA Source df SS MS F £>F Guilt 1 7033.125 7033.125 1.18 .28 Gender 1 138.601 138.601 0.02 .87 Guilt*Gender 1 5.408 5.408 0.00 .97 Error 76 453573.98 5968.07

PAGE 71

62 Table 4 Mean Integration Sco res for High (HSG) and Low (LSG ) Sex Guilt Males and Females Group x 160.01 141.26 (68.50) (83.30) Subjects HSG LSG Group X Males (n=40) 161.59 142.32 151 95 (69.39) (91.15) (80i55) Females (n=40) 158.44 140.20 149.32 (69.36) (77.04) (72.*94) Note. Standard deviations are given in parentheses.

PAGE 72

63 Table 5 Analysi s of Variance Summary Table for Extremity Ratings of Constructs ~ ANOVA Source df SS MS F £>F Guilt 1 2784.80 2784.80 0.67 .41 Gender 1 11956.05 11956.05 2.88 .09 Guilt*Gender 1 627.20 627.20 0.15 .69 Error 76 315605.50

PAGE 73

64 Table 6 Mean

PAGE 74

65 Table 7 Analysis of Varia nce Summary Table for Negative Ratin gs of Sexual Experiences ~~~~ ANOVA Source df SS MS £>F Guilt 1 23.112 23.112 0.35 .55 Gender 1 248.51 248.51 3.81 .05 Guilt*Gender 1 3 6 12 3.612 0.06 .81 Error 76 4958.15 65.23

PAGE 75

66 Table 8 Mean Negativity Scores for High (HSG) and Low (LSG) Sex Guilt Males and Females Males (n=40) Females (n=40) Group x Subjects HSG LSG Group X 35

PAGE 76

67 Table 9 Analysis of Variance Summary Table for Positive Ratings of Sexual Experiences ANOVA Source df SS MS F p_>F Guilt 1 189.112 189.112 2.12 .14 Gender 1 6.612 6.612 0.07 .78 Guilt*Gender 1 7.812 7.812 0.09 .76 Error 76 6789.850 89.340

PAGE 77

68 Table 10 Mean Positivity Scores for High (HSG) and Low (LSG) Sex Guilt Males and Females Subjects HSG LSG Group X Males

PAGE 78

69 Hypotheses Related to Content Hypothesis 13 stated that no differences would exist in the constructs used by male and female university students to describe their sexually satisfying experiences. Also, as stated in hypothesis 14, there would be no differences in the constructs used by high and low sex-guilt students. Hypothesis 13 was rejected; the chi-square statistic for gender differences in content used to describe sexual satisfaction was significant. The large chi-aquare value was indicative of differences in the way male and female university students in this study construed sexual satisfaction: chi-square (16, N = 1600) = 73.45, £ < .001. The category frequencies for males and females are presented in Figure 1 The analysis of category distribution as a function of high or low sex-guilt resulted in the failure to reject hypothesis 14: (chi-square ( 1 6, N = 1 600 ) = 22.75, £ < .14) but since the critical chi-square value is 26.30 at the .05 level of significance, the finding of a critical value of 22.75 seems worthy of attention. The frequency of category occurrence used by high and low sex-guilt individuals is displayed in Figure 2. Each of the categories with high enough frequencies to meet chi-square assumptions were then tested for significance. Significant gender differences were found

PAGE 79

70

PAGE 80

71 5U3UUUOJTAU3 sseujnjaoaoj sjBTjfXEnO auiBj^xa 1.U3U13AJOAUI UOT^E^UaT^O 3i XI C o H 4-1 X) -H M 4-1 CO XDN3n03HJ

PAGE 81

72 for frequency of sexual satisfaction constructs in the following categories: 1 Status 2. Factual Description 3. Self-Reference 4. Alternatives 5. External Experience 6. Egoism 7. Tenderness Significant sex-guilt differences were found in two categories: 1 Self-Reference 2. Environment The gender effect was significant at the .05 level in 7 of the 17 categories. The guilt effect was significant in 2 of the 17 categories (see Table 11). Because of infrequency of assignment, 5 categories were eliminated. Those categories were as follows: 1. Social Interaction (n = 0) 2. Organization (n = 0) 3. Intellective (n = 2) 4. Imagination (n = 7) The category "Environment" was added due to the frequency of occurrence (n = 38) and the absence of a suitable category for subject reference such as, "took place at the

PAGE 82

Table 11 Results of Independent Samples Chi-Square Test 73 Category df = 1 Gender Guilt Forcefulness Self -Sufficiency Status Factual Description Self -Reference Alternatives Sexual Morality External Appearance Emotional Arousal Egoism Tenderness Time Orientation Involvement Extreme Qualifiers Environment chi-square= 1 .28 chi chi-square= 0.47 chi chi-square= 5.40** chi chi-square= 5.53** chi chi-square= 7.53*** chi chi-square=27.59**** chi chi-square= 2.31 chi chi-square= 0.21 chi chi-square= 9.80**** chi chi-square= 0.03 chi chi-square= 7.75*** chi chi-square= 3.87* chi chi-square= 0.01 chi chi-square= 0.26 chi chi-square= 0.27 chi chi-square= 0.94 chi -square=0.82 -square=2.38 -square=0.06 -square=2.46 -square=5. 50** -square=0.1 6 -square=0.008 -square=0.86 -square=0.20 -square=0.51 -square=0.31 -square=0.06 -square=1 .0 -square=0.06 -square=0.27 -square=6.73*** *£<.05. **p_<.02. ***p_<. 01 ****£<. 01

PAGE 83

74 beach," "in the hot tub," "great atmosphere," and "places other than the bedroom." Table 1 1 lists the chi-square critical values for all categories for both gender and guilt. The direction of the findings that reach levels of significance at .05 are illustrated in Figures 1 and 2. Summary The data for the 12 structural null hypotheses were analyzed by using a 2 X 2 multivariate analysis of variance (MANOVA) procedure. The chi-square test for independent samples was used for the two categorical hypotheses. Main effects and interactions were examined for statistical significance at the .05 level. Overall, the results of the structural hypotheses multivariate analysis led to a failure to reject the contention that there were no differences in the personal construct paradigms of high and low sex-guilt male and female university students. Though not significant, there were notable gender differences in complexity of construct use (FIC), extremity ratings, and negative ratings of sexual experiences. Significant gender differences were found in constructs used to describe sexual experiences. There were minimal differences in the constructs used by high and low sex-guilt students participating in this study.

PAGE 84

CHAPTER FIVE SUMMARY, DISCUSSION, LIMITATIONS, AND IMPLICATIONS FOR FUTURE RESEARCH Summary The primary purpose of the present investigation was to examine the effects of gender and sex-guilt upon the cognitive responses that male and female university students generate to describe sexual satisfaction in their personal sexual experiences. Of equal importance was the exploration of the nature of constructs which individuals in this study would elect to represent concepts of sexual satisfaction. Hypotheses for these effects were formulated from a personal construct theory perspective. Eighty students were selected from a prescreeening of 297 students available from the introductory psychology subject pool during the fall semester 1987. It was required that participants were from the white ethnic group, aged 18 through 25, heterosexual, and have had, or be currently engaged in a sexual relationship. They were assigned to high or low sex-guilt groups based on scores derived from the Mosher Sex-Guilt scale. The 80 university students who participated in this study then completed a 10 X 10 Sexual Satisfaction Repertory Grid. 75

PAGE 85

76 No significant differences were found in the structural properties of sexual satisfaction constructs rated by male and female high and low sex-guilt university students. However, there were observable gender differences in this study with male students showing more complexity in their use of personal constructs of sexual satisfaction. Female university students tended to rate their sexual experiences more extremely and more negatively than did male university students. The content hypotheses were analyzed using the chi-square statistic. Significant gender differences were found in the types of constructs the male and female university students generated for sexual satisfaction. These results are discussed in this chapter. Also included in this chapter are limitations and implications of the study and recommendations for future research. Discussion Hypotheses 1 through 1 2 refer to the cognitive structures of sexual satisfaction. The cognitive structures of interest to this investigation were (a) differentiation (FIC), (b) integration, (c) extremity, and (d) negativity and positivity. It was indicated in hypotheses 1 through 3, respectively, that no differences in the complexity of cognitive differentiation of sexual satisfaction would

PAGE 86

77 emerge between the male students' ratings and the female students' ratings. In addition, there would be no significant differences due to level of sex-guilt, nor would differences occur in combination of these independent variables. The multivariate analysis of variance did not reach significance at the .05 level. However, the analysis of variance for univariates seemed evidential of differences in cognitive complexity of sexual satisfaction. It would appear from the results of testing that male students in this study were somewhat more differentiated or cognitively complex in sexual satisfaction constructs than were the female students. Due to the exploratory nature of this research, care must be taken to limit the conclusions drawn from the data. Indeed, there is a paucity of reported personal construct research relative to gender differences in cognitive structure, offering little previous research bearing directly on this issue. Recently, Neimeyer and Metzler (1987) reported significant differences between males and females along measures of cognitive structure. Similarly, males were found to be much more differentiated (£ < .003), but, their study measured vocational constructs. Females were reported to demonstrate higher levels of integration (p_ < .05). No conclusions were drawn from these observed gender differences but directions for future research were identified.

PAGE 87

78 Relatedly, males in this study had higher differentiation mean scores than did females. In fact, the high sex-guilt males were the most differentiated of the four groups. One possible explanation of this is Bieri's (1955) "vigilance" hypothesis. Bieri suggested that subjects differentiated more between negative, disliked persons, or unpleasant situations than they did between positive events or close associates. He posited a tendency for individuals to be more guarded when confronted with negative or unknown parameters, suggesting that cognitive differentiation serves an adaptive function to help the individual anticipate outcomes. Harrell (1970) offered an alternative interpretation of "vigilance", proposing that social demands create the expectation that one would like a particular person or situation. If that is not the case, she or he is behooved to have a good reason for this aberration and therefore must refine perceptions of disliked persons or situations. The "vigilance" hypothesis offers a plausible explanation for the higher differentiation scores reported for high sex-guilt male students. Considering the definition of sex-guilt as a generalized expectancy for self-mediated punishment for violating or for anticipating violating standards of proper sexual conduct (Mosher & Cross, 1971), it would follow that this group would be the most differentiated in attending to sexual stimuli. It would

PAGE 88

79 seem congruent that this group would be the most cautious and complex in attending to perceived sexual threats. These interpretations may be reasonable, but the results may be explained in other ways as well. Hypotheses 4 through 6 in the present study indicated that no significant differences would occur among integration scores due to gender, sex-guilt level, or an interaction of the two. Integration scores reflect the degree of statistical association between constructs suggesting that the dimensions are organized into an interrelated system of perceptions. Individuals with a well integrated system of constructs understand nuances of application of the independent constructs in his or her repertoire. Unlike the differentiation and integration research just cited (Neimeyer & Metzler, 1987), no differences were found in this study among the groups. This suggests that there was a comparable facility of application of sexual satisfaction constructs among the high and low sex-guilt male and female university students in this study regardless of the degree of complexity of those constructs. Hypotheses 7 through 9 were tested for gender and sex-guilt main effects and interactions for differences in the use of extreme ratings to describe sexually satisfying or dissatisfying experiences. No significant differences were found in extreme ratings used by male and female

PAGE 89

80 university students. Females in this study, however, tended to rate sexual experiences more extremely than did males. Higher extremity scores suggest more meaningfulness of the construct to the subject. There is considerable evidence that "elicited" constructs, the procedure performed in this study, produces more extreme ratings by subjects (Adams -Webber, 1979; Landfield, 1971) than do ratings across "provided" constructs. This suggests, according to personal construct literature, more personal meaningfulness to the ratings of ones personal constructs than constructs supplied from another source. This does not answer the question of whether females generally tend to use more extreme ratings than do males or if this is an isolated occurrence. Of additional interest would be gender comparisons in extreme ratings across varying domains. Possibly the extremity of ratings differences in this instance attest to stronger feelings of investment in the constructs associated with the females' sexual experiences. The final structural null hypotheses of this study, 10 through 12, were tested for differences in positive and negative ratings of the sexual satisfaction constructs used to describe sexual experiences. The hypotheses were that no differences would exist in the ratings used by male and female university students, high and low sex-guilt students, nor high and low sex-guilt male and

PAGE 90

81 female students. Critical values for hypotheses 10 through 12 did not reach .05 levels of significance, therefore, the researcher failed to reject these hypotheses. It is evident from the univariate analysis (p_ < .05) that there were some gender differences related to negative ratings. Females in this study tended to rate sexual experiences more negatively than did males. It would be too simplistic to ascribe this observance to any one aspect of the female sexual response. There are many psychosocial and physiological factors which could account for these rating differences. Kaplan (1974) discussed the increasing prevalence of complaints about lack of sexual interest and negative feelings about sexual behavior. This has become so common that it has been classified as inhibited sexual desire (ISD). Males also exhibit ISD but not as often as females. Although this may be a plausible explanation for the more frequent negative ratings used by females in this study, it is only one possible explanation. Surprisingly, sex-guilt in this investigation, did not have a significant influence upon any of the five dependent variables related to cognitive structure of sexual satisfaction. This is not in keeping with the review of the literature related to the impact of sex-guilt on sexual attitudes and behavior. This is discussed in more detail later in this chapter.

PAGE 91

82 Null hypotheses 1 3 and 1 4 concerned the content of the repertory grid. The hypotheses were tested for differences in the ways university students (by gender and level of sex-guilt) construed sexual experiences. Gender differences in the use of constructs to describe sexual satisfaction were found to be significant in seven categories (see Table 11). Male students in this study made significantly more references to status, factual descriptions, external appearance, and statements denoting self importance (egoism) that did female students. Females made significantly more self references, used multiple descriptions for an experience (alternatives) and referred to feelings of love, gentleness, and kindness denoting tenderness more often than did males. Guilt had an effect in two categories in this study — self reference and environment. In both categories, low sex-guilt individuals were more likely to describe sexual satisfaction in terms relating directly to him or herself and to make statements about surroundings than were high sex-guilt individuals. Gender differences were found to occur in the ways male and female students described sexually satisfying experiences. These differences occurred in directions which are in accord with the literature. Kaplan (1974), Masters, Johnson, & Kolodny (1986), Hite (1982), Carroll,

PAGE 92

83 Volk, and Hyde (1985) described male sexual response as much less sensitive to psychological influences and relatively impervious to the quality of the relationship, especially when they are young. On the other hand, females have been found to focus more on the psychological involvement as a major component of sexual pleasure. It is important to take note of the large frequencies in categories in which no significant differences were found in this study but were representative of responses which reflect gender agreement on constructs for sexual satisfaction. The types of responses having to do with the status category were statements such as, "embarrassed to be seen in public with him," "proud, boastful," "accomplishment," and "not proud to be out with her." Factual descriptive responses were observable or measurable such as "naked," "coarse skin," "clothed," "drunk," or "abstainer." External appearance constructs included "ugly," "physically dreadful," "attractive," and "they were the hottest girls." Egoism constructs were "I conquered these chicks," "convenience," "with a bore," "the girl flipped out over me," and "waste of my time." Self-reference had to do with direct self statements such as "I was naive," "I was happy," "I felt bad about myself," and "I lacked desire." Constructs that conveyed several different meanings were placed in the alternative category. Those

PAGE 93

84 included "unsatisfied, ashamed"; "ashamed, noncommitted" ; "good time, quick, fun"; and "happy, new, and exciting." Tenderness terms included "gentle," "caressing," "calm," "caring," and "sweet." It may be concluded that male and female university students in this study were more similar than different in their manner of construing sexual experiences. However, there were significant gender differences in the constructs generated by the university students to describe these experiences. Sex-guilt did not appear to have an effect on the responses of the university students participating in this study. Limitations of the Study One apparent limitation of this study was the size of the grid used. Although it was selected because of administrative constraints in consideration of the subjects* time and the sensitive nature of the investigation, a larger, more complex grid may have been a better choice. The resulting 10X10 grid was a relatively small representation of the possible field of sexual satisfaction constructs observed in this study. Another limitation was the homogeneity of the sample and possible restricted range of sexual experiences from which the data were drawn. The data also are applicable to a white college-aged population only. Whether the data

PAGE 94

85 are representative of other populations can be empirically determined in future research. However, generalizability of the results is restricted. A final limitation was the use of the revised Mosher Sex-Guilt scale. Although the scale was revised to accommodate a recurring change in the range of guilt scores in recent years (Mosher, 1987), the revised version may not have been in use a long enough time to provide substantial evidence of its utility. The absence of an effect due to the level of sex-guilt may reflect more of a problem of instrumentation than an accurate estimate of the role the disposition of sex-guilt plays in a person's construct system. Implications for Future Research The study presented here was exploratory in nature and represents a beginning step in understanding the psychological constructions of sexual satisfaction among university students. The results of this study provide information about the cognitive structures related to sexual satisfaction by operationalizing the term in the form of a repertory grid. The testing of the null hypotheses by using a multivariate analysis of variance failed to yield significant differences at the .05 level for effects of gender and guilt. Therefore, it seems questionable to

PAGE 95

86 interpret univariate tests, but there were interesting gender differences in differentiation, extremity of ratings, and negative ratings of sexual experiences reported in this study. Future researchers may wish to address these factors and examine other possible differences between male and female university students' understandings of sexually satisfying experiences. If males are more differentiated on this dimension, what might that say about self-disclosure? There is some evidence which corroborates the occurrence noted in this study of gender differences in extremity ratings (Neimeyer, Brown, Metzler, Hagans, & Tanguy, 1988). The notion of female negativity toward sexual experiences presents important implications for counselors and should be investigated further and in more detail. It is also important to extend and refine the rating system for content categories. This study revealed differences in the way male and female university students described sexually satisfying experiences. Similarities of content were also noted. Would these content categories and cognitive structures be similar with differing age groups, non-student groups, ethnic groups, or socioeconomic groups? There appear to be numerous research possibilities using refinements of the methodological approach initiated in this study. This

PAGE 96

87 would result, hopefully, in even more meaningful and worthwhile contributions to the understanding of human sexuality.

PAGE 97

APPENDIX A INFORMED CONSENT

PAGE 98

INFORMED CONSENT PART I This Inventory is the first part of a research project focused on sexual attitudes and experiences. If you are willing to participate in Part II, a pencil-andpaper questionnaire, write your name in the space provided on the answer sheet. Include your telephone number in the space labeled Social Security Number. Your name is requested solely for the purpose of contacting you to schedule the time for Part II participation. When all subjects have been contacted, the names and telephone numbers will be destroyed to provide confidentiality. If you have any questions or concerns regarding this topic, please contact me at 475-1050. Alma R. Suchman Principal Investigator I have read the above passage and understand that my participation in this research is voluntary. I may be included in Part II of this research if I include my name and telephone number. This is only for the purpose of scheduling the research time and all identification with 89

PAGE 99

90 responses will be destroyed when all subjects have been contacted. Signed Date

PAGE 100

91 INFORMED CONSENT PART II You have volunteered to participate in research which will involve paper and pencil responses to an instrument that focuses on sexual experiences. This will require approximately 75 minutes of your time. Your participation and identification with any response is strictly confidential. Should this subject cause embarrassment, you are free to discontinue the testing, refuse to respond to any question, and/or limit the scope of disclosure. Participation in this research is voluntary and no monetary compensation is being offered to you for participation. If you have any questions or concerns regarding this topic and your experience as a subject, please contact me at 475-1050. Alma R. Suchman Principal Investigator I have read the above passage and understand that I am free to withdraw my consent at any time. I also

PAGE 101

92 understand that my responses are confidential within the limits defined by law. Signed Date

PAGE 102

APPENDIX B REVISED MG INVENTORY

PAGE 103

REVISED MG INVENTORY Rate each of the items from to 6. Record your answer on the machine scoreable answer sheet by filling in the bubble opposite the item number. Please do not omit any items; O's must be filled in to be read by the computer. NOT AT ALL EXTREMELY TRUE OF TRUE OF (FOR) ME (FOR) ME 12 3 4 5 6 ************************************************************ 1 "Dirty" jokes in mixed company do not bother me. 2. "Dirty" jokes in mixed company are something that makes me very uncomfortable. 3. Masturbation is wrong and will ruin you. 4. Masturbation helps one feel eased and relaxed. 5. Sex relations before marriage should be permitted. 6. Sex relations before marriage are wrong and immoral. 7. Sex relations before marriage ruin many a happy couple. 8. Sex relations before marriage are good in my opinion. 9. Unusual sex practices might be interesting. 10. Unusual sex practices don't interest me. 11. When I have sexual dreams I sometimes wake up feeling excited. 12. When I have sexual dreams I try to forget them. 13. "Dirty" jokes in mixed company are in bad taste. 14. "Dirty" jokes in mixed company can be funny depending on the company. 15. Petting, I am sorry to say, is becoming an accepted practice. 16. Petting is an expression of affection which is satisfying. 17. Unusual sex practices are not so unusual. 18. Unusual sex practices don't interest me. 19. Sex is good and enjoyable. 20. Sex should be saved for wedlock and childbearing. 21 "Dirty" jokes in mixed company are coarse to say the least. 22. "Dirty" jokes in mixed company are lots of fun. 23. When I have sexual desires I enjoy it like all healthy human beings. 24. When I have sexual desires I fight them for I must have complete control of my body. 94

PAGE 104

95 NOT AT ALL EXTREMELY TRUE OF TRUE OF (FOR) ME (FOR) ME 1 2 3 4 5 6 ************************************************************ 25. Unusual sex practices are unwise and lead only to trouble. 26. Unusual sex practices are all in how you look at it. 27. Unusual sex practices are O.K. as long as they're heterosexual. 28. Unusual sex practices usually aren't pleasurable because you have preconceived feelings about their being wrong. 29. Sex relations before marriage, in my opinion, should not be practiced. 30. Sex relations before marriage are practiced too much to be wrong. 31. As a child, sex play is immature and ridiculous. 32. As a child, sex play was indulged in. 33. Unusual sex practices are dangerous to one's health and mental condition. 34. Unusual sex practices are the business of those who carry them out and no one else's. 35. When I have sexual desires I attempt to repress them. 36. When I have sexual desires they are quite strong. 37. Petting is not a good practice until after marriage. 38. Petting is justified with love. 39. Sex relations before marriage help people adjust. 40. Sex relations before marriage should not be recommended. 41 Masturbation is wrong and a sin. 42. Masturbation is a normal outlet for sexual desire. 43. Unusual sex practices are awful and unthinkable. 44. Unusual sex practices are all right if both partners agree. 45. Masturbation is all right. 46. Masturbation is a form of self-destruction. 47. If I had sex relations, I would feel all right, I think. 48. If I had sex relations, I would feel I was being used, not loved. 49. Masturbation is all right. 50. Masturbation should not be practiced. Please fill in the bubble opposite the item number with the number which applies to you. 51 AGE: Fill in if your age is _1_7 or under. Fill in 1 if your age is 1 8 Fill in 2 if your age is 19.

PAGE 105

96 if your age is 20 if your age is 21 if your age is 22 if your age is 23 if your age is 24 Fill in 8 if your age is 25 Fill in 9 if your age is 26^ or over. Fill

PAGE 106

APPENDIX C REVISED MG INVENTORY INSTRUCTIONS

PAGE 107

REVISED MG INVENTORY INSTRUCTIONS INSTRUCTIONS : This inventory consists of 50 sentences such as "When I have sexual dreams I try to forget them." You are to respond to each sentence as honestly as you can by rating your response on a 7-point scale from 0, which means NOT AT ALL TRUE OF (FOR) ME to 6, which means EXTREMELY TRUE OF (FOR) ME. Ratings of 1 to 5 represent ratings of agreement-disagreement that are intermediate between the extreme anchors of NOT AT ALL TRUE and EXTREMELY TRUE for you. The sentences are arranged in pairs of 2 to permit you to compare the intensity of TRUENESS for you. This limited comparison is often useful since people frequently agree with only 1 item in a pair. In some instances, it may be the case that both sentences or neither sentence is true for you, but you will usually be able to distinguish between sentences in a pair by using different ratings from the 7-point range for each item. Rate each of the 50 sentences from to 6 as you keep in mind the value of comparing items within pairs. Record your answer on the machine scoreable answer sheet by filling in the blank opposite the item number with your rating from to six. Please do not omit any items; zero's must be filled in to be read by the computer. 98

PAGE 108

APPENDIX D PROCEDURES FOR ADMINISTERING THE REP-TEST

PAGE 109

PROCEDURES FOR ADMINISTERING THE REP-TEST Confidentiality This study is being conducted to provide a better understanding of human sexual response. The results of the instruments you are about to complete will be used to help make counseling more effective. The information you provide will be completely confidential Care has been given to protect your privacy and anonymity Do you have any questions about confidentiality? Introduction Each of us has a unique way of understanding and interpreting our sexual experiences. The instrument you are beginning is a method of exploring your way of understanding some sexual events in your life. There are no right or wrong answers. 1 Identifying the experiences. To begin, here are 10 cards marked on front and back with the letters A through J. You will write on the lined side of the card. Only you will see these cards. They will NOT be collected. You may either take the cards with you when you leave or destroy them on completion of the repertory grid. You will be asked to recall 10 sexual experiences. The Sexual Experience sheet in your packet is to guide and assist you in this recall. 100

PAGE 110

101 Please think, back over your sexual history and on card A write a word or phrase which helps you to recall a sexually satisfying experience that included but was not necessarily limited to Handholding. The word or phrase may include a partner's name, the time of day, date, place or nature of the experience. Use whatever method helps to fix this event in your mind. Next, take the card marked, B (continue with other sexual experiences as the interviewer structures them for you. ) 2. Elicitation of constructs. The next part of this process involves thinking about these 10 experiences in groups of three. For this, you will use the Rating Scale form in your packet. To begin, take cards lettered A. D. and E (Triad #1). Think about these 3 experiences. Are 2 of them alike in some important way which distinguishes them from the third experience? When you decide which 2 experiences are alike put them together and write the word or phrase that describes how these 2 experiences are alike under the column designated "Alike" on the Rating Scale form. Now, write what you consider to be the opposite of this characteristic under the column designated "Opposite" on your Rating Scale. This opposite does not have to describe the third experience. Now take cards lettered B, F. and I (Triad #2). Think about these experiences until you remember an

PAGE 111

102 important way in which 2 of them are alike and which distinguishes them from the third experience. Write the word or the phrase that describes how 2 of the experiences are alike. If possible try to write a different description for each comparison. Now, write the opposite. Continue this comparing and contrasting with each or the ten sorts requested on the SORT page in your packet. When you have completed this, STOP. You will continue with instruction from the instructor. 3. Rating of each of the 10 sexual experiences. For the third part, you will use the form on which you have written descriptive words and phrases and the cards of 1 experiences. Rate each experience on the 10 scales by putting a circle around the number which best represents your description of the particular experience. For example, your first scale is / Rate the experience toward the 6 on the left if you see the experience as very Rate the experience toward the 6 on the right if you see the experience as very Rate the experience on the mid-point if you see the experience as in between the 2 characteristics, or if you are unsure which would apply to the experience, or if neither characteristic applies. STOP.

PAGE 112

103 4. Indication of preferred pole. The final step is to indicate next to each of the 2 descriptive phrases in each pair which is the side you prefer with a (+), and the least favorable side with a (-).

PAGE 113

APPENDIX E RATING SCALE

PAGE 114

3.2 D D D D D 3 E w < u in z H O "S. 3 B. cr < UJ C5 E > '11 D D D D D D o 105

PAGE 115

APPENDIX F SEXUAL EXPERIENCES

PAGE 116

SEXUAL EXPERIENCES Please think back over your sexual history and write a word or phrase which brings to mind different experiences which included but may not have been limited to the following behaviors. A. Recall a sexually satisfying experience that included but might not have been limited to Handholding B. Recall a sexually dissatisfying experience that included but might not have been limited to Hugging and Kissing C. Recall a sexually dissatisfying experience that included but might not have been limited to Disrobing D. Recall a sexually satisfying experience that included but might not have been limited to Hugging and Kissing E. Recall a sexually dissatisfying experience that included but might not have been limited to Nude Body Caresses F. Recall a sexually satisfying experience that included but might not have been limited to Sexual Intercourse G. Recall a sexually satisfying experience that included but might not have been limited to Disrobing 107

PAGE 117

108 H. Recall a sexually dissatisfying experience that included but might not have been limited to Sexual Intercourse I. Recall a sexually dissatisfying experience that included but might not have been limited to Handholding J. Recall a sexually satisfying experience that included but might not have been limited to Nude Body Caresses

PAGE 118

APPENDIX G SORTS

PAGE 119

SORTS The 10 cards will be sorted into 10 groups of three (Triads). The triads are listed below. Each is presented by the interviewer. Time is given between each sort for the subject to consider the special ways in which the experiences presented are alike. 1 A, D, E 2. B, F, I 3. E, G r J 4. D, F, H 5. A, B, C 6. D, H, J 7. C, G, I 8. B, H f F 9. A, C, G 10. E, F, J 110

PAGE 120

REFERENCES Abramson, P. R., & Handschumacher I. W. (1978). The Mosher sex guilt scale and the college population. Journal of Personality Assessment 42 635. Abrarason, P., Mosher, D. Abramson, L., & Woychowski, B. (1977). Personality correlated of the Mosher guilt scales. Journal of Personality Assessment 41 375-382. Adams-Webber, J. R. (1979). Personal construct theory New York: John Wiley & Sons. Adams-Webber, J., & Mancuso, J. C. (1983). Applications of personal construct theory New York: Academic Press. Bannister, D., & Mair, J. M. M. (1968). The evaluation of personal constructs New York: Academic Press. Barbach, L. (1984). For each other New York: New American Library. Bardwick, J. (1971). The psychology of women New York: Harper & Row. Barrett, J. (1972). Gerontological psychology Springfield, Illinois: Charles C. Thomas. Baucom, D. H. & Aiken, P. A. (1984). Sex role identity, marital satisfaction, and response to behavioral marital therapy. Journal of Consulting and Clinical Psychology 52 438-444. Bern, S. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology 42 155-162. Bentler, P. M. (1968a). Heterosexual behavior assessment-I. males. Behavior Research and Therapy 6_, 21-25. 111

PAGE 121

112 Bentler, P. M. (1968b). Heterosexual behavior assessment-II, females. Behavior Research and Therapy 6_, 27-30. Bernstein, D. (1982). The formulation of an instrument to assess the interpersonal meanings of sexual experience (Doctoral dissertation, University of Florida, 1982). Dissertation Abtracts International 43, 06B. Bieri, J. (1955). Cognitive complexity-simplicity and predictive behavior. Journal of Abnormal Social Psychology 51 263-268. Bieri, J., Atkins, A. L., Briar, S., Leaman, R. L., Miller, H., & Tripodi, T. (1966). Clinical and social judgement New York: Wiley. Bonarius, J. C. (1968). Personal constructs and extremity of ratings. Proceeding of the XVIth International Congress of Applied Psychology (pp. 595-599). Amsterdam: Swets and Zeitinger. Brady, J. P., & Levitt, E. E. (1965). The scalability of sexual experiences. Psychological Records 15, 275-279. Carroll, J. L., Volk, K. D., & Hyde, J. S. (1985). Differences between males and females in motives for engaging in sexual intercourse. Archives of Sexual Behavior 14, 131-139. Comfort, A. (1972). The joy of sex New York: Crown. Conley, J., & O'Rourke, T. (1973). Attitudes of college students toward selected issues in human sexuality. Journal of School Health 43 286-292. Denney, N. W. Field, J. K. & Quadagno, D. (1984). Sex differences in sexual needs and desires. Archives of Sexual Behavior 1 3 233-245. Derogatis, L. R. (1975). Derogatis sexual functioning inventory Baltimore: Clinical Psychometrics Research. Derogatis, L. R. (1978). Derogatis sexual functioning inventory (rev. ed.). Baltimore: Clinical Psychometrics Research.

PAGE 122

113 Epting, F. (1972). The stability of cognitive complexity in construing social issues. British Journal of Social and Clinical Psychology 11 122-125. Epting, F. (1984). Personal construct counseling and psychotherapy New York: John Wiley & Sons. Eysenck, H. J. (1971). Personality and sexual adjustment. British Journal of Psychiatry 12, 553-608. Eysenck, H. J. (1972). Personality and sexual behavior. Journal of Psychosomatic Research 16 141-152. Eysenck, H. J. (1976). Sex and personality Austin: University of Texas Press. Farley, F. H. & Davis, S. (1980). Personality and sexual satisfaction in marriage. Journal of Sex & Marital Therapy 6_, 56-62. Fjeld, S. P., & Landfield, A.E. (1961). Personal construct consistency. Psychological Reports 8, 127-129. ~ Frank, E., Anderson, C, & Rubinstein, D. (1979). Marital role strain and sexual satisfaction. Journal of Consulting and Clinical Psychology 47 1096-1103. Fransella, F., & Bannister, D. (1977). A manual for repertory grid technique New York: Academic Press. Freud, S. (1927). The ego and the id London: Hogarth Press. Gagnon, J. H. (1977). Human sexualities Glenview, Illinois: Scott, Foresman. Galbraith, G. G. Hahn, K. & Leiberman, H. (1968). Personality correlates of free-associative sex reponses to double-entendre words. Journal of Consulting and Clinical Psychology 32 193-197. Garrison, D. A. (1983). A sex-role orientation to the interpersonal meaning of sexual experience (Doctoral dissertation, University of Florida, 1983). Dissertation Abstracts International 45 04B.

PAGE 123

114 Gerrard, M. (1987). Sex, sex guilt, and contraceptive use revisited: The 1980's. Journal of Personality and Social Psychology 52 975-980. Grater, H. & Downing, N. (1981). The meaning of sexual experience Unpublished manuscript. University of Florida, Gainesville, Florida. Green, S. E., & Mosher, D. L. (1985). A causal model of sexual arousal to erotic fantasies. The Journal of Sex Research 21 1-23. Guttman, L. (1950). The basis for scalogram analysis. In S. A. Stouffer, L. Guttman, E. A. Suchman, P. F. Lazarsfeld, S. A. Start, & J. A. Clausen, (Eds.), Measurement and prediction (pp. 60-90). Princeton, New Jersey: Princeton University Press. Harmatz, M. & Novak, M. (1983). Human sexuality New York: Harper & Row. Harrell, E. E. (1970). Affective stimulus value extent of contact and cognitive complexity Unpublished master's thesis, University of Florida, Gainesville, Florida. Heath, D. H. (1978). Personality correlates of the marital sexual compatibility of professional men. Journal of Sex & Marital Therapy 4_, 67-82. Heiman, J. (1980). Female sexual response patterns. Archives of General Psychiatry 37 1311-1316. Hite, S. (1976). The Hite report New York: Macmillan. Hite, S. (1982). The Hite report on male sexuality New York: Ballantine Books. Hoon, E., & Hoon, P. (1978). Styles of sexual expression in women: Clinical implications of multivariate analyses. Archives of Sexual Behavior 7_, 105-116. Hudson, W. W. (1982). The clinical measurement package : A field manual Homewood, Illinois: The Dorsey Press. Isaacson, G. I., & Landfield, A. W. (1965). The meaningf ulness of personal and common constructs. Journal of Individual Psychology 21 160-166.

PAGE 124

115 Izard, C. E. (Ed.). (1979). Emotions in personality and psychopathology New York: Plenum Press. Janda, L., Magri, M. & Barnhart, S. (1977). Affective guilt states in women and the perceived guilt index. Journal of Personality Assessment 41 79-84. Janda, L. H., & O'Grady, K. (1976). Effects of guilt and response modality upon associative sexual responses. Journal of Research in Personality 10 457-462. Janda, L. H. & O'Grady, K. (1980). Development of sex anxiety inventory. Journal of Consulting and Clinical Psychology 48 169-175. Jankowicz, A. D. (1987). Whatever became of George Kelly? Applications and implications. American Psychologist 42 481-487. Kaplan, H. S. (1974). The new sex therapy Nwe York: Brunner-Mazel. ~ '~~ Kelley, K. (1985). Sex, sex guilt, and authoritarianism: Differences in responses to explicitly heterosexual and masturbatory slides. The Journal of Sex Research 21 68-85. Kelly, G. A. (1955). A psychology of personal constructs New York: W. W. Norton. Kelly, G. A. (1963). A theory of personality New York: W. W. Norton. Kinsey, A. (1953). Sexual behavior in the human female Philadelphia: W. B. Saunders. Koch, P. B. (1983). The relationship between sexrelated attitudes and beliefs and the sexual concerns experienced by college students (Doctoral dissertation, Pennsylvania State University, 1983). Dissertation Abstract International 44 06A. Kolodny, R. D. (1981). Evaluating sex therapy: Process and outcome at the Masters & Johnson institute. Journal of Sex Research 17 301-318. Krafft-Ebing, R. (1924). Psychopathia sexualis New York: Physicians and Surgeons Book Company.

PAGE 125

116 Lamb, C. (1968). Personality correlates of human enjoyment following motivational arousal. Journal of Personality and Social Psychology 9_, 237-241 Landfield, A. W. (1971). Personal construct systems in psychotherapy Chicago: Rand McNally & Company. Landfield, A. W. (Ed.). (1977). Interpretive man. Nebraska symposium on motivation 1976 Lincoln: Nebraska Press. Landfield, A. W. & Cannell, J. E. (1987). Ways of assessing functionally independent construction, meaningfulness, and construction in hierarchy Manuscript submitted for publication. Langston, R. D. (1973). Sex guilt and sex behavior in college students. Journal of Personality Assessment 37, 467-472. Lemon, N., & Warren, N. (1974). Salience, centrality, and self-relevance of traits in construing others. British Journal of Social and Clinical Psychology 1_3, 119-124. LoPiccolo, J., & Lobitz, W. C. (1972). The role of masturbation in the treatment of sexual dysfunction. Archives of Sexual Behavior 2_, 163-171. LoPiccolo, J., & LoPiccolo, L. (1978). Handbook of sex therapy New York: Plenum Press. LoPiccolo, J., & Steger, J. (1974). The sexual interaction inventory: A new instrument for assessment of sexual dysfunction. Archives of Sexual Behavior 3_, 585-595. Love, R. E., Sloan, L. R. & Schmidt, J. J. (1976). Viewing pornography and sex guilt: The priggish, the prudent, and the profligate. Journal of Consulting and Clinical Psychology 44 624-629. Lowen, A. (1969). Betrayal of the body New York: Macmillan. Masters, W. H. & Johnson, V. E. (1966). Human sexual response Boston: Little, Brown. Masters, W. H. & Johnson, V. E. (1970). Human sexual inadequacy. Boston: Little, Brown.

PAGE 126

117 Masters, W. H., Johnson, V. E., & Kolodny, R. C. (1986). Masters and Johnson on sex and human loving Boston: Little, Brown. May, R. (1969). Love and will New York: W. W. Norton. Mosher, D. L. (1961). The development and validation of a sentence completion measure of guilt (Doctoral dissertation, Ohio State University, 1961). Dissertation Abstracts International 22 2468. Mosher, D. L. (1965). The interaction of fear and guilt in inhibiting unacceptable behavior. Journal of Consulting Psychology 29 161-167. Mosher, D. L. (1966). The development and multitraitmultimethod matrix analysis of three measures of three aspects of guilt. Journal of Consulting Psychology 30 35-39. Mosher, D. L. (1968). Measurement of guilt in females by self -report inventories. Journal of Consulting Psychology 32 690-695. Mosher, D. L. (1973). Sex differences, sex experience, sex guilt, and explicitly sexual films. Journal of Social Issues 29 95-112. Mosher, D. L. (1979). The meaning and measurement of guilt. In C. E. Izard (Ed.), Emotions in personality and psychopathology (pp. 105-129). New York: Plenum Press. Mosher, D. L. (1987). Revised Mosher guilt inventory Manuscript submitted for publication. Mosher, D. L., & Cross, H. (1971). Sex guilt and premarital sexual experiences of college students. Journal of Consulting and Clinical Psychology 36 27-32. Mosher, D. L., & O'Grady, K. (1979). Sex guilt, trait anxiety and females' subjective sexual arousal to erotica. Motivation and Emotion 2_, 235-249. Mosher, D. L., & Vonderheide, S. G. (1985). Contributions of sex guilt and masturbation guilt to women's contraceptive attitudes and uses. The Journal of Sex Research 21, 24-39.

PAGE 127

118 Mosher, D. L. (in press). Revised Mosher guilt inventory. In W. L. Yarber, & C. M. Davis (Eds.), Sexuality related measures; A compendium. NT:NP. Murphy, M., & Neimeyer, R. A. (1987). Auto-Rep: Software reference manual [Computer program], Memphis: Memphis State University. Neimeyer, G. J., Brown, M. T., Metzler, A., Hagans, C, & Tanguy, M. (1988). The impact of sex, sex-role orientation, and construct type on vocational differentiation, integration, and conflict Unpublished manuscript. Neimeyer, G. J., & Metzler, A. E. (1987). Sex differences in vocational integration and differentiation. Journal of Vocational Behavior 30 167-174. Pedersen, F. A. (1958). Consistency data on the role construct repertory test Unpublished manuscript, Ohio State University, Columbus, Ohio. Perlman, S. D., & Abramson, P. (1982). Sexual satisfaction among married and cohabiting individuals. Journal of Consulting and Clinical Psychology 50 458-460. Piaget, J. (1948). The moral judgment of the child Glencoe, Illinois: Free Press. Pinney, E., Gerrard, M., & Denney, N. (1987). The Pinney sexual satisfaction inventory. The Journal of Sex Research 23 233-251 Podell, L., & Perkins, J. C. (1957). A Guttman scale for sexual experience: A methodological note. Journal of Abnormal Social Psychology 54 420-422. Rodriguez, G. J. (1981). A comparison of male heterosexual and homosexual fantasies and tentative norms for the meaning of sexual experience (Doctoral dissertation, University of Florida, 1981). Dissertation Abstracts International 43 01B. Ryle, A., & Breen, D. (1972). A comparison of adjusted and maladjusted couples using the double dyad grid. British Journal of Medical Psychology 45 375-382.

PAGE 128

119 Sarrel, L., & Sarrel, P. (1984). Sexual turning points New York: Macmillan Publishing Company. Schmidt, G. (1982). Sex and society in the eighties. Archives of Sexual Behavior 11 91-97. Segal, J. (1984). The sex lives of college students Wayne, Pennsylvania: Miles Satndish Press. Shaw, M. (1981). Recent advances in personal construct technology New York: Academic Press. Slater, P. (Ed.). (1976). Explorations of interpersonal space New York: John Wiley & Sons. Snyder, D. & Berg, P. (1983). Determinants of sexual dissatisfaction in sexually disstressed couples. Archives of Sexual Behavior 1_2, 237-246. Teens speak out on sex and birth control. (1987, February 23). Sexuality Today p. 3. Traupmann, J., Hatfield, E., & Wexler, P. (1983). Equity and sexual satisfaction in dating couples. British Journal of Social Psychology 22 33-40. Wanlass, R. L. Killmann, P. R., Bella, B. S., & Tarnowski, K. J. (1983). Effects of sex education on sexual guilt, anxiety, and attitudes: A comparison of instruction formats. Archives of Sexual Beahvior 12, 487-502. Waterman, C. K. & Chiauzzi, E. J. (1982). The role of orgasm in male and female sexual enjoyment. The Journal of Sex Research 1 8 146-159. Wijesinghe, 0. B. A., & Wood, R. R. (1976). A repertory grid study of interpersonal perception within a married couples psychotherapy group. British Journal of Medical Psychology 49 287-293. Wright, K. J. T. (1970). Exploring the uniqueness of common complaints. British Journal of Medical Psychology 43 221-232. Zaken-Greenberg, F., & Neimeyer, G. (1986). The impact of structural family therapy training on conceptual and executive therapy skills. Family Process 25 599-608.

PAGE 129

Zilbergeld, B. (1978). Male sexuality; A guide to sexual fulfillment Boston: Little, Brown. Zuckerman, M. (1973). Scales for sex experience for males and females. Journal of Consulting and Clinical Psychology 41 27-29. 120

PAGE 130

BIOGRAPHICAL SKETCH Alma Rodgers Suchman was born on January 7, 1939, in De Funiak Springs, Florida. She attended Florida State University in Tallahassee from 1956 through 1960, graduating with a Bachelor of Science degree in education. In September 1977, Alma began graduate studies at the University of Florida. She received her Master of Education and Specialist in Education degrees in May of 1981. Alma completed an internship and post-internship at the University of Florida's Psychological and Vocational Counseling Center. She expects to graduate with the Doctor of Philosophy degree in April 1988. On November 7, 1981, Alma married David Ira Suchman. David is a professor and counselor at the University of Florida Counseling Center. Alma is the parent of John Bailey, Cynthia Holley, and Amy Bailey. She is also the grandmother of twin granddaughters Jessica and Allyson Holley. 121

PAGE 131

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degre^of Doctor of Philosophy. L [aJjl M a vrkH Wittmer, Chair sor of Counselor Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degre^of Doctor of Philosophy. Profess Eppng ~Y or arx Psychalog ogy I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. A. (jin. Harry A. prater, Jr. Professor of Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Phyllis M. Meek Associate Professor of Counselor Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. /P JiielviY V— ^vC£j, ?.-vt .//} J^qdelyir L. Resnick Professor of Counselor Education

PAGE 132

This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. April 19! jx J a c&^j^w^ Dean, College of Education Dean, Graduate School

PAGE 133

UNIVERSITY OF FLORIDA MaB 08285 352 3"


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EO62IN04K_SXW3I2 INGEST_TIME 2017-07-11T22:06:26Z PACKAGE UF00101043_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES