UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characteristics of the individual and the transition : mediating factors in the response to infertility Koropatnick, Stephanie J. 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_spring_koropatnick_stephanie.pdf [ 6.02MB ]
JSON: 831-1.0054143.json
JSON-LD: 831-1.0054143-ld.json
RDF/XML (Pretty): 831-1.0054143-rdf.xml
RDF/JSON: 831-1.0054143-rdf.json
Turtle: 831-1.0054143-turtle.txt
N-Triples: 831-1.0054143-rdf-ntriples.txt
Original Record: 831-1.0054143-source.json
Full Text

Full Text

CHARACTERISTICS OF THE INDIVIDUAL AND THE TRANSITION: MEDIATING FACTORS IN THE RESPONSE TO INFERTILITY by STEPHANIE J. KOROPATNICK A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Counselling Psychology)  We accept this thesis as conforming to the required standards  THE UNIVERSITY OF BRITISH COLUMBIA December 1991  © Stephanie J. Koropatnick 1991  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of (',ounce  dint. l  The University of British Columbia Vancouver, Canada  Date )er 	/ ,  DE-6 (2/88)  ,,,Ch.o./08/1  Abstract  Using Schlossberg's (1981) model of adjustment to transitions in adulthood as a framework, an exploratory study was conducted of the factors mediating responses to involuntary primary infertility. Forty-three women and twenty-eight men who were seeking medical attention for infertility concerns at an Alberta Infertility clinic were recruited to complete five standardized selfreport instruments and one researcher-designed survey measuring fourteen predictor variables and ten outcome variables. Predictor variables included four Characteristics of the Transition, including: diagnosis, duration, perceived outcome and perceived chances of success, as well as ten Characteristics of the Individual, including: age, sex, race, religion, SES, sex-role identity, self-esteem, locus of control, general health and previous experience with a similar transition. Outcome variables included nine symptoms of distress, including somatization, obsessive compulsiveness, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation and psychoticism. Canonical correlational analyses indicated two significant links between characteristics of the individual and responses to infertility. The first link demonstrated that men and women with high self-esteem, high SES, higher age and an internal locus of control tended to respond to infertility with fewer symptoms of distress overall, scoring particularly low on measures of anger and hostility. The second link described a group of individuals who were among the oldest members of the sample and who reported low self-esteem and an undifferentiated sex-role identity. These men and women appeared to also report heightened feelings of distress overall, with particularly high levels of anxiety and obsessive compulsiveness. From the results, two groups who may be at risk of experiencing difficulties with infertility were identified. They include: Men and women who are recently infertile and have yet to receive a definitive diagnosis and older men and women who demonstrate low self-esteem and a poorly defined self-image. The counselling needs of each group were briefly discussed and recommendations made for possible directions of future research.  ii  TABLE OF CONTENTS Page  Contents Abstract Table of Contents List of Tables List of Figures Acknowledgements  vi vii viii  Chapter 1: Statement of the Problem Introduction The Model Characteristics of the Transition Characteristics of the Individual Purpose of the Study Assumptions, Limitations and Delimitations Assumptions Limitations Delimitations  1 1 4 6 9 14 15 15 15 17  Chapter 2: Review of the Literature Introduction Historical Approach to Psychology and Infertility Impact of Infertility Factors Mediating the Impact of Infertility Summary of Findings Hypotheses Research Hypotheses Null Hypothesis #1 Null Hypothesis #2 Alternative Hypotheses Alternative Hypothesis #1 Alternative Hypothesis #2  19 19 19 22 35 53 55 55 55 55 56 56 56  iii  Chapter 3: Methodology The Sample Characteristics Relating to Infertility Comparison of Respondents to Non-Respondents Instrumentation Personal Information Survey Bern Sex-Role Inventory Coopersmith Self-Esteem Inventory Rotter Personal Beliefs Questionnaire Derogatis SCL-90-R Adjustment to Infertility Scale Procedure Analysis of Data  57 57 58 60 61 61 62 64 65 66 68 69 71  Chapter 4: Results  72 72 74 75 76 76 76 78 78 83 84 84 86 89  Assumptions and Data Handling Descriptive Data: Independent Variables Descriptive Data: Dependent Variables Canonical Correlational Analyses Characteristics of the Transition Hypothesis One Characteristics of the Individual Hypothesis Two Additional Analyses Comparison of Canonical Variates by Gender Comparison of Canoncial Variates by Race Comparison of Canonical Variates by Sex-Role Summary Chapter 5: Discussion & Conclusions Discussion of the Findings Characteristics of the Transition Characteristics of the Individual Summary  iv  90 90 91 93 98  Chapter 5 continued  Implications for Counsellors  Recently Infertile  Older/ Negative Self-Image  Implications for Researchers  Review of Methods and Materials  Directions for Future Research  Conclusion  99 99 101 104 104 107 109  References Appendix A: Appendix B: Appendix C: Appendix D: Appendix E: Appendix F:  110 116 119 120 121 122 123  Personal Information Survey Research Referral Form Letter of Explanation Consent Form Follow-up Letter Schlossberg's Model of Adaptation to Transition  v  List of Tables  TABLES Page  Table 1: Demographic Characteristics: Responders v. Non-Responders  59  Table 2: Characteristics of the Transition: Responders v. Non-Responders  60  Table 3: Descriptive Data: Independent Variables  74  Table 4: Descriptive Data: Dependent Variables  75  Table 5: Loadings for First Canonical Variable: Characteristics of the Transition  78  Table 6: Results of Canonical Correlations  79  Table 7: Loadings for First & Second Canonical Variables: Characteristics of the Individual  81  Table 8: Loading of First & Second Canonical Variates: Dependent Variables  82  Table 9: Comparison of Males and Females on First & Second Canonical Variate Pairs  84  Table 10: Canonical Correlations: Race Excluded  85  Table 11: Loadings for First & Second Canonical Variates: Race Excluded  85  Table 12: Comparison of Group Means by Race  86  Table 13: Canonical Correlations: Sex-Role Excluded  87  Table 14: Loadings for First & Second Canoncial Variates: Sex-Role Excluded  87  Table 15: Comparison of Groups Means by Sex-Role  88  vi  List of Figures Page  Figure  Figure 1: Linearity of Correlation Between First Canonical Variates: Characteristics  of the Transition  77  Figure 2: Linearity of Correlation Between First Canonical Variates: Characteristics  of the Individual  79  Figure 3: Linearity of Correlation Between Second Canonical Variates: Characteristics  80 of the Individual  vii  Acknowledgements  In this space I would like to acknowledge and thank the many people whose support and assistance have been invaluable to my efforts in completing this research project. Despite the seemingly endless lonely hours spent in front of my computer, this thesis has been far from a solitary effort. I wish to thank the following people for their help: Dr. Tony Pattinson and the staff at the Infertility Clinic in the Dept. of Obstetrics and Gynaecology at Foothills Hospital in Calgary, for providing me with the space and the subjects needed to collect my data, and for making me feel welcome and comfortable throughout the process; Dr. Walter Bo1dt, for his kind and patient manner as much as for his surefooted guidance through the hazardous world of multivariate statistics; the staff at Education Computing Services, for guiding me through the equally hazardous minefield of mainframe computer software; and most importantly, Dr. Judith Daniluk, for the literally countless hours of reading, editing, making suggestions, re-reading and re editing, always with a sure hand, a supportive smile and a sense of humour. I would like also to thank my family and friends, who have been remarkably patient and understanding throughout the past three years. I want especially express my heartfelt appreciation to my husband, Bill, for his support and encouragement. Finally, I would like to dedicate this thesis to the memory of my father, Fred MacDonald, whose unfailing belief in me continues to be a source of inner strength and motivation for me in my work.  viii  Statement of the Problem  CHAPTER ONE Statement of the Problem  Introduction  For an increasing number of North American couples, infertility is an unwelcome disruption of life goals and plans. It is estimated that between 10 and 17% of couples who plan children are unable to conceive after at least one year, and often several years of trying (Davis, 1986; Kraft, Palombo, Mitchell, Dean, Meyers & Wright Schmidt, 1980; Morse & Dennerstein, 1985; Snarey, Son, Kuehne, Hauser & Valliant, 1987). For these couples, the inability to participate in the natural transition to parenthood is profoundly distressing, and may be seen as a crisis in adult development. Using Schlossberg's (1981) model of transitions in adulthood, the present study views infertility, defined as the inability to conceive a viable pregnancy after one year of regular, unprotected, sexual intercourse, as a critical non-event transition. The focus of this study is upon identifying factors which may be of relevance in understanding an individual's responses to infertility. Within our culture, it is generally assumed that a marriage will produce children; "The agreement to have children together is an assumption underlying most marriages" (Woollen, 1985, p.473). In marriage ceremonies the expectation that the couple will have children is nearly always implicit and is often explicitly prescribed as a requisite of a 'proper' or fulfilled marriage. Family, friends and even strangers often express their eagerness for the newlywed couple to 'sanctify' or complete the marital union with a child, with frequent questions and encouragement about the timing of childbirth (Sandelowski & Jones, 1986). Most couples, moreover, tend to be in agreement with this cultural imperative and plan to have at least one child. In fact, only 5% of couples choose voluntarily to remain childless (Veevers, 1980). Of the 95% majority of married couples who plan for and expect to become parents, it is estimated that at any given time, between one-in-five and one-in-twelve couples (Matthews & Matthews, 1986; Morse & Dennerstein, 1985) are experiencing difficulty conceiving and 1  Statement of the Problem  maintaining a pregnancy. These estimates refer predominantly to married couples who are currently seeking medical assistance for fertility issues and include couples with primary infertility, who have never carried a pregnancy to full term as well as those with secondary infertility, who have previously given birth to one or more children. They do not include married couples or single individuals who have yet to discover their infertile status and may in fact currently be using birth control. Nor do these estimates include couples who have previously had problems with fertility and are no longer seeking medical assistance for it (Halpern, 1989; Matthews & Matthews). It is possible, therefore, that as many as 15 to 20% of adult North Americans have problems with fertility in terms of conceiving and carrying a viable pregnancy (Morse & Dennerstein). Furthermore, it is suggested that these numbers may be on the increase (Bryant, 1990). Becoming a parent is generally assumed to be one of the fundamental transitions in adult life. Throughout childhood and adolescence, the female socialization process is strongly geared toward potential motherhood (Maccoby & Jacklin, 1974; Russo, 1979). From the onset of the menstrual cycle, girls become aware of their body's readiness for pregnancy, and the fear of becoming pregnant prematurely is promoted as the primary prohibition to teenage sexuality (Hendricks, 1985). Many religious and sociocultural influences also effectively prescribe motherhood as the primary purpose of women (Deutsch, 1947; Russo, 1979). Girls and women subsequently come to assume fertility and eventual parenthood as inevitable. Boys and men are also socialized to anticipate fatherhood as a valued role, although in less direct ways. While there are no childhood rituals specifically concerned with the role of fatherhood, in many families a great deal of emphasis is placed upon the young man's responsibility, even obligation, to 'carry on the family name'. Erikson (1963) identified generativity, or the ability to pass on knowledge (and implicitly genetic material) as one of the fundamental and essential requisites for normal adult development. Further, a man's fertility is strongly linked with his virility and masculinity (Pohlman, 1970). In many cultures, the more children produced in a marriage, particularly male children, the greater respect accorded to the man. In view of the high degree of importance placed upon child-bearing and rearing in North 2  Statement of the Problem  American society and the inevitability with which these events are anticipated by both men and women, the discovery of infertility is frequently experienced as a "developmental crisis" (Kraft, et al., 1980, p.620). Indeed, the pattern of responses observed in infertile subjects over a variety of studies is similar in many ways to a 'crisis reaction', and numerous writers have adopted a crisis model in examining the experience (Lalos, Lalos, Jacobsson & Von Schoultz, 1986; Menning, 1980; Valentine, 1986). These researchers and clinicians have observed behavioral indicators of shock, denial, anger, depression and loss of control in individuals and couples experiencing infertility. Unlike a crisis, which is typically a sudden event or series of events, with a definite beginning and end and a relatively brief duration, infertility may be an active issue for a couple for anywhere from eighteen months to several years. While viable pregnancies are achieved for between 50 and 60% of couples who seek medical assistance for infertility (Daniluk, 1988), for other infertile couples it may be difficult to identify a point at which the crisis is over. Couples who choose adoption or a child-free lifestyle may need to arbitrarily close the issue of infertility and make a choice to move onward. Other couples may never make that choice and may continue pursuing pregnancy long after reasonable hope is exhausted. The term 'crisis' usually connotes some sort of a dramatic event, which occurs suddenly and with no forewarning (Hopson, 1981). While infertility may in fact have a dramatic impact on the lives of those affected, it is difficult to construe it as an 'event'. Rather, infertility can be more thoroughly understood when seen as a series of 'non-events', as each menstrual cycle passes, failing to culminate in the much-anticipated and hoped-for pregnancy. Except in rare cases, such as when an emergency hysterectomy must be performed, the recognition that a problem exists is likely to be less a dramatic revelation, than a gradual dawning. Despite the high incidence of infertility and the importance of parenthood in the lives of men and women, infertiltiy continues to be viewed as a rare condition, both in popular thought and within the academic world. A review of more nearly two dozen texts concerning adult development revealed only a handful which even make a cursory mention of childlessness as an 3  Statement of the Problem  alternative and none which refer to infertility. Individuals and couples enduring the experience tend to feel marginalized and victimized by their condition. While medical advances have identified organic correlates in as many as 90% of the cases of infertility, a large proportion of the academic research produced in the field to date has continued to focus on identifying the psychological etiology of infertility. Although numerous psychological symptoms have been found to correlate with infertility, no progress has been made in discovering which factor or factors may play a causative role (Pantesco, 1986). The present study adopts an alternative view that infertility plays a causative role in a number of psychological symptoms. Infertility has been reported to have a negative intrapersonal and interpersonal impact. The factors which mediate this impact are of particular interest in this study. Schlossberg (1981) has developed a model of adult transition which may provide a more comprehensive framework for understanding of the impact of infertility on the life of an individual adult. The model is based within a theoretical framework which views adult development as an ongoing process of adapting, or not adapting, to psychosocial transitions. Transitions may be the result of normative events such as first job, marriage or retirement, or non-normative events like sudden wealth or job loss, or may be triggered by crisis events such as family deaths or natural disasters. Schlossberg (1981) has identified a range of factors which may interact to influence or shape the short-term and long-term effects of the transition. These include characteristics of the event, characteristics of the environment and characteristics of the individual. The present study uses Schlossberg's model of psychosocial transitions as a framework for attempting to understand the factors that serve to mediate the impact of infertility on men and women currently living through the experience. The Model  Schlossberg (1981) has defined psychosocial transition as "an event or non-event (which) results in a change in assumptions about oneself and the world and thus requires a corresponding change in one's behavior and relationships (and which) may provide both an opportunity for 4  Statement of the Problem  psychological growth and a danger of psychological deterioration" (p. 5-6). The inclusion of 'nonevents' or the "non-occurrence of anticipated events" (p.5) in this definition, makes Schlossberg's model particularly useful for understanding the impact of infertility. In this context, infertility is the non occurrence of the anticipated event of parenthood. -  Schlossberg (1981) suggests that adaptation to a psychosocial transition may be positive or negative, with the potential for growth or deterioration. The flexibility of this definition affords an opportunity to utilize the model in examining the complexity of responses reported by couples experiencing infertility. Through popular usage, 'adaptation' implies a positive set of responses which function to integrate a change or development into one's life. Hopson (1981) suggests that this popular connotation of adaptation interferes with its usefulness in describing transition outcomes. He suggests that more clarity in the understanding of transitions outcomes may by achieved by replacing 'adaptation' with 'responses to transition' (p.37). This more general descriptor accomodates the diversity of possible responses to any change, which range on a continuum from 'non-adaptive', through 'mere survival' and 'managing', to 'growth' (Hopson; Schlossberg). In the present study, 'responses to transition' will be used to describe the adaptive, non-adaptive and neutral ways individuals and couples behave while experiencing infertility and will be assessed in two ways. Responses to the stress of infertility may be observed in a variety of emotional and behavioral symptoms. Responses commonly reported by infertile individuals include depression, anger, confusion, fear, hypersensitivity, disorganization, distractability, moodiness, and obsessive behaviors and thoughts (Menning, 1980; Valentine, 1986). The present study is interested in the degree to which infertile men and women exhibit symptoms of distress, as well as in whether there are differences in the type of symptoms reported which may be attributable to characteristics of the transition and/or characteristics of the individual. Symptoms of distress will therefore be measured, using a self-report symptoms checklist. While symptoms of distress may be gauged as a measure of the emotional impact of infertility, to form a more complete picture of the range of responses to infertility it may also be 5  Statement of the Problem  necessary to assess the degree to which infertile men and women are adapting or adjusting to the transition. As a lack of reported symptoms may not necessarily be indicative of positive adaptation, a separate measure is used to assess the individual's level of adjustment to infertility. Schlossberg (1981) has identified three major groups of factors or variables which may be studied to understand the impact of a particular event or non-event on an individual. The three categories of factors identified include the following: the characteristics of the transition, which include factors unique to the specific event or non-event; characteristics of the environment, involving elements of the environoment prior to, during and following the transition; and characteristics of the individual, including psychosocial factors unique to the individual experiencing the transition (Schlossberg). According to the literature, support from family and from friends tends either to be nonexistent or actually experienced as negative (Davis, 1987; La_los et al., 1986; Menning, 1980; Sandelowski & Jones, 1986). For infertile individuals, all three levels of interpersonal support identified by Schlossberg (1981); intimate relationships, the family unit and the network of friends, may be problematic. The extent of the difficulties associated with obtaining social support is of sufficient complexity to warrant an independent investigation, which is beyond the scope of the present study. Based on the available research, the categories of particular interest in the present study include the characteristics of the infertility transition and of the infertile individual.  Characteristics of the T sition In recognition of the subjectivity of experiences, Schlossberg (1981) has identified seven characteristics of the transition, in terms of the way an individual perceives or defines their experience, which could affect the way he or she responds, including whether: it represents a gain or a loss, it is perceived as positive or negative, the perceived origin is an internal or an external source, it is on-time or off-time, the onset is perceived to be gradual or sudden, it is seen as temporary, permanent or of uncertain duration, and the degree of stress experienced. Among the most salient findings in recent literature on the perceptions of those experiencing infertility is the issue of loss (Bowers, 1985; Sawatsky, 1981; Valentine, 1986; 6  Statement of the Problem  Woollett, 1985). There are many losses represented by infertility, including "loss of potential children; loss of genetic continuity; loss of pregnancy, childbearing and breast-feeding experience; loss of a life goal; and loss of control over one's body" (Valentine, p.64.). It may be assumed therefore, that for those individuals seeking to 'cure' or ameliorate their infertile status, infertility invariably represents a loss. Given the universality of this experience, it is assumed for the purposes of the study that infertility represents a loss. The experience of infertility is almost universally perceived with negative affect. Despite a current trend toward childlessness as an active choice, this option is pursued by as few as 5% of the married population (Veevers, 1980). For the 95% majority of the married population who wish to have children, the inability to fulfill their expectations of biological parenthood is indeed regarded as negative. Thus the perception of infertility as negative is also assumed. Schlossberg (1981) suggests that transitions with a gradual onset are likely to be more easily adapted to than those which occur suddenly, because they may be anticipated and prepared for. The onset of the infertility transition may be considered gradual, but it is seldom anticipated or easy to prepare for. Although for a small minority of couples the diagnosis is swift, conclusive and irreversible, as in a case of emergency hysterectomy (Menning, 1977), the majority of infertile couples come to a full realization of their situation only gradually, as month after month passes with no pregnancy. Even after the decision is made to seek medical intervention, which generally can be received only after a year of unsuccessful attempts to conceive, it may be many more weeks, months or even years before a conclusive diagnosis is made. Therefore, for the purposes of the study, a gradual onset is presupposed. Schlossberg (1981) suggests that individuals "adapt...more easily to transitions in which the source is internal" ( p.9). While it may appear to an observer that the source of infertility is internal, within the very body of the individual, infertile people often perceive their condition as being imposed upon them. The removal of the ability to make the choice to have biological children means that, for infertile couples and individuals, the source of the infertility is often perceived to be external. Thus, for the purposes of the study, it is assumed that the source of the 7  Statement of the Problem  transition is external. The perception that infertility is imposed from an external source may be exaggerated if the cause of infertility is identified as a malfunction in the reproductive system of the marital partner. For this reason, the diagnosis received may be an important factor in understanding an individual's response to infertility. Conversely, it may also be expected that the member of the couple in whom the etiology of the infertility has been diagnosed may experience greater distress than an individual who is not diagnosed as infertile, but who is married to an infertile person (Daniluk, 1988). These factors will be measured, as individuals will be asked to indicate whether the diagnosis implies a female factor, a male factor, a couple factor or an unknown factor as the cause of infertility in the couple. Most couples perceive the duration of infertility as very uncertain, as a seemingly endless series of tests and treatments continues the cycle of hope and disappointment. Even after a clear diagnosis is made, the efficacy of many of the treatment procedures remains uncertain (Halpern, 1989; Leader, Taylor & Daniluk, 1984). Schlossberg (1981) notes that individuals who perceive the disruptive phase of a transition to be neverending or of uncertain duration are likely to experience higher levels of distress than are individuals who believe the changes to be permanent or temporary. The present study investigates this by including the individual's perception of the infertility as temporary, permanent or uncertain as among the predictors of distress levels. In addition, the actual duration of the expereince of infertility may be an important factor in predicting distress levels. Duration is measured by the length of time couples have been attempting to conceive as well as the length of time since seeking medical attention. Timing is also an issue of particular significance to the infertility experience. Whether a transition is on-time or off-time may be seen within the context of both chronological and psychological development. For those who make a transition 'off-time', either earlier or later than the norm, negative social and psychological consequences may be encountered as the individual may be made to feel 'deviant'. Schlossberg (1981) notes that events such as getting married (and) having children...are linked in people's minds with age" ( p.9.). With the increasing trend for 8  Statement of the Problem  couples to postpone childbearing until careers and financial comforts are established, many choose to begin a family when they are both psychologically and chronologically 'behind schedule'. Infertility may not, in fact, be discovered until the woman is nearing the end of what is generally assumed to be the 'fertile' years. The pressure to succeed in finding a solution to the infertility is often exacerbated by the inexorable ticking of the 'biological clock'. Thus the age of the individual, particularly of the woman, may be an important factor to observe. As age is a characteristic of the individual, this information is collected and analyzed with the second group of factors. For infertile couples, hope becomes a commodity of great importance. As long as the infertility 'work-up' continues, couples often find themselves caught in a 'cycle of hope and disappointment', measuring time by the woman's menstrual cycle (Davis, 1987; Kraft et al., 1980). For couples for whom a conclusive diagnosis cannot be made, the 10% to 20% whose infertility is of unknown cause, this cycle can be particularly tortuous. As Halpern (1989) notes, "infertile couples are vulnerable people, susceptible to hope however tentatively it is held out before them. Where others, learning that a particular treatment has a one-in-10 rate of success, understand that it fails 90 percent of the tithe, the infertile tend to focus exclusively on the fact that it works 10 percent of the time" (p.148). Thus the degree of hope the individual maintains, as estimated by his or her perception of the chance of a successful outcome, is included as a factor in predicting the amount of distress felt and the level of adjustment to infertility.  Characteristics of the Individual The other group of predictive factors identified in Schlossberg's (1981) model which are included for investigation in the present study are the characteristics of the individual. The factors which may be useful in predicting or understanding an individual's responses to infertility include: psychosocial competence; gender and sex-role identification; age; physical health; race, ethnic origin or cultural identification; socio-economic status; values, and; previous experience with transitions of a similar nature. The construct of 'psychosocial competence' includes a positive self-attitude, a realistic 9  Statement of the Problem  world attitude and an active behavioral orientation (Schlossberg, 1981; Tyler, 1978). Positive selfattitudes may be described in terms of a sense of self-esteem and an internal locus of control or sense of self-responsibility. Individuals with positive and flexible self-concepts may be likely to respond more adaptively, and with less distress to any transition, than those who have rigid or negative self-images (Schlossberg). One component of psychosocial competence is self-esteem. This has particular significance for the experience of infertility, as the inability to fulfill the role of parent may constitute a serious threat to an individual's self-image. Infertility may also have a profound effect on the individual's 'body-image', as many begin to see their bodies as defective and unattractive (Davis, 1987; Kraft et al., 1980). These negative feelings may spread to other aspects of self-concept, particularly sexual identity, with the physical malfunction coming to represent, in the individual's mind, all that is wrong with them as a person. As noted by Adler & Boxley (1985) "men and women who report low levels of...self-esteem may be at a higher risk for problems associated with infertility" (p.278). Conversely, high self-esteem may help to moderate or 'buffer' the impact of infertility. The buffering effects of self-esteem have been cited by numerous authors (Adler & Boxley, 1985; Chiappone, 1984; Coopersmith, 1981; Sabatelli et al.1988). What is most salient, however, is that individuals with high self-esteem and individuals with low self-esteem appear to respond to infertility in different ways. It is the nature of these differences that the present study attempts to explore as self-esteem is included as a predictor of distress levels in response to infertility. Another element of psychosocial competence may be described as an internal locus of control. Like self-esteem, the nature of the relationship between locus of control and infertility is unclear. Individuals who perceive themselves to be in control of the events that affect them tend to respond favorably to most life transitions (Schlossberg, 1981). However, when faced with the overwhelming experience of 'loss of control' reported by infertile men and women (McCormick, 1980), an internal locus of control and an active approach to life may be of little use. As such, it seems important to include locus of control in an examination of the factors that may mediate the impact of infertility. 10  Statement of the Problem  A second characteristic cited by Schlossberg (1981), which may be important in understanding responses to transtions, is the gender of the individual. This is of particular interest in the study of the infertility transition. It may be expected that males and females will respond differently to the experience of infertility. Infertile women have reported to greater feelings of loss and depression than men (Chiappone, 1984; Daniluk, 1988) as well as a tendency to assume the 'blame' and feel guilt for the couple's infertility (Chiappone; Daniluk; Lalos et al., 1986; McEwan, Costello & Taylor, 1987). This may be largely a reflection of social beliefs which tend to see infertility as primarily a 'female problem' and hold the woman responsible for its solution (Menning, 1977; Pantesco, 1986). Women also tend to experience distress in response to stressors affecting the significant persons in their lives (Schlossberg, 1981), and may therefore 'take on' the pain felt by an infertile husband. Alternatively, women have been shown to employ a greater range and variety of coping mechanisms in response to stress (Chiappone, 1984), suggesting an advantage over men in their ability to reduce feelings of distress. Further, although men appear to respond with less intensity than women, it is clear that males do experience the negative impact of infertility. For men, infertility often engenders feelings of humiliation and embarrassment, (Kraft et al., 1980; Valentine, 1986) depression (Feuer, 1983; Kedem, Mikulincer & Nathanson, 1990; Mahlstedt, 1985; Mazor, 1979; Walker, 1978) and anxiety (Kedem et al.). Many men associate the inability to impregnate a woman with sexual inadequacy and a failure to fulfil the male role (Kraft et al., 1980; Snarey et al., 1987). Clearly, the differences between men and women in response to infertiilty do not appear to be a simple matter of more and less, but rather, it is expected that males and females may exhibit differences both in the level and the nature of their responses to infertility. These differences may be apparent in the pattern of symptom and adjustment scores achieved. Another factor which may be as important as the individual's gender is the degree to which he or she identifies with traditional definitions of maleness and femaleness, or 'sex-roles'. The individual's sex-role identity may be an important factor in predicting his or her response to infertility. The ability to reproduce and raise a family is intimately and inextricably connected to 11  Statement of the Problem  both psychological and social definitions of what it means to be a man or a woman. Many women assimilate the pervasive sociocultural and interpersonal influences and come to view motherhood as central to their self-identity, as an integral "part of their biological and psychological makeup" (Frias & Wilson, 1985, p.43). Women who have strongly identified with traditional views of womanhood, and who define their value as women in "uterine terms, and (who) cannot choose to complete their image through achieving one or more pregnancies...feel stripped of their self-worth" (Hendricks, 1985, p.152). Thus a woman's identification with traditionally defined femininity may be related to the level and type of distress she experiences in response to infertility. Similarly, men who identify with traditional views of manhood, and are unable to fulfill their role by producing a family to provide for and protect, may feel a threat to their identity as males (Humphrey, 1977; MacNab, 1985; Mahlstedt, 1985). However, high masculinity has been related to higher levels of self-esteem (Adler & Boxley, 1985). Thus the degree to which a man identifies with traditional sex-roles may also be important in predicting his responses to infertility. As noted previously, the individual's age may may in an important factor in understanding his or her responses to infertility. Older individuals may react differently to the experience of infertility than do younger persons. Age may be related to increased distress as, the closer the infertile couple comes to the end of the women's reproductive years, the shorter is the time remaining to find a successful resolution to infertility. At the same time, however, age may be related to more effective coping strategies as older individuals may be expected to respond with greater maturity. Older individuals may also be more likely to have previously encountered difficult life experiences which may help to prepare them for facing infertility. Another factor, often related to age, as many couples delay attempts to reproduce until they feel financially secure, is socio-economic status. Individuals with higher levels of education may be more likely to seek and obtain information regarding infertility, thereby increasing their feelings of being in control of an uncontrollable situation. Thus it is expected that both age and socioeconomic status, including income and education levels, may be related to the levels of distress and 12  Statement of the Problem  adjustment reported by infertile individuals and will be measured. Race or ethno-cultural background may also be an important element in understanding the impact of infertility. While virtually all racial and cultural groups value having children, some groups place a stronger emphasis on reproduction as the central theme of marriage (Pohlman, 1970). Therefore, information on race will be collected, as it may be expected that the cultural or ethnic background of the individual may be a factor in mediating the impact of infertility, with members of some ethnic groups experiencing better adjustment and lower distress than members of other groups. Similarly, religious affiliation, and the degree to which an individual identifies with the teachings and values of their chosen religion, may also represent a significant cause of strain for the individual experiencing infertility. There may be pressure from the religious community to reproduce, to 'sanctify' the marriage, while simultaneously proscribing the use of medical interventions, many of which are seen as morally wrong (McEwan et al., 1987; Menning, 1982). McEwan et al. found that women who followed non-Protestant religions reported greater distress than Protestant women. Alternatively, Graham, Thompson, Estrada and Yonekura (1987) reported that highly religious women who blamed God for a miscarriage reported lower distress than less religious women who miscarried. Thus, an individual's religion and religiosity may play a role in either exacerbating or reducing the distress engendered by the experience of infertility and will be included for measurement. In addition, the general health of the individual, both prior to the discovery of infertility and during the time in which it is an active issue, may influence the individual's response to infertility. An individual who has been unhealthy for some time, or who is experiencing other health-related crises concurrently with infertility, may fmd it difficult to rally the resources necessary to effectively cope with infertility. Conversely, an individual who has enjoyed robust health prior to the onset of infertility may be more able to remain optimistic of their chances to achieve a pregnancy. Thus, the question of whether the individual's general health is poor, fair, good or excellent may be useful in predicting and understanding his or her responses to infertility. 13  Statement of the Problem  As suggested above, previous exposure to similar experiences may also influence an individual's ability to respond adaptively to infertility. Schlossberg (1981) suggests that individuals who have undergone similar transitions in the past will likely be better equipped to deal effectively with the present experience, especially if they were able to derive some positive outcome from the previous experience. An individual who, for example, had previously experienced a family death, may have learned important coping strategies for dealing with loss, as well as having gained insight and greater closeness with other family members. Another individual undergoing the same type of experience, however, may have coped unsuccessfully with their loss and subsequently suffered continued anguish. Therefore, knowing whether an individual has previously experienced transitions of a similar nature may be important in predicting his or her success in responding to infertility and is included in the investigation. Purpose of the Study  Each of the characteristics of the individual and the characteristics of the transition discussed above may play a role in influencing the individual's pattern of responding to the experience of infertility. Eleven characteristics of the individual, including sex, age, socioeconomic status, sex-role orientation, self-esteem level, locus of control, ethnic background (race), religion, general health and previous similar experience, as well as four characteristics of the transition, including length of time attempting to conceive, diagnosis received, perceived chance of success and perceived outcome have been used as predictor variables. Criterion variables include nine elements or symptoms of distress and a measure of adjustment to infertility. It is not expected that a single factor will emerge as a dominant predictor of response patterns. Rather, the purpose of the present study is to identify a combination or group of factors which covary in such a way as to create a picture or profile of individuals who are at greater risk of experiencing high levels of distress in response to the experience of infertility, as well as a profile of those individuals who are at lower risk of distress. Furthermore, the need for counselling professionals who are cognizant of the issues faced by infertile couples and individuals is becoming increasing apparent (Daniluk, Leader and Taylor, 14  Statement of the Problem  1987; Seibel & Taymor, 1982). A second purpose of the present study is to increase the level of recognition and understanding within the counselling profession, of both the nature and the extent of the impact of infertility on those individuals and couples affected. In an effort to contribute to the growing body of counselling literature which seeks to understand the nature and impact of infertility as a significant life event, it is hoped that this study will provide valuable information for counsellors, to be used as a guide for assessing risk factors as well as potential sources of strength, as they provide support for couples and individuals enduring the transition of infertility. Assumptions. Limitations and Delimitations Assumptions  The most fundamental assumption underlying the present study is in regards to the nature of causality. While it is beyond the capacity of a correlational research design to suggest that infertility 'causes' the various psychological symptoms associated with the experience, the author rejects the notion that psychological symptoms 'cause' infertility. This latter assumption has motivated a significant body of research, with varied, often contradictory findings (Noyes & Chapnick, 1964; Pantesco, 1986). The present study approaches the question from a fundamentally different theoretical perspective and suggests that the question of whether infertility 'causes' psychological symptoms, or the reverse, is not relevant. Rather, the present study represents an attempt to understand the nature of the relationship between characteristics of the individual and of the infertility experience and the patterns of response to a complex and profoundly distressing human event. Limitations  There are a few important limitations to the present study which must be addressed. The limitations are consistent with those inherent in survey research. The representativeness of the sample is the most important concern as this affects the ability to generalize the results of the study. Three important concerns may be raised regarding the representativeness of the sample. Due to the nature of the research question, the sample has not been selected randomly and is, of necessity, a 'sample of convenience', restricted to a specialized population. The population the 15  Statement of the Problem  sample is recruited from consists of infertile adults who are seeking medical support for infertilityrelated concerns. This is the population the results may logically be generalized to. Two other concerns may be raised regarding the extent to which the sample is representative of infertile adults in demographic terms. The sample is predominantly (84%)white and anglophone, with only minimal representation from other racial and ethnic groups. There are considerable restrictions, therefore, in the ability to generalize the results of this study to multicultural populations. In addition, there is some concern regarding the recruitment of the sample from a single clinic located in an urban area. Foothills Hospital's central location makes it the primary venue for infertility services in southern Alberta, serving both urban and rural populations. It may be argued, therefore, that the sample population may be representative of white, anglophone, rural and urban infertile adults who seek expert medical attention to address their concerns. Another important concern is with the validity of responses collected on self-report measures. With this form of instrument there is always some difficulty in obtaining candid responses which are unaffected by social desirability factors. Measures were taken to minimize these effects. Subjects were assured anonymity prior to consenting to participate and were further assured that medical and other treatments received in the clinic were in no way to be influenced by their participation in the study. In addition, the importance of providing uncensored responses, to help in developing a means of support for other people experiencing infertility, was explained and stressed during these interviews. Nonetheless, it is likely that, despite the assurances and encouragements, the self-consciousness engendered by the act of completing self-report measures may have caused some respondents to unintentionally moderate their responses, thus potentially affecting the validity of responses. While the use of survey and correlational design in the present research project serves to place minor limitations on the researcher's ability to control elements of the process of data collection, these limitations are not uncommon in research designs of this kind. Measures have been taken to address each of these concerns in the present study. 16  Statement of the Problem  Another area for concern is in the use of a single trial method of sampling. The infertility transition may take many months or years to complete and may involve many changes in mood and affect. While the focus of the study is primarily on the mediating factors, which may be assumed to be constant, the level of functioning at the time of assessment may not be representative of the individual's functioning throughout the experience and may limit the validity of the findings. However, the results may be representative of a random selection of infertile adults at a given time. Finally, there are limitations inherent in the ability to accurately measure variables used in the study. Of particular concern is the measurement of the construct of 'psychosocial competence', (Schlossberg, 1981; Tyler 1978), a complex construct including a variety of attitudinal and behavioral components. The measures used in the present study represent an attempt to approximate two of the three primary components of psychosocial competence, 'positive selfattitudes' (self-esteem) and 'realistic world attitudes' (locus of control). There is no corollary measure to assess the construct's third primary component of an 'active behavioral orientation'. As no appropriate measure was available for assessment of an active behavioral orientation, it would be necessary to create such an instrument. Such an exercise may comprise the focus of a separate study, and was therefore deemed to be beyond the scope of the present study. Delimitations  Schlossberg's (1981) model of transitions in adult development has been used in the present study as a guide for identifying and selecting appropriate variables for study in the investigation of individual differences in response to infertility. Although Schlossberg identified three sets of factors which may be viewed as significant in mediating the individual's ability to adapt to a transition, including characteristics of the transition, characteristics of the environment and characteristics of the individual, the present study selected only the characteristics of the transition and the characteristics of the individual for investigation. Although characteristics of the environment were not directly studied, these factors were not ignored and were, in fact, largely controlled for. Two elements of the environment were controlled through the selection process. Physical surroundings, including such things as climate 17  Statement of the Problem  and geography, were consistent across the sample, as subjects were recruited from a single geographical region within the term of one month. The availability of institutional supports was also comparable for all subjects, as each was receiving specialized medical support at the time of study. The third element of the environment identified by Schlossberg is interpersonal support. The availability and effectiveness of interpersonal support is an area of specific relevance to a study of the impact of infertility and has, in fact, generated a notable collection of research efforts. The results of these studies consistently suggest that interpersonal support is problematic for infertile individuals and couples, often appearing to increase levels of distress experienced (Davis, 1987; Lalos et al., 1986; Menning, 1980; Sandelowski & Jones, 1986). As these results appear to be highly consistent, ineffective interpersonal support is assumed in the present study. A final word must be addressed to the issue of recruiting and the resulting number of subjects surveyed. An initial design included the use of two separate sites for recruitment of subjects, including the Infertility Clinic at Foothills Hospital in Calgary as well as the clinic in the Department of Obstetrics and Gynaecology at Grace Hospital in Vancouver. This plan was devised primarily in response to concerns that an insufficient number of subjects may be recruited from a single clinic. However, in less than one month of data collection, more than two-thirds (106) of the overall target number of subjects (150) had been contacted. Furthermore, the rate of return for completed questionnaires was encouraging (approximately 70%). A decision was made to refrain from further data collection, thereby including some control over environmental factors by recruiting from a single geographical location. Although a larger sample size would undoubtedly provide more complete information, as well as enhancing the validity of the findings, a review of the available data suggests that the sample is representative of a range of scores on many of the variables. Moreover, it must be noted that the present study is exploratory in nature and seeks primarily to understand whether there are differences between individuals in ways of responding to infertility. Any patterns or trends suggested by the results of the present study may be understood as indicative of directions for future research. 18  Review of the Literature  CRaIERIWQ Review of the Literature Introduction  This chapter will present a review of the literature relevant to the present study's inquiries regarding psychological and emotional responses to infertility. Following a brief summary of the history of research regarding the relationship of infertility to psychological factors, the chapter will focus on studies which examine the psychological impact of infertility. The chapter will conclude with a review of the literature seeking to identify personal and situational factors which mediate or predict the nature and intensity of the impact of infertility on men and women. Historical Approach to Psychology and Infertility  Historically, infertility research is characterized by attempts to determine the causes and possible cures or treatments for the condition of infertility. Within the past two decades, great strides have been made in the medical sector of the field in discovering congenital and traumabased physical irregularities which correlate with and contribute to repeated difficulties in reproduction. From approximately 50% of cases which were physically explainable thirty years ago, currently up to 90% of cases can be traced to a physical malfunction in either the male or female (Leader & Taylor, 1984). In consideration of this rapid expansion of medical understanding of the factors affecting fertility and the concurrent rapid progress in medical microtechnology, it is not inconceivable that twenty more years, if not fewer, will bring the answers to the puzzles of the remaining 10% of currently 'unexplainable' instances of infertility. Not surpisingly, considerable attention has been focussed on the issue of unexplained infertility. In a recent paper, McBain and Pepperell (1987) outlined a number of possible explanations for 'unexplained infertility'and suggest that further investigations may soon result in explanations, if not solutions, for those cases. A large portion of the research attention is aimed however, not at discovering the organic source and subsequent treatment for these currently unexplained cases of infertility, but at seeking support for the common assumption that psychological or emotional 'blocks' within the individual 19  Review of the Literature  play a causative role in impeding reproduction. The present study rejects the idea of a causative link between psychological factors and infertility. Two reviews of this literature follow, each of which concludes that a search for a psychogenic cause of infertility appears to be futile, and suggests that the direction of any link may in fact be the reverse (Noyes & Chapnick, 1964; Pantesco, 1986). Noyes and Chapnick (1964) reviewed 75 research articles focussing on the relationship between psychology and infertility, which were published between 1939 and 1963. Utilizing a five-item evaluation format, the authors, each representing a separate discipline (one medical, the other psychological), independently rated each paper on a scale of 0 to 2 on each of the following dimensions: a clearly stated hypothesis regarding the causative relationship of psychologic factors to infertility; objectivity of methods and materials for measurement of variables; statements or evidence suggesting that psychologically oriented treatments may be palliative to a condition of infertility; statisical support for a conclusion that psychological factors affect infertility, and; the contribution made by the article to the furtherance of knowledge in the field. There was a fairly high degree of agreement (r= 0.64) between reviewer ratings on each paper. Out of twenty possible points for each article (2 points x 5 diminsions x 2 raters), the average score achieved by the 75 papers was 8.6, indicating that the majority of articles approaching the question of psychogenic bases for infertility failed to meet even the most basic standards of experimental rigor, as measured on the rating scale. Interestingly, the three highest scoring papers (two achieving scores of 17, one with 18) described well-controlled studies which provided evidence to refute the popular belief that adoption increases fertility. Noyes and Chapnick (1964) found that "many authors took it as axiomatic that psychic factors influence infertility, and that psychogenic infertility may be diagnosed by the exclusion of organic factors" ( p.553). Moreover, the large number of psychic factors cited as causative of infertility suggests researcher bias, as the cited research "certainly contains enough variety to support almost any preconceived opinion" (p.553). Within the 75 articles, 4.0 different factors were listed and many articles cited multiple factors. Largely because of the "vagueness of 20  Review of the Literature  hypotheses"(p.553) found in most of the articles reviewed, there was some discrepancy between the two reviewers in recording the factors indicated by each author. While the two researchers agreed on 181 cases of specific factors being cited, each reviewer recorded a number of cases which were not recorded by the other, for a total of 487 recorded instances of psychogenic factors being cited as causes of infertility. Despite the abundance of possibilities, "this analysis has yielded no evidence that specific psychologic factors can affect fertility"(p.554). More recently, Pantesco (1986) critically reviewed a sample of thirty-one research articles published between 1942 and 1983 and found that the majority sought to establish the existence of psychogenic causes of infertility in women. He noted that, until the early 1970's, the emphasis in the literature was predominantly on identifying psychodynamic reasons for the inhibition of pregnancy in women. These included such constructs as the following: masculine-aggressive personality, feminine immature personality, functional derangements, anxiety about one's feminine role, unconscious rejection of pregnancy, and hostility (to mother or to husband); many of the same constructs recorded by Noyes and Chapnick (1964). Pantesco concluded that a "biased emphasis on women's physical and emotional culpability" (p.733) has, until recently, coloured the approach to infertility research. An important aspect of the sexist bias inherent in the older infertility research lay in the fact that the overwhelming majority of studies focussed exclusively on women as source as well as the primary victim of infertility. Two-thirds of the thirty-one studies reviewed by Pantesco focussed exclusively on identifying psychopathology and psychosomatic disorders in infertile women. Less than one-third of the articles reported observations of both members of the couple. Only one study focussed exclusively on male subjects. Pantesco noted that, while recent researchers have acknowledged the importance of studying males as important elements in the "system of the couple" (p.734), the inclusion of men as subjects of study remains infrequent. He also noted, however, that men's restricted availability for research and unwillingness to participate constitute a major obstacle to rectifying this imbalance. Pantesco's review included a critical look at methodological issues in research relating  21  Review of the Literature  psychogenic factors to infertility, and found much of it problematic. Like Noyes and Chapnick (1964) he found a body of research characterized by vaguely stated hypotheses, loosely defined variables, small sample sizes, reliance on nonstandardized measures, poorly analyzed data and a reliance on assumptions in place of evidence. Both reviews also noted the lack of communication between researchers in the field. The result has been a body of literature with often contradictory and largely inconclusive findings regarding psychogenic causes of infertility (Noyes & Chapnick; Pantesco). While almost half a century of research aimed at seeking psychogenic causes to unexplained infertility has been singularly unsuccessful, the belief in the existence of such a relationship remains strong in popular mythology and such beliefs continue to inspire research efforts into the question (see Domar, Seibel & Benson, 1990). While no infertility research has yet shown a priori psychological symptomatology, to support a claim for causation, a number of psychological symptoms have been found to be associated with infertility. Depression, anxiety, obsessive-compulsive behavior and feelings of guilt have all been observed in subjects experiencing infertility (Daniluk, 1988; Lalos et al., 1986; Seibel & Taymor, 1982; Walker, 1978). Among these symptoms, depression has been most consistently reported. In light of the inconclusive nature of research seeking a psychological cause for infertility, an alternative explanation is warranted to account for the prevalence of these various psychological symptoms noted among infertile individuals. The present study adopts the view that psychological symptoms found to correlate with infertility can better be understood as the effects, rather than the cause, of a profoundly distressing developmental crisis. Impact of Infertility Much of the early exploratory literature describing the psychological impact of infertility consists of anecdotal reports from infertility support group meetings as well as interviews conducted by social workers, physicians and nurses working with infertile couples (Daniels, Gunby, Legge, Williams & Wynn-Williams, 1984; Kraft, et al., 1980; Menning, 1980; Menning, 1982). Although these studies lack the rigor of experimental methodology, they have 22  Review of the Literature  provided more recent researchers with a rich source of information on the impact of infertility. A handful of studies have utilized interview methods with infertile subjects in formal attempts to explore the impact of the experience in an in-depth manner. The following studies are representative of the scope and nature of this research approach. Using a structured interview format, Valentine (1986) sought to identify elements of the emotional impact of infertility on married couples. Twenty-six subjects aged 25 to 38, including twelve couples and two women, took part in a two hour semi-structured interview designed to encourage in-depth examination of the individual's emotional responses and coping strategies associated with infertility. Issues and concerns raised were reported in terms of five main topics or themes, including emotional impact, crisis, loss, multiple stressors and coping strategies. In describing the emotional impact of infertility, subjects reported feelings of "sadness, depression, anger, confusion, desperation, hurt, fear, embarrassment, humiliation, disappointment, unfairness and unfulfillment..(as well as)...reactions such as disorganization, distractability, exhaustion and fatigue, moodiness, unpredicatability and obsessive behaviors and thoughts" (Valentine, 1986, p. 63). Other themes identified by Valentine (1986) included the prevalence of feelings of loss, for example "loss of potential children; loss of genetic continuity; loss of pregnancy, childbearing and breast-feeding experiences; loss of a life goal; and loss of control over one's body."(p. 64). Subjects also identified numerous sources of stress, including intrusive medical practices and the insensitivity and ignorance of family and friends, as well as the marital relationship. Valentine also describes the various coping strategies commonly reported by infertile interview subjects, including becoming 'obsessed' with infertility, avoiding contact with parents or pregnant friends and seeking support from family, friends and infertility support groups. Unfortunately, the usefulness of these results is limited due to some fairly important methodological problems. The sample is very small and is skewed toward upper-middle class whites, making generalization of the results to other populations difficult. More seriously, however, no mention is made of the method by which the reported themes were developed, leaving  23  Review of the Literature  the impression that no systematic approach was employed. The study is also marred by a somewhat confused organization and theoretical approach. Of the five topics around which the body of the article is organized, only the four discussed above represent elements drawn from the content of the interviews. The fifth topic is a discussion of the crisis model as it applies to the experience of infertility. Although this appears to be a theoretical model, it is not identified as such, but is instead included with the results. In fact, no theoretical basis is stated for the research. These concerns represent significant flaws in Valentine's (1986) study and make generalization of the results tenuous. The descriptions of emotional responses, such as sadness, depression, anger, despair, disappointment and obsessive thoughts, as well as the many forms of loss experienced by infertile couples are supportive of clinical findings (Daniels et al., 1984; Kraft et al., 1980; Menning,1980) . Thus, although flawed, Valentine's study supplements and supports the literature describing the serious emotional impact of the experience of infertility. Utilizing similar research strategies, Woollett (1985) approached the question of how infertile individuals cope with the impact of infertility. Subjects included forty-two females and eight males, with primary or secondary infertility. Employing a semi-structured interview format , Woollett focussed on the individual's feelings and actions in response to their infertility as well as on the impact of the experience on their relationships with others. Four primary issues or themes were raised by a majority of interview subjects. Infertility represented a major disruption in an individual's life and prompted attempts to  regain control. All subjects sought medical help, which initially provided a sense of control. This strategy also provided a means for subjects to redefine their goals and expectations. Many subjects, however, reported that medical investigations represented a loss of control, as medical personnel virtually took over their lives, dictating the scheduling of monthly appointments and even sexual intercourse. Interview subjects reported seeking ways to understand their infertility and often became  'infertility experts'. This strategy appeared most successful in reducing stress in cases where resolutions were relatively swift. Although many of the interviewees were highly knowledgeable  24  Review of the Literature  about the procedures and investigations employed, many appeared highly optimistic regarding their chances of successful conception, despite relatively low success rates of many of the procedures (Woollett, 1985). Subjects unanimously reported contending with negative images, held by themselves and others, associated with infertility. Many women and most of the men appeared to perceive their difficulties as evidence of sexual inadequacy. Several subjects reportedly generalized their feelings of incompetence to include a sense of failure in interpersonal, intrapersonal and work skills. Some interview subjects appeared to deal with negative feelings by denying infertility, despite ongoing infertility investigations. As Woollen (1985) notes "strategies of denial seemed to be more readily available to men than women " (p.479). Finally, a major issue identified by the subjects of Woollett's study was the experience of coping with the many losses associated with infertility. In addition, subjects identified a variety of strategies for coping with infertility. Aside from those already mentioned, which included seeking medical help, redefining the problem in medical terms, becoming infertility experts and denial of infertility, many subjects attempted to seek support from friends and family. Like the subjects in Valentine's (1986) study, however, many of Woollett's (1985) subjects found it diffiicult to obtain helpful support from others because of their lack of understanding or insensitivity. As a result, many subjects reported feelings of isolation and having increased demands placed on the marriage to meet emotional needs, creating an additional strain on the relationship. Some subjects reported making attempts to create positive self-images, redefining life goals with regards to childbearing and parenting and seeking new ways to meet these needs. Woollett (1985) indicates, however, that these strategies were seldom pursued with enough vigor to operate as successful coping mechanisms. Woollen concludes that the primary methods utilized for coping with infertility tend to be focussed on the search for a medical solution to the problem. While this approach is successful in a majority of cases, Woollen suggests that an over-reliance on this approach may result in a tendency to avoid or ignore alternative strategies and solutions. The methodological flaws in this study are similar to those seen in Valentine (1986).  25  Review of the Literature  Although Woollett's (1985) sample is larger (50 individuals), the very low representation of males (8), renders any comments regarding the results from this group highly tentative. Furthermore, as with Valentine, no theoretical or methodological models were described, suggesting none were utilized. This study and the previous one must be regarded as exploratory and as such, they are not without value. The absence of experimental rigor in these studies and others of their kind has allowed the latitude to explore the issues and concerns raised by infertile couples with a depth and breadth not usually possible in most controlled studies. Research of this kind provides a rich source of information and direction for those wishing to examine the impact of infertility in more controlled settings, using standardized measures and methodology. Indeed, the following studies represent a growing body of research using structured formats and measurements to address a number of questions regarding the impact of infertility . Sandelowski and Pollock (1986) report on a well-designed phenomenological study aimed at discovering the 'meaning of infertility' to women enduring the experience. Forty-eight women who were currently infertile were interviewed by one of the two primary researchers. The first 22 interviews were conducted in subject's homes with women receiving medical attention from private physicians. These interviews were unstructured and focussed on the general topic of "what it's like not to be able to have a baby when you want to" (p.140). Using 'phenomenal analysis', the information gathered from these interviews was systematically analyzed for dominant and recurrent themes. Second interviews were then conducted with 20 of these women as well as four more women recruited from private physicians, to validate and further refine the themes. At this time, a second sample of 22 women was recruited from a public infertility clinic for the 'medically indigent'. Structured interviews were conducted with these women to validate the themes. Three main themes were identified in the women's descriptions of their experience of infertility. The most commonly identified theme was 'ambiguity', a feeling of being "'in limbo,' 'on hold', of 'dangling', and living in a 'gray area' (p.142). The ambiguity theme represented a collection of similar feelings and experiences reported by almost 94% of respondents. Elements 26  Review of the Literature  of the ambiguity expressed by the majority of women include a sense of uncertainty regarding the causes or reasons for their infertility as well as feelings of ambivalence toward medical personnel. Between 12% and 19% of subjects expressed some of the following feelings of ambiguity: confusion and frustration regarding life goals and plans; lack of confidence in ability to cam/ a pregnancy successfully; discomfort and ambivalence about the safety and effectiveness of the medical procedures used suspicion and uncertainty regarding own past and future fertility, and; insecurity about the amount of control they have in their lives. A second theme identified was 'temporality', described as "a heightened consciousness of chronological and biological time" (p.143). The elements of the temporality theme expressed by most subjects included: a concern with time limits repreatedly being set and reset; frustration over wasted time or time 'running out; time-consuming rituals and the need to follow them according to a strict schedule ; awareness of time marked by menstrual cycles; the need to continually plan, wait and review; the delay of other plans contingent on the timing of childbearing; and for a few, a sense of time being slowed. The third theme these researchers identified in the experiences of infertile women was 'otherness', or a "feeling or separation and deviance" (p.144). The major elements of the 'otherness' theme were the following: a preoccupation with making comparisons between self and other women; a feeling of being unjustly 'singled out' for their fate; a sense of being estranged or excluded by others; a belief that other women, including fertile women as well as formerly infertile women, cannot understand how they feel. A few of the women interviewed expressed their feelings of 'otherness' as a belief they were somehow defective. Efforts were made to use systematic procedures and controls in this study. To increase the generalizability of the findings from the original sample of volunteers, who were predominantly white and middle class (75% of this sample had average family yearly incomes over $36000, 64% had university or technical training), the researchers sought a socioeconomically different group to corroborate their thematic findings. Of the second group of 22 women, 90% were black, over 85% had an average family yearly income of below $36000, only one had university or technical  27  Review of the Literature  training and 27% had not completed high school. Although the inclusion of these two disparate groups demonstrates an effort to be more fully representative of North American women, there is evidence that these two groups were treated differently. Interviews with the first group lasted from two to four hours, as compared with interviews as short as thirty minutes with the second group. Several differences were found in the experiences of these two groups of women, with the clinic group providing significantly fewer examples of virtually all elements of the themes than the private practice group. Whether this contributed to, or is a result of, the shorter interview periods, is unclear. Although no statistical comparisons were performed, as the study was not designed to address any questions of contrast, the researchers suggest that there may be real differences in the way these two groups of women interviewed perceived their infertility experiences. The poorer women appeared less likely to express feelings of ambiguity, temporality and otherness in regards to infertility specifically, perhaps because of being relatively familiar with these experience in general, already having an "acquaintance with doom and fate...(and a)... well-developed sense of separation from mainstream American life" (p.144). It is important to note that, although no attempt was made to quantify the differences through statistical comparisons, these researchers suggest that the way an individual woman experiences infertility may be, in part, a function of her socioeconomic status. In other words, they suggest that socioeconomic status may be a mediating factor in the impact of infertility, such that women with a lower socioeconomic status appear to actually report fewer stressful experiences associated with infertility than women with higher socioeconomic status. Although these differences may be largely reflective of the differences in the format, content and even the settings of the interviews conducted with the two groups, the findings nevertheless point to the need for further investigation of the effects of socioeconomic status on men's and women's experience of infertility. Using the same data collected from the forty-eight women interviewed by Sandelowski and Pollock (1986), Sandelowski and Jones (1986) identified a second thematic structure. After 28  Review of the Literature  examining the structure and content of the interactions related by interview subjects, these researchers suggested that coersion and rejection characterize the social exchanges of infertile women. Coersive interaction were identified by many infertile women, who described that they often felt forced to explain their childlessness to family, friends and even strangers, who asked questions which were perceived as intrusive, pushy and overly personal. Several women also reported feeling forced to 'put on a happy face', to avoid alienating or offending others with their negative feelings. Many reported feeling forced to control the expression of their pain and frustration, in recognition that friends and spouses quickly tired of hearing about it. Women also described the effects of their infertility on others, reporting that family and friends habitually withheld information from them and avoided talk related to pregnancies in their company, making them feel 'left out' and 'different'. Interactions wherein infertile women openly or privately rejected the advice and support offered by others comprised the second social pattern identified. Infertile women described developing a covert set of rules relating to who could give advice and support and what kind of advice was acceptable. Believing that only those who have gone through the same experience, or one very similar, could possibly understand their feelings, many women tended to reject any advice or support from 'normals', perceiving much of it as worthless or offensive. Moreover, the women appeared to divide the world into separate spheres, one containing the 'sisterhood' of infertile women, with everyone else, even formerly infertile women, occupying the second sphere. The primary weakness of this study, as suggested above, lies in the method by which the samples were chosen and used. Although an effort was made to promote generalizability of the findings to a more diverse population, the primary thematic findings of this and the former study appear to be predominantly reflective of the experience of white middle class infertile women, with the information gathered from the poor and black women being used primarily to support, rather than expand, the themes. Despite this limitation, these two studies provide a comprehensive and in. depth picture of the emotional and psychological turmoil experienced by many infertile women. 29  Review of the Literature  An early study by Wiehe (1976) focussed on how individuals "psychologically react to their infertility" (p. 863) and how they handle their reactions. Twenty-two couples, who had reportedly been aware of their infertility for two to six months and were described as having 'unsuccessfully sought medical attention for infertility', were recruited from among the applicants to a private adoption agency in the U.S. Two self-report pencil and paper surveys were mailed to subjects. The primary measure was a researcher-developed version of the Semantic Differential, designed to reflect attitudes toward specific elements of the experience of infertility. A second measure consisted of scales measuring aspects of personality and adjustment. The mean score for infertile subjects on the Semantic Differential was 4.5 on a scale of one to seven, indicating a neutral to somewhat positive attitude toward infertility. These scores were compared with those of the personal adjustment scales, resulting in some significant correlations. Subjects who indicated more positive attitudes toward aspects of infertility tended to score high in defensiveness and also tended to score low on the self-awareness, suggesting that individuals who have relatively poor self-awareness and who tend to be more defensive in their approach to selfperceptions are less likely to report perceiving infertility in a negative light. The results of the study indicate the surprising fmding that some infertile subjects hold a neutral attitude toward elements of the infertility experience. As this finding appears to be contrary to expectations, based on the results of many of the studies previously cited, some attempt at explanation is warranted. Wiehe (1976) offers three interpretations of his result. First, he suggests that response bias to the Semantic Differential may have obscured subjects' true attitudes. He dismisses this possibility because of the measure's 'unobtrusive' qualities, and because subjects were assured that their responses would not affect their adoption privileges. It is possible, however, that, despite such assurances, couples in the process of application and adjudication for adoption may have felt pressured to avoid any appearances of negativity. As noted elsewhere "adoption workers expect...(applicants)...to demonstrate that they had resolved their feelings about their infertility" (Valentine, 1986, p.66).  30  Review of the Literature  Wiehe continues in his explanation of his findings by discrediting the results of earlier studies which indicated that depression and disappointment were the reactions to infertility, and that these reactions were readily apparent to both observers and to infertile subjects. Regardless of possible methodological flaws in the two studies he cites, substantial evidence of depression and disappointment in response to infertility have been found repeatedly since the publication of Wiehe's report (Daniluk, 1988; Kedem et al. 1990; Feuer, 1983; Menning, 1980; Valentine, 1986; Woollett, 1985). The third and most compelling explanation offered by Wiehe (1976) for the surprisingly neutral attitude toward infertility expressed by the subjects in his study is suggested by the fact that this sample had been aware of their infertility for between two and six months. Wiehe suggests that these subjects, having only recently learned of their infertility and having not yet achieved parenthood through adoption, may have been in a period of denial of their feelings. He suggests that the couples represented in the other studies cited, who had been dealing with infertility for two to ten years and had either successfully adopted or had made an "adjustment ..to non-parenthood" (p. 866) were more able, from a 'safe distance', to acknowledge their negative feelings regarding infertility. This explanation is supported by the fmding that a positive evaluation of infertility was related to poor self-awareness and high defensiveness. An alternative explanation for the unexpected results, not identified by Wiehe, is that the sample of infertile subjects recruited for his study represented a unique sub-population within infertile people and was not therefore representative of infertile individuals in general. The sample was drawn from the list of applicants for an adoption agency. It would be reasonable to expect that most couples seek to adopt only after having exhausted all hope of medical intervention in achieving a pregnancy, a process which takes many months and often years to complete. In fact, the majority of infertile couples seeking interventions, both medical and adoptive, have been aware of their infertility for a minimum of one year. Yet the sample in Wiehe's (1976) study reportedly had been aware of their infertiilty for only two to six months and were already seeking adoption. This fact alone may qualify this group as a distinct sub-group. 31  Review of the Literature  Whether any of these explanations is valid is impossible to ascertain, as no information on this sample, other than that already reported, is provided. What is clear is that these results are not consistent with the majority of research findings, which indicate that infertility is an intensely felt and predominantly negative experience. The following study describes some of the negative elements of infertility. Bell (1981) conducted an investigation into the 'psychological adjustment' of infertile men and women. This researcher recruited twenty couples attending an infertility clinic with presentations of primary infertility. One group of ten couples were seeking treatment for infertility for the first time. A second group of ten couples had received diagnoses and had been undergoing treatment for some time. Interviews were conducted separately with each of the forty subjects and were designed to elicit responses regarding feelings about infertility, as well as to clinically assess levels of anxiety, depression, as well as the individuals' social, sexual and marital adjustment. Each subject also completed questionnaires measuring sexual motivation, attitude to marriage, social adjustment, anxiety and depression. Descriptive data presented appear to indicate that female subjects reported higher levels of anxiety and depression than did male subjects, but no statistics were reported describing the extent of these differences. Those subjects who scored higher on measures of anxiety and depression also tended to score higher on the measure of social maladjustment. Twenty percent of the females were reported to exhibit clinically significant levels of 'social maladjustment', although it is not stated whether these women represented the pre-diagnosis group or the treatment group. Bell (1981) states that the extent to which the observed symptoms may be attributable the infertility itself, or to disturbances pre-dating the infertility "is unclear" (p. 3). The lack of clarity in his conclusion is symptomatic of the vagueness of his purpose and of his methodology. For example, the recruitment of two distinct (if similar) subject groups suggests a comparison study, but no comparisons were reported. Further, his failure to report on the extent of the differences reported by men and women on the measure of anxiety and depression is curious. In addition, the small sample size (40 individuals) makes generalization of the findings tentative. Nonetheless,  32  Review of the Literature  these defects do not discredit the evidence suggesting that anxiety and depression, as well as decreased sexual and marital satisfaction are among the negative effects of infertility. Bell closes with an assertion that longitudinal studies are required to address the impact of infertility and to direct the "provision of supportive counselling" (p. 3) for infertile couples. The following study utilitzes just such an approach to directly address the emotional impact of the infertility investigation on the individual. Many authors have described the tremendous fluctuations in emotions throughout the often long drawn out phases of the infertility work-up as one of the distinguishing and distinctly stressful characteristics of the infertility experience (Davis, 1987; Kraft et al. 1980; Halpern, 1989; Woollen, 1985). Edelmann, Connolly and Robson (1989) have reported the preliminary findings of a long-term study which follows couples from their initial contact with an infertility clinic throughout the medical investigation. Data was collected from all new patients attending an infertility clinic, with results from the first four couples to complete a 22-week investigation being reported. During the initial medical interview Edelman et al. (1989), collected demographic and background information and asked each patient to complete questionnaires measuring personality, general health, depression, state and trait anxiety, marital adjustment, interpersonal support and sex role orientation. These measures were repeated at the end of the 22 weeks. To chart the week-toweek changes in affect and distress, each patient was asked to complete a questionnaire assessing levels emotional distress each week and to keep a weekly diary describing any stressors experienced, both infertility related and otherwise. Scores on the personality and psychological measures for the eight individuals fell within normal ranges. State and trait anxiety scores for both members of two couples were high, as compared with standardized norms, but were comparable to norms obtained from a population of surgical patients. Although no statistics were reported, scores obtained at follow-up tended, in general, to be lower than those obtained at intake. The authors suggest that this may be reflective of elevated levels of distress experiences at the initial visit. Due to the small size of the sample, no 33  Review of the Literature  predictive statistics are reported regarding the relationship between scores at intake and the weekly levels of emotional distress. Charts depicting the weekly changes in distress level scores exhibited distinct fluctuations in mood and affect throughout the 22-weeks for three of the four couples. These charts were compared with the dates of the various test procedures conducted, and of the stressors reported by the patients. For all four couples, the 'peaks' (reflecting high levels of negative affect) coincided strongly with infertility related stressors, such as testing dates, the beginning of drug treatment, the onset of menses and receipt of diagnosis, as well as such events as a visit from relatives with an infant. Data collected from such a small sample cannot be generalized to a larger population, and the authors make no attempt to do so. This study is of interest, however, because it demonstrates the reported experience of many infertile couples, that, rather than being a single crisis which can be responded to and then recovered from, infertility consists of a series of large and small crises, eliciting a series of response-and-recovery cycles. Further, the use of case illustrations dramatically illuminates the differences between individuals. Although there were similarities in general terms, the eight different patients exhibited eight different patterns of response to the experience, highlighting the need to identify characteristics which might identify those individuals who are most likely to require support and counselling though the experience. The research studies reviewed here describe the emotional impact of infertility. These studies have provided evidence that men and women tend to respond to infertility with attempts to regain control over their bodies and their lives and experience feelings of loss, depression, anxiety, hopelessness, and anger. Although many of these responses are exhibited in a significant subgroup, or even a majority of infertile people, none of these responses is universally reported. It is evident that some individuals tend to exhibit greater distress and emotional turmoil in response to infertility than do others. The following section reviews a variety of studies which have more closely examined the characteristics of the individual and of the infertility experience which may have impact in moderating or otherwise influencing the impact of infertility. 34  Review of the Literature  Factors Mediating the Impact of Infertility The following review will examine studies which have identified one or more factors as influential in mediating the impact of infertility on men and women. The influence of gender is perhaps the most commonly reported mediating factor, but the nature of gender influences remain unclear and often controversial. Evidence has been reviewed suggesting differences in the symptomatic distress reported by men and women. For example, Bell (1981) and Daniluk (1988) have reported differences in men's and women's assessments of their marital relationship while undergoing treatment for infertility, but these differences are in opposite directions; Bell reporting a more critical assessment by females than by males, with Daniluk reporting a trend toward more positive assessments by females than by males. Brand (1989) explored sex differences in acceptance of infertility in a sample of fifty-nine couples attending an Infertility clinic. Ninety-two percent of couples were diagnosed with female factor infertility, with the remaining eight percent receiving a diagnosis of male factor. Brand interviewed each subject, utilizing a semi-structured interview format designed to elicit subjective responses regarding the effects of infertility on the individual and the relationship. Using a chi-square analysis, Brand (1989) reported a number of significant differences between men and women in their behavior in response to infertility. Women were found to talk more frequently and more willingly about infertility. As well, women reported feeling more intense levels of disappointment and feelings of being 'overwhelmed'. Brand also reports that men and women subjects agreed that the 'psychological impact' of infertility was greater for women than for men. The 'significant' gender differences reported by Brand (1989) are derived from an analysis of interview data which elicited subject's opinions on their own reactions and the reactions of their spouse. No standardized measures were utilized. Although the method of data collection is essentially qualitative, the reporting of statistical results suggests an effort to misrepresent the study as quantitative. The results of Brand's study must therefore be regarded as unreliable. While the methodological flaws described above are sufficiently serious to view Brand's  35  Review of the Literature  (1989) study with serious misgivings, the underlying, unacknowledged gender bias which informed the study is of greater concern. Aside from whatever validity Brand's results may or may not have, the 'gender differences' reported in the study are problematic because they are presented without interpretation or any real attempt at discussion, as indicative of women being more negatively affected by the stress of infertility than men. By comparing males to females and interpreting the lower reported symptomatology reported by men as a lack of symptomatology, Brand ignores the fact that men and women characteristically demonstrate different patterns of response to stress. The differences in the ways males and females respond to and report stressful events are many. The following study offers a comprehensive description of gender differences in response to infertility. Chiappone (1984) used Schlossberg's (1981) model of transition as a framework for examining the differences, in impact and coping strategies, between men and women experiencing infertility. Chiappone recruited 111 couples through advertisements in Resolve newsletters. Volunteers were mailed packages including standardized measures of subjective distress, coping strategies, depression, self-esteem, sense of mastery and a researcher-designed questionnaire assessing the impact of infertility, feelings about infertility and the availability of social supports. Chiappone (1984) found several significant differences between men and women in their responses to infertility. Women rated the negative impact of infertility on self-image, marriage, sexuality and career significantly higher than did men, reporting that infertility had affected all four areas, whereas men reported that their self-image, marriage and sexuality were affected somewhat, but that their careers were not at all affected. Men and women also showed significant differences on a measure of their feelings about infertility, with women tending to indicate significantly greater feelings of frustration, anger, depression, anxiety, defectiveness, desperation, shock, and isolation than men. Women also indicated significantly more willingness to do anything, more feelings of stress in the sexual relationship and a greater sense of being misunderstood, but also indicated greater feelings of strength and closeness to their spouses. Women also indicated significanity less feelings of 36  Review of the Literature  hopefulness for a child, of coping okay, of success, of control in their lives, of being attractive sexually, of happiness and of relief. It must be noted that, despite the statistical significance of the gender differences, the response pattern of male and female subjects was very similar. Although no correlational analyses were performed supporting the pattern, it is apparent from a review of the data that there are more similarities than differences between the feelings about infertility reported by these men and women. This pattern was also seen on other measures, with women's average scores on emotional distress and depression being significantly higher than men's and their scores on mastery and self-esteem being significantly lower. Although all of these differences were statistically significant, none were large. These patterns suggest that both men and women experience infertility as deeply distressing. The differences between the sexes in their responses to infertility appears to largely one of degree, with women displaying a tendency to report feelings, both positive and negative, with somewhat greater intensity than men.  For both men and women, a number of factors were found to be predictive of depression and of the duration of infertility. For men, high mastery and positive self-esteem were negatively related to depression. For both men and women, depression scores were positively correlated with the tendency to report intrusive thoughts and feelings. For women, a report of intrusive thoughts was the best predictor of depression. In addition, as with men, women who scored high in mastery tended to report low levels of depression. For both women and men, self-esteem was found to negatively correlate with duration of infertility. In addition, for women only, the duration of infertility was found to correlate positively with the number of coping strategies utilized, such that the longer a woman was infertile, the more coping strategies she used. Women also reported a significantly broader range of coping strategies. Although male subjects reported using only 51% of the strategies included on the Ways of Coping Checklist, women reported that they used 63% of the strategies listed. Women and men were also found to differ in the type and variety of coping strategies utilized, with women using considerably more 37  Review of the Literature  types of strategies than men. For example, women sought social support, engaged in wishful thinking and self-blame, utilized problem-solving strategies and adopted tension-reduction strategies significantly more than did men. The ways of coping most often reported by men were strategies such as 'maintaining pride', 'keeping a stiff upper lip', 'refusing to think about it' and 'taking things one step at a time; strategies which could appropriately be described as methods involving suppression or avoidance. Methodologically, Chiappone's (1984) study is not without problems. Over fifty separate t tests were conducted. Most of these were based on individual items of standardized measures, with only a minimal effort to distinguish clusters or patterns in the responses. With such a large number of individual calculations being performed, it is possible that a proportion of the significant differences were a result of chance error. A second, and perhaps more serious area of weakness in the study is in its structural composition. Although it is clear that Schlossberg's model of responses to transition is the underlying framework defining the investigation, the way in which this framework is utilized is somewhat confusing. Gender is the only one of Schlossberg's characteristics of the individual identified by the research questions as a predictor variable. Selfesteem, another characteristic of the individual, is used in the correlational analyses as a criterion variable. For these questions, Chiappone uses a selection of variables to predict scores on depression and duration of infertility, but the rationale for the particular selection of scores in each equation is not evident. Despite these cautions, Chiappone's (1984) study offers a comprehensive and inclusive explication of the different ways men and women may respond to the experience of infertility. While Chiappone's findings support the position that women tend to respond with greater evidence of symptomatology than do males, a careful review of her findings reveals that the differences, although significant, are not large. In general, Chiappone's results indicate that men and women tend to report similar feelings in response to infertility, and that women tend to exhibit greater intensity in both positive and negative responses. This pattern is consistent with socially prescribed gender roles which allow women more freedom in the expression of emotion than men. 38  Review of the Literature  While it appears clear that men may be more likely to report lower levels of distress and higher levels of adjustment than women, Chiappone's (1984) results show that men may also be more likely to employ coping strategies involving avoidance, denial and suppression of emotions. As noted by Murray and Callan (1988) "sex roles depicting males as stronger and less emotional may inhibit (men) from expressing or reporting their (feelings)" (p. 243). These authors suggest that an individual's sex role orientation, rather than their gender may in fact be the important mediating element in the observed differences between males and females in the levels of distress each group reports in response to events such as infertility. The following study directly address the influence of gender roles on an individual's responses to infertility. Adler and Boxley (1985) conducted a study examining the impact of sex-role identity for individuals experiencing infertility. These researchers recruited a large sample (N=164) of men and women, including 103 infertile individuals, 41 formerly infertile individuals and 20 people with no history of infertility, with each group being approximately half men and half women. Data collection involved the completion of standardized measures of sex-role orientation, marital adjustment, self-defeating behaviors, self-esteem, psychiatric symptoms and body-image. For each subject group, masculinity and femininity scores were used to predict each of the criterion 'coping' variables of self-esteem, body image, marital adjustment, psychiatric symptoms and selfdefeating behavior. For all three groups, masculinity for both males and females was strongly positively correlated with self-esteem. For infertile men and women, masculinity was also positively correlated with several aspects of body image. Femininity was negatively correlated with the body image subscale relating to build, as well as with the measure of self-defeating behavior. In addition, higher femininity scores tended to covary with higher levels of marital adjustment. In addition, femininity was positively related with the body-image subscale assessing facial-image in the group of formerly infertile patients. Subjects were identified as androgynous, masculine, feminine or undifferentiated in terms of sex-role and compared on criterion measures. For infertile subjects, those identified as  39  Review of the Literature  androgynous scored significantly higher on self-esteem measures than did those identified as feminine or undifferentiated. In addition, a significant difference was found for androgynous men and women on overall body-image, with individuals identified as undifferentiated scoring significantly lower than the other three sex-role groups. In the formerly infertile group, androgynous subjects scored significantly higher than feminine or undifferentiated subjects on self-esteem and significantly lower than these two groups on psychiatric symptoms. In addition, undifferentiated subjects scored significanity higher on selfdefeating behaviors than either the androgynous or the masculine subjects. No significant group differences were reported for fertile subjects. Comparisons were also made of the three fertility status groups, with no differences being found between groups on any of the coping variables. Finally, the authors compared males with females on each of the outcome variables, with no significant differences between sexes being found on any of the measures for either the infertile or the formerly infertile subjects. The single significant difference reported was on the levels of marital adjustment of fertile subjects, with fertile women reporting higher levels of marital adjustment than fertile men. In general, the study appears to be methodologically sound. It is apparent that Adler and Boxley (1985) have made attempts to include and control possible extraneous or confounding variables (the effects of sex, education and religiosity were statistically controlled in the regression analyses) and the number of separate equations performed is not unreasonable. There is, however, an area for concern, specifically regarding the authors' lack of thoroughness in reporting of results. In at least two cases, statistically significant relationships are referred to, but no descriptive data is provided. In other cases, the authors have chosen to include in their tables only those results which achieved statistically significant levels, preventing the reader from examining the data and drawing his or her own conclusions. While these omissions do not invalidate the results of the study, they do lead the reader to question why more complete results were not included. The results which are included indicate that, in general, men and women who identified  40  Review of the Literature  with masculine or androgynous characteristics tended to exhibit higher self-esteem, a more positive body-images and to report fewer psychiatric symptoms and self-defeating behaviors, than did those whose sex-role identify was undifferentiated. Individuals who identify with feminine sexrole traits also indicated a more positive evaluation of some aspects of body image and of their marital relationships. In describing both the positive and negative ways sex-roles may affect an individual's self-image, Adler and Boxley (1985) support the need to examine an individual's sexrole identity in gaining a more comprehensive understanding of the complexity of factors involved in shaping an individual's response to the stress of infertility. Although few researchers have looked exclusively at psychological factors associated with infertility in males, a recent research project by Kedem, Mikulincer, Nathanson and Bartoov (1990) represents a comprehensive effort to fill the void. Employing a two-part design, these researchers conducted a between-group comparison of infertile and fertile men on a number of psychological dimensions, as well as examining objective and subjective elements of the experience of infertility to predict individual differences on the psychological variables within the experimental group of infertile men. One hundred and seven infertile males were recruited through attendance at an Israeli university semen analysis clinic. The researchers collected demographic data as well as objective measures of infertility (duration, diagnosis and whether infertility was primary or secondary). Subjects completed questionnaires measuring the degree of hope for a successful outcome, the degree of stress posed by infertility as well as whether negative events were attributed to internal, stable & global factors or to external, unstable, specific factors. Subjects also completed a selfesteem questionnaire and a symptom checklist measuring six symptoms of stress, including sexual inadequacy. A control group of thirty men who had not sought treatment for infertility was recruited from occupational groups listed by the infertile men and carefully selected to match proportionally to the experimental group on the basis of the demographic variables. These men completed the measure of self-esteem and the symptom checklist, including the additional 'sexual inadequacy' 41  Review of the Literature  questions. In the comparison study, a few significant differences were found between infertile and fertile males. While infertile males tended to score lower on self-esteem than their fertile counterparts, they indicated higher scores on symptom scales of anxiety and somatization. No differences between groups were reported on the other four symptoms measured. In the second stage of the study, regressions were performed to predict the relationship of the objective, attributional and subjective elements of the men's infertility to each of the seven criterion measures (self-esteem and six symptoms). These analyses yielded two significant results: Men who perceived their infertility to be a global problem and who rated infertility as more stressful tended to score higher in depression. Also, men who perceived their infertility as global, who viewed the prognosis as hopeless and whose wives had also been diagnosed with fertility problems, tended to score higher on the measure of sexual inadequacy. Both the primary strengths and the weakness in this study lie in the selection of the control group of fertile men. The control group was painstakingly selected to provide an exact proportional representation of the demographics found in the infertile research sample, thereby significantly increasing the validity of the fmdings. This strategy also resulted in a fairly small comparison group, restricting the likelihood of discovering significant differences between the groups. Despite the small size of the comparison group, some important psychological differences were found between fertile and infertile men. The findings that infertile men report lowered selfesteem and increased levels of anxiety and somatic complaints is important. Unlike the majority of studies which compare infertile men to infertile women and interpret men's lower reported distress as lack of distress, this study compared men to men, thereby controlling for any gender effects and providing evidence that males also respond to the stress of infertility with feelings of depression, sexual inadequacy and decreased self-esteem. An exploratory study on the impact of infertility on men was conducted by Feuer (1983). In this unpublished research, Feuer tested 93 infertile men on six variables.Interviews were 42  Review of the Literature  conducted to assess levels of masculinity, locus of control and social isolation. Standardized questionnaires were used to measure depression, self-esteem and the quality of the marital relationship. Men representing two diagnostic conditions, oligospermia (low sperm count) and azoospermia (zero sperm count) were compared on the several variables. All of the infertile men reported significant levels of depression. Men diagnosed as oligospermic demonstrated a more severe impact, in levels of depression, quality of marital relationship, and social isolation than did men diagnosed with azoospermia. Feuer (1983) suggests that the feelings of uncertainty surrounding a diagnosis of oligospermia, as contrasted with the finality of an azoospermic diagnosis, may account for the increased levels of distress observed in these men. In interviews, each of the infertile men identified some aspect of his life which was negatively impacted by his infertility. Feuer (1983) also noted that men tended to reveal more evidence of negative impact in interviews than they reported on standardized measures, noting the men's 'tendency to deny' the impact of infertility. Like the previous study, Feuer (1983) focussed on men's perceptions and feelings regarding infertility. Both studies demonstrated that the distress experienced by infertile men is significant and real. Also highlighted by Feuer was men's apparent tendency to under-report their distress, suggesting that results indicating higher levels of distress for women than men (Bell, 1981; Chiappone, 1984) may be more a reflection of men's 'tendency to deny' than of actual differences. On the other hand, the distress reported by men may actually be different in quality and content than distress reported by women. For example, while both men and women have been shown to report elevated levels of anxiety, depression and social isolation (Bell, 1981; Chiappone, 1984; Feuer; Kedem et al. 1990; Sandelowski & Jones, 1986) only men have been reported to display somatization (Kedem et al.) or denial (Feuer, Chiappone; Woollett, 1985). It is this form of gender differences that is of interest in the present study. Aside from variables concerned with gender and gender-roles, a variety of other factors may be instrumental in mediating or influencing the impact of the stress of infertility. As already noted, diagnosis (Feuer, 1983) attributions (Kedem et al., 1990) as well as a sense of mastery and 43  Review of the Literature  self-esteem (Chiappone, 1984) have been shown to have a noticeable effect on shaping the individual's response to infertility. The following study examines the mediating impact of a factor which may be described as a characteristic of the transition, specifically, whether infertility is primary or secondary. Callan and Hennessey (1988) sought to determine the "life circumstances" (p,137) which influence a woman's psychological adjustment to infertility. Suggesting that motherhood (previous successful reproduction) may be an important mitigating factor in adjustment, these researchers compared 53 primary infertile women who were undergoing IVF treatment with 24 mothers who were seeking IVF treatment for infertility related to a subsequent pregnancy. Each subject was sent a questionnaire package, which elicited demographic data and information on fertility history, as well as assessments of personal happiness, including standardized measures of well-being, self-esteem, general affect and life satisfaction. Both groups of women scored within normal range on all measures of personal happiness. No differences were found between infertile mothers and infertile non-mothers in measures of selfesteem or on reports of positive or negative affect. Non-mothers did report lower satisfaction with life and less happiness over the past month and, on the semantic differential, rated life as less interesting and more lonely than did mothers. In comparison studies, the crucial factor which must be studied is the similarity of sample groups. The two groups are similar in demographic terms, but close inspection reveals that primary infertile women appear to have married and begun their attempts to have children at a somewhat later age than the secondary infertile women. Although these differences are minimal, they may in fact reflect differences between the groups in social or psychological factors such as sex-role identity, socio-economic status or career and occupational goals. As no data are reported in the study on any of these factors, these queries are not answerable. While the relevancy of the possible differences between the groups may be debated, they are on a minor scale. What is clear is that both groups of women were highly motivated to become parents and were actively seeking pregnancy at the time of the study. These researchers suggest 44  Review of the Literature  that women who choose to pursue the often lengthy and arduous procedures involved in IVF "are a select group among the infertile...(and)...less motivated and possibly less adjusted women are unwilling to bear these demands and do not pursue IVF treatments" (p.139). As women who do not pursue IVF treatments were not included in this study, this conclusion remains conjecture. However, the exclusive use of women seeking IVF treatments does impede the ability to generalize these authors' findings to other groups of infertile women or to men. The major contribution of the study are the scores which suggest that these non-mothers perceive their lives as less satisfying, less interesting and more lonely than did mothers, suggesting that a characteristic of infertility, in this case, whether it is primary or secondary, may have an impact on some women's response to the experience. The following study utilizes a longitudinal approach to explore the impact of a number of important factors related to the transition, including time spent dealing with infertility, stage of the medical investigation and diagnosis received as well as the gender of the individual. Daniluk (1988) conducted a comprehensive longitudinal study of several factors associated with the psychological impact of infertility. Forty-four couples undergoing medical examinations for primary infertility were recruited from patients attending a university clinic specializing in the treatment of infertility. Subjects were surveyed on four occasions; immediately following their initial medical visit, four weeks later, one week after diagnosis and six weeks following diagnosis. Four questions defined the scope of the study, which examined whether changes occurred on measures of psychological distress, marital adjustment, relationship quality, sexual satisfaction and perceived need for counselling during the course of the medical investigation for infertility; at what point or points during the investigation these changes occurred; the nature of the changes; and whether there were differences in the changes experienced by men and women. No significant differences across testing sessions were found on measures of sexual satisfaction, marital adjustment or relationship change on the basis of gender, diagnosis, identified source or time spent trying to conceive. Significantly higher levels of psychological distress were reported for both men and women at the initial medical interview, prior to any testing or diagnosis, 45  Review of the Literature  than at subsequent testings. Significant differences were reported between male and female participants on measures of marital adjustment at all four testing sessions, with females consistently scoring higher than males. Differences in sexual satisfaction scores were reported by men and women receiving a neutral (unexplained) diagnosis as compared with those receiving a positive (treatable) or negative (untreatable) diagnosis. Individuals who received the diagnosis of unexplained infertility reported a higher level of sexual dissatisfaction than men and women with a definite diagnoses. This finding supports the contention that uncertainty is among the most stressful elements of the infertility experience. The primary strength of this study lies in its longitudinal approach. By examining the experience of infertility from the perpective of four distinct phases or stages, this researcher emphasizes the complex and long-term nature of the transition to childlessness. The study also demonstrates that characteristics of the transition such as duration and diagnosis may be important in shaping an individual's response to infertility. Although no differences were reported on the dependent measures on the basis of time spent dealing with infertility, the finding that distress levels peaked at the initial interview suggests that distress may decrease as men and women proceed through the infertility investigation. In addition, some evidence of increased distress was shown for individuals whose diagnosis was unknown, suggesting that this characteristic of the transition may be important in shaping the individual's response to infertility. Finally, Daniluk (1988) demonstrates that, although both men and women experienced elevated distress in response to infertility, women tended to maintain higher levels of marital satisfaction than men, suggesting that, in terms of their relationships, women displayed higher levels of adjustment than men. The implication of this finding is that the mediating influence of sex, perhaps the most basic characteristic of the individual, may be much more complex than a matter of more or less distress reported in response to infertility. Sabatelli, Meth and Gavazzi (1988) studied the mediating effects of several characteristics of the individual as well as some characteristics of the infertility on married couples' adjustment to 46  Review of the Literature  involuntary childlessness. A sample recruited from Resolve membership lists consisted of 52 women and 29 men (twenty-nine couples plus 23 women) who were at various stages throughout the infertility investigation. The study included two major questions. The first question was exploratory and sought to quantify the changes attributed by an individual to infertility. The second question, which comprised the chief investigation, addressed a number of predictor variables or mediators of adjustment to infertility. The researchers used measures of self-esteem, economic independence, marital locus of control, marital commitment and satisfaction, past coping strategies used as well as the duration of infertility to predict subjects' scores on measures of depression, anxiety and adjustment to infertility. On the first question, exploring the changes attributed to the infertility experience, a number of interesting results appeared. In general, women reported changes almost twice as often as did men. Women reported decreases in self-confidence, frequency of contact with friends and comfort with friends, but also reported increases in emotional support from spouse as well as in conflicts with spouse. The greatest changes were reported in the area of sexuality, with over half of the women reporting decreases in satisfaction, frequency and comfort with sex. Men also reported decreases in this area, with about half reporting decreases in frequency and satisfaction with sex. Thus the majority of changes attributed to the infertility experience were negative. The chief investigation of the study revealed five predictor variables which were moderately to strongly related to coping efficacy. Men and women who reported high self-esteem were very likely to indicate high levels of adjustment to infertility and low depression and anxiety. Women who reported a strong marital commitment and an internal locus of control were likely to report high adjustment to infertility and low depression and anxiety. Men who reported a high marital commitment and an internal marital locus of control also showed a moderate tendency score higher in adjustment to infertility and lower in depression. Men and women who indicated high marital satisfaction also scored somewhat higher in adjustment to infertility and somewhat lower in depression. For both men and women, the longer they had been experiencing infertility, the less anxiety was reported. For men, longer duration was  47  Review of the Literature  also related positively to adjustment and negatively to depression. Thus, for both men and women, self-esteem, marital commitment, internal locus of control, marital satisfaction as well as more time spent dealing with infertility appeared to be important predictors of coping efficacy in response to infertility. Further analysis revealed that, for women, self-esteem and commitment to marriage accounted for the majority of the variance on all three measures of coping efficacy. Coping scores for the twenty-nine couples were compared and although couples' adjustment to infertility tended to covary, wives reported lower adjustment and higher anxiety than their husbands. Two coping strategies showed moderate to strong correlations with coping efficacy. Women and men who reported having reframed or redefined past crisis situations indicated high positive adjustment to infertility. For women, reframing was also strongly related to low reported levels of anxiety and depression. 'Spirituality' and 'social networking' strategies for women and a 'help seeking' strategy for men were mildly to moderately related to lower depression scores. However, men and women who reported using passive coping strategies, including avoidance and denial, indicated a strong tendency to display poor adjustment to infertility and a moderate to strong tendency to indicate high levels of anxiety and depression. Thus, the tendency to approach events passively appeared to be less effective in mitigating the negative impact of transitional crises such as infertility than was the tendency to reframe or redefine situations into more manageable terms. Although this study offers a number of interesting fmdings, there are some important concerns which must be addressed. Although it is clear that the investigators sought to study involuntary childlessness, it is not made clear whether subjects were limited to those with primary infertility. Further, as the sample is drawn from a population of Resolve members, it may not be representative of the majority of infertile couples, but may be skewed toward a sub-population with a pro-active approach to coping, suggesting higher levels of adjustment. As coping strategies and adjustment levels were major elements of the investigation, this may be an important consideration in using the results of the study for predictive purposes. In addition, as Chiappone (1984) found, women are more likely to seek support through  48  Review of the Literature  sources external to the marriage than are men. It is not surprising therefore, almost twice as many women than men volunteered to participate. The relatively small size of the male sample (29), therefore makes generalization of the results for men, tentative. With these caveats in mind, a number of interesting results may nonetheless be extracted from the study. Sabatelli and his colleagues (1988) demonstrated that a number of personal or attitudinal characteristics of the individual, including high self-esteem, an internal locus of control regarding their marriage and a high degree of marital commitment tended to have a positive impact on adjustment to infertility and helped to minimize feelings of depression and anxiety. In addition, an important characteristic of the transition, time spent dealing with infertility also appeared to contribute to increased adjustment and decreased distress. Finally, the finding that the use of reframing as a coping strategy in past crises predicted better adjustment to infertility supports the contention that, as Schlossberg (1981) noted, "the individual who has successfully weathered a transition in the past will probably be successful at adapting to another transition of a similar nature" (p.15). The mediating effect of a number of demographic, medical, cognitive and environmental factors on men's and women's adjustment to infertility was investigated in a complex and comprehensive study by McEwan, Costello and Taylor (1987). Sixty-two women and forty-five men (N=107) were recruited from among the patients at an Infertility clinic. Data for predictor variables was collected through medical files and interviews. Information from medical files included time trying to conceive, time seeking medical support, diagnosis, infertility status (primary or secondary), medication prescribed, and the medical estimate of probability of conception, as well as demographic information such as age, sex, religion and socioeconomic status. Interviews were directed to elicit information on stressful events experienced within the past year, perceptions of the availability of social support, cause of infertility, perceived chances for success, feelings about infertility and the acceptability of alternatives to conception. Standardized questionnaires were used to measure the dependent variables of emotional disturbance, perceived social role performance, sexual satisfaction, quality of sexual relations and 49  Review of the Literature  drinking patterns. Initially, a comparison of scores for males and females on the measure of emotional disturbance revealed that significantly more women than men (40% of women and 13% of men) reported serious symptoms of distress. When compared with the expected rates of distress found in the general population, however, (37% of women and 1% of men), the infertile males reported proportionately greater distress levels. This suggests, as do Chiappone's (1984) results, that although women tend to report greater levels of distress than men, many men do experience serious levels of distress in response to infertility. Dependent measures were condensed into composite factors of related elements of adjustment, which were then used in regression equations with sets of predictor variables. For women the strongest predictors of the first composite, emotional and social adjustment, were religion and increased age, with recent life events and a confiding relationship with their spouse also showing a postive correlation with positive adjustment. In addition, women who made selfattributions regarding infertility and those who had not received a diagnosis tended to show significantly more distress. Two other composites, sexual adjustment and alcohol consumption were identified, but no significant relationships with predictor variables were found. For men, no significant relationships were reported between any of the predictor variables and the three composites factors of sexual adjustment, emotional and social adjustment and alcohol consumption combined with family unit relationships. A final analysis was performed to test the relationship of distress levels to women's tendency to accurately predict their chances of successful conception. Although both distressed and non-distressed women were aware of the official medical prognosis, the results suggest that distressed women tended to underestimate and non-distressed women tended to overestimate the probability of success. This complex and sophisticated study provides several useful fmdings and presents only minor areas for concern, methodologically. Data from a relatively small number of subjects (62 women and 45 men) is used to answer a large number of questions. In recognition of this, results 50  Review of the Literature  for male subjects were reported in a limited fashion, leaving the burden of answers to be provided by female subjects, who outnumber males by a ratio of only 1.4 to 1. McEwan et al. (1987) suggested that women reported higher levels of distress than did men in response to infertility and provided evidence of several factors with a mediating impact on infertile women's adjustment levels. Those individuals who appeared to be at highest risk of suffering serious emotional distress in relation to infertility were young, non-Protestant women who blamed themselves for their infertility, who had no previous experience with stressful events (characteristics of the individual) who did not feel able to confide in their husbands (characteristics of their environment) and who had not yet received a diagnosis and were therefore still facing a high degree of uncertainty (characteristics of the transition). This study suggests that characteristics of the individual, their environment and the transition may all be important in shaping the complex nature of a woman's response to infertility. The following study explored the relationship between a number of psychological and personality factors for infertile women. Fouad and Fahje (1989) recruited 61 infertile women through infertility support groups and the offices of specialists and administered measures of selfesteem, depression, locus of control and perceived social support. Information on time dealing with infertility and whether infertility was primary or secondary was also collected. When compared with standardized norms provided with the instruments, the infertile women studied indicated average levels of self-esteem, a slight tendency to be internally controlled, mild levels of depression, a higher than average need for support, less than average support received and a moderate level of satisfaction with the support received. These results suggest that, in general, these women were somewhat distressed, but appeared nonetheless to be fairly well adjusted. A number of significant relationships were found between the four measures. High levels of depression were negatively related to internal locus of control and positively related to greater perceived needs for social support and lower levels of satisfaction with support received. These results suggest that women who perceived their lives as controlled by external forces and who  51  Review of the Literature  perceived the support they received as inadeqate were at greater risk of experiencing feelings of depression in response to infertility. High self-esteem was positively related to internal locus of control, high levels of satisfaction with support received and to a minimal discrepancy between support required and support received, suggesting that the more comfortable an infertile woman was with herself, the less support she required, the happier she was with the support received and the more likely she was to perceive events and consequences as within her own control. In addition, although no hypothesis addressed this relationship, a strong negative correlation between self-esteem and depression was found, indicating that women who indicated high levels of self-esteem also tended to report low levels of depression. A critical review of Fouad and Fahje's (1989) investigation reveals a fairly well-constructed correlational study. The primary concerns regard the possible influence of the duration of infertility as well as the individual's infertility status. Although this information was apparently collected, it is not clear how it was used, if at all. With this caution noted, however, the study does highlight several interesting relationships. The finding that infertile women with high self-esteem and high internal locus of control were less likely to report being depressed is suggestive that these personality factors may act as mediators in reducing the negative impact of infertility. Further,results suggesting that women who perceive that available social support is adequate to meet their needs and who are satisfied with the social support received, are less likely to exhibit signs of depression, add strength to Schlossberg's (1981) contention that effective social support may provide some protection from the emotional impact of infertility. Obversely, these researchers found that women with low self-esteem reported that their support needs were being inadequately met by the available resources. These women and those with an external locus of control also indicated high levels of depression. These results suggest that women with low self-esteem and a limited sense of control of their lives are at high risk for feelings of isolation and depression in response to infertility. Each of these results suggests that 52  Review of the Literature  self-esteem and locus of control may be important characteristics of the individual to consider in an examination of the factors shaping an individual's reponse to infertility. Summary of Findings  The studies reviewed demonstrate that a variety of factors appear to be useful in gaining an understanding of the nature and extent of the impact of infertility on an individual. Included in this review were characteristics of the individual, characteristics of the environment as well as characteristics of the transition. Of the characteristics of the infertility transition investigated, two were shown to mediate its impact. The duration of infertility, often measured as the length of time since discovery of infertility, appeared to be related to lower distress, such that the more time an individual spent living with infertility, the better able he or she was to adjust to the transition (Daniluk, 1988; Sabatelli, 1988) and, for women, the more coping strategies used (Chiappone, 1984). On the other hand, Chiappone also found that longer time spent dealing with infertility was related to decreases in self-esteem. While these results appear to be contradictory, they may not be. Although men and women may gradually develop effective strategies for coping with the ongoing distress engendered by infertility, the individual's perceptions of their own effectiveness over time, particularly if no 'cure' is found, may suffer. Diagnosis also appeared to play a role in mediating the impact of infertility as individuals who had received a specific diagnosis tended to exhibit less distress than those whose diagnosis remained uncertain (Daniluk, 1988). Similarly, Feuer (1983) found that men who received the unequivocal diagnosis of azoospermia demonstrated less distress than did men whose diagnosis of oligospermia left an element of uncertainty and possible hope. It appears that, as long as diagnoses or the efficacy of prescribed treatment remains uncertain, infertile men and women remain in a state of anxiety and distress. It may be that, once a conclusive diagnosis is made, even if the implication is that no treatment is available, infertile couples and individuals may begin to move from a state of being in transition, into a process of adjustment or adaptation. Several characteristics of the person's social environment were also investigated. Among 53  Review of the Literature  those factors which appeared to facilitate better adjustment to infertility were the ability to confide in one's spouse (McEwan et al., 1987) and the 'perceived fit' between needs for support and support received (Fouad & Fahje, 1989). These results suggest that, as Sclossberg (1981) noted, effective social support may be an important factor in helping an individual to cope with the changes engendered by experiences such as infertility. The largest group of factors investigated as possible mediators of the impact of infertility were a collection of characteristics of the individual. The most often investigated personality characteristic is self-esteem. High self-esteem was repeatedly indicated as the strongest correlate of better adjustment and lower levels of distress for both men and women (Chiappone, 1984; Fouad & Fahje, 1989; Sabatelli et al., 1988). In a similar way, internal locus of control (Fouad & Fahje, 1989; Sabatelli et al.), a sense of mastery (Chiappone, 1984) and a proactive approach to coping (Chiappone) have also been shown to be related to higher levels of adjustment and decreased distress in response to infertility, while a tendency to blame oneself for the infertility was related to poorer adjustment (McEwan et al., 1987). These findings suggest that an individual's sense of self-worth and efficacy, or 'psychosocial competence' (Schlossberg 1981) may be among the most important elements in shaping his or her response to the experience of inferility. An interesting pattern of findings emerged regarding the nature of sex differences in responses to infertility. While Chiappone (1984), Daniluk (1988) and Bell (1981) reported differences between men and women in the intensity of responses, with women consistently reporting more anxiety, depression and other elements of distress than men, as well as greater marital adjustment (Daniluk). Chiappone's results suggest that women tend to report both positive and negative feelings with greater intensity and frequency than men. These results, as well as those of Kedem et al. (1990) and Feuer (1983) indicate that men experience significant levels of distress in response to infertility and that men's and women's responses may actually differ in quality as well as intensity, with men appearing more likely to exhibit evidence of somatization (Kedem et al.) and denial (Feuer; Chiappone; Woollett, 1985), and women appearing more likely to exhibit depression (Chiappone; Woollett) and to utilize a variety of coping strategies  54  Review of the Literature  (Chiappone). It appears that sex-role identity may also play a role in the observed differences between men and women, as men and women who identify with androgynous or masculine sexrole characteristics tended to demonstrate higher levels of coping and lower levels of distress (Adler & Boxley, 1985). A variety of other characteristics of the individual were also indicated as having a possible impact on levels of subjective distress and adjustment for some people. Religious beliefs appeared to have some impact on adjustment, such that women who described themselves as Protestant indicated higher adjustment and less distress than women who identified with other religious groups (McEwan et al., 1987). In addition, a strong sense of commitment to the marital relationship was shown to be related to better adjustment (Sabatelli et al., 1988). Finally, age appeared to be related to adjustment such that older women exhibited less distress than younger women (McEwan et al.). These last two findings appear to suggest the possibiiliy of a relationship between maturity or stability and the ability to cope effectively with the experience of infertility.  Hypotheses Research Hypotheses Null Hypothesis # 1: There will be no statistically significant canonical correlation at the  a  = 0.05 level on the mean scores on a weighted linear composite of scores for the dependent variables of Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Psychoticism and Adjustment to Infertility for the Independent variables of Diagnosis, Duration, Perception of Outcome and Perception of Success and the joint effects of these factors.  Null Hypothesis # 2: There will be no statistically significant canonical correlation at the = 0.05 level on the mean scores on a weighted linear composite of scores for the dependent variables of Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Psychoticism and Adjustment to Infertility for the Independent variables of Self Esteem, Locus of Control, Sex-Role, General Health, Age, Sex, Religion, Race, Socio-economic  55  Review of the Literature  Status and Previous Experience and the joint effects of these factors. Alternative Hypotheses  In the event that the either of the null hypotheses are unsupported by the data, the following alternative hypotheses have been generated for testing: Alternative Hypothesis #1: The canonical correlation between the dependent variables of  Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Psychoticism and Adjustment to Infertility and the independent variables of Diagnosis, Duration, Perception of Outcome and Perception of Success will be significant at the  a = 0.05  level. Alternative Hypothesis #2: The canonical correlation between the dependent variables of  Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Psychoticism and Adjustment to Infertility and the independent variables of Self-Esteem, Locus of Control, Sex-Role, General Health, Age, Sex, Religion, Race, Socio-economic Status and Previous Experience will be significant at the  a = 0.05 level.  56  Methodology  CHAPTER THREE Methodology The Sample  A sample of 106 subjects was recruited from the population of couples attending the Infertility clinic in the Obstetrics and Gynaecology Department of Foothills Hospital at the University of Calgary in Alberta. Individuals and couples interviewed included those in all stages of the infertility transition, from those making their initial visit to the clinic, to those whose fertility investigations were ongoing for several months or years. Eligible participants were men and women who were members of couples experiencing primary infertility, defined as never having carried a pregnancy to full term, with no known history of psychiatric illness or use of moodaltering drugs. Demographic information (Characteristics of the Individual) for all participants was collected through the Personal Information Survey. A summary of the results is shown in Table 1. In total, seventy-one (71) of the 106 volunteers who agreed to participate in the study returned completed questionnaire packages, representing a 67% response rate. Of these 43 (61%) were female and 28 (39%) were male. Men ranged in age from 27 to 48 with a mean age of 33. Women ranged in age from 23 to 40, with an average age of 31 years. Seventy-three percent of respondents indicated family income levels at or above $40,000 per year and 92% reported incomes of $25000 or greater. Education was measured by level achieved, with 34% reporting completion of high school and 65% reporting the completion of two or more years of college or university. Most subjects (99%) reported their general physical health as good (category 3) or excellent (cat. 4). Race or ethnic origin categories were derived from a logical grouping of those ethnic groups identified by respondents in an open-ended question. The great majority (84%) identified themselves as White, Canadian, or of British or Northern European origin (cat. 1); 4% reported being French Canadian (cat. 2); 6% reported being Asian or Chinese (cat. 3); with the remaining 6% (Other (cat. 4)) identifying themselves as Native, Jewish or Latin American. An auxiliary 57  Methodology  question was included with this item, asking whether having children is regarded as an important cultural or family value. Ninety-three percent (93%) indicated 'Yes' in response to this item. Respondents also completed an open-ended question regarding their religious affiliation. The largest single group (51%) reported no religious affiliation, 20%identified themselves as Catholic, 13% indicated affiliation with Protestant churches, including United, Lutheran and Anglican, another 14% identified themselves with traditional Christian groups, including the Mennonite Church or Church of God, and 3% indicated other religions such as Judaism or Muslim. Characteristics Relating to Infertility  A number of questions relevant to the transition were also included on the Personal Information Survey (See Table 2). The first related to the 'duration' of the transition, measured in months. Subjects reported that they had been attempting to conceive for between 12 months and 144 months, with an average of approximately 48 months, or four years. In addition, subjects reported a range of 5 months to 132 months ( X= 28.16 ± 27.6, or approximately two and onehalf years) since first seeking medical attention for infertility. Subjects were also asked to identify their diagnosis received, by category. The largest group (41%) identified their infertility as 'Female Factor', with only 10% 'Male Factor' and 4% identified as 'Couple Factor'. Almost half of the sample had no clear diagnosis, however, with 21% reporting an 'Unknown factor' diagnosis and 24% having received no diagnosis. On an item asking respondents to describe their perception of their infertility status, 55% reported being 'very uncertain', 25% perceived infertility as 'a temporary condition' and 18% identified it as 'a permanent condition'. Subjects were also asked to assess their chances of achieving a successful pregnancy. A majority (62%) assessed their chances of success as less than 50%, with 37% reporting a less than 25% chance of success and another 25% of respondents reporting between 25% and 50% chance of success. Twenty-five percent of respondents thought their chances of success were between 50% and 75% and 15% of the sample believed that their chances were better than 75%. 58  Methodology  Table 1:	 P.I.S. Demographic Characteristics: Responders v Responders Variable Category 30.71 ± 4.59 Age Females 32.68 ± 5.24 Males 61% (1)Female: Sex (2)Male: 39% (1)White/Anglo: 84% Race (2)French Canadian: 4% /Ethnic Background 6% (3)Asian: 6% (4)Other: (1)< $15000/Yr 3% Income 6% (2)$15-25000/Yr 18% (3)$25-40000/Yr (4)> $40000/Yr 73% 1% Education Completed (1)Elementary 34% (2)High School (3)College or Univ. 54% 11% (4)Post Graduate General Health (1 - 4) 1=Poor, 4=Excellent 3.58 ± .5 20% Religion (1)Catholic: 13% (2)Protestant: (3)Christian (Other): 14% 3% (4)Other: 51% (5) None: 93% (1)Yes Children Valued 7% (2)No:  59  Non-Responders T Stat NonResp 31.06 ± 3.55 -.31 32.93 ± 3.79 -.18 -.72 49% 51% 1.08 89% 6% 6% 0% -.63 0% 6% 23% 69% .08 3% 37% 51% 9% 3.45 ± .5 1.16 26% 23% 1.81 11% 6% 34% -1.34 83% 17%  P Value .76 .86 .48 .28  .53  .93  .25  .08  .19  Methodology  Table 2: P.I.S. Characteristics of the Transition Responders v Non-Responders Variable  Category  Responders NonResp  Time Trying No. of Months	 47.77 ± 28.9 Time Seeking Medical Help No. of Months	 28.16 ± 27.6 (1) Female Factor: 41% Diagnosis (2)Male Factor:	 10% (3)Couple Factor: 4% (4)Unknown: 21% (5)No Diagnosis: 24% (1) Uncertain: 55% Perception of Outcome (2)Temporary:	 25% (3)Permanent: 18% (1) < 25%: 37% Perceived Chance (2) 25 to 50%: 25% of Success (Hope) (3)50 to 75%: 25% (4)> 75%: 15% (1) Yes: 45% Previous Experience (2) No: 55%  T Stat  46.23 ± 23.9 .28 32.81 ± 19.3 -.97 48% 34% 3% -1.39 11% 6% 34% 34% -2.45 31% 34% -.17 29% 20% 17% 63% 1.71 37%  P Value .78 .33  .17  **.02 .87  .09  About half (45%) of the respondents reported being faced with similar life experiences in the past. More than half of those who responded positively to this question identified experiences which they found similar to infertility. These included such experiences as their own or a family member's serious illness, the death of family members, serious financial loss, divorce, abortion as well as multiple miscarriages and previous failed pregnancies. Comparison of Respondents to Non-Respondents  All volunteers recruited through the medical clinic completed the Personal Information Survey, which was collected by the researcher immediately. Therefore, it was possible to compare the characteristics of those who completed and returned the questionnaire packages with those volunteers who did not. Using BMDP3D (Sockne & Forsythe, 1988), a series of Hotelling ttests were performed to compare respondents with non-respondents on each item of the Personal  60  Methodology  Information Survey. Cases with missing scores were excluded from these calculations. As can be seen in Tables 1 and 2, no significant differences were found between respondents and non-respondents on age, sex, income, education, general health, race, religion, duration of infertility, diagnosis, hope or previous relevant experience. A significant difference (p < .05) was shown between respondents and non-respondents on perceptions of the outcome of their infertility, with respondents being more likely to report their infertility as uncertain (55%) than as temporary or permanent and non-respondents tending to be evenly split in their perceptions of the outcome. Instrumentation  The response package completed by each subject included three instruments measuring predictor variables: the BEM Sex-Role Inventory (Bem, 1974), the Coopersmith Self-Esteem Inventory (Coopersmith, 1981) and Rotter's Personal Belief Questionnaire (Rotter, 1966). Two criterion measures were also included: the SCL-90-R (Derogatis, 1975) and the Adjustment to Infertility Scale (Sabatelli, Meth & Gavazzi, 1988). Personal Information Survey (PIS)  The Personal Information Survey (Appendix A) was designed by the researcher to elicit demographic, personal and diagnostic information of specific relevance to the present study. The questionnaire consisted of 14 items and required subjects to answer fact-based questions by selecting from an array of alternatives or by giving a brief written response to open-ended items. Subjects took 5 to10 minutes to complete the survey. Demographic data elicited by the P.I.S. included gender, age, ethnic background, socioeconomic status (yearly income and level of education attained) and religion. These factors have been identified by Schlossberg (1981) as characteristics of the individual which may be important in assessing the impact of negotiating a psychosocial transition such as infertility. In addition, the P.I.S. included questions regarding the subject's general health and previous experience with transitions of a similar nature. These were among the characteristics of the individual identified by Schlossberg (1981) as of potential importance in adapting to a 61  Methodology  transition. Recognizing that the 'similar' previous experiences may include a limitless number of events, and that the degree of similarity is subjective, the question addressing previous experience was left open ended and included an invitation to briefly describe the experience. There were no predetermined categories of similar transitions, nor of their impact. The only information used in statistical analysis was the Yes/No response. Six questions addressed subject-specific characteristics of the infertility experience. Duration of infertility was measured as the length of time (in months) each subject had been attempting to conceive as well as the length of time since each subject originally sought medical treatment for infertility. In addition, the individual's subjective perception of whether their infertility was permanent, temporary or of uncertain duration was assessed, as well as the subject's perceptions of the chances (in percent) of achieving a successful pregnancy. Finally, subjects were asked to identify whether a diagnosis had been made, and if so, whether it indicated a male factor, a female factor or a couple factor as the primary cause of infertility, or if the cause was as yet unknown.  Bem Sex-Role Inventory (BSRD The Bem Sex-Role Inventory -Short Form (Bem, 1974) was used to to assess the individual's degree of identification with 'traditional' definitions of masculinity and femininity. The BSRI is a measure of sex-role identity, a psychological or attitudinal construct relating to the way in which individuals perceive themselves relative to social definitions of 'Masculinity' and 'Femininity'. Sex-role identity has been identified as an important characteristic of the individual, relevant to the assessment of adaptation to developmental transitions (Schlossberg, 1981). Sexrole identity is of particular relevance to the transition instigated by infertility, as many people equate the ability to have children with their 'sexual identity' (Hendricks, 1985; Humphrey, 1977; MacNab, 1985; Mahlstedt, 1985). The BSRI-Short Form is a self-report scale consisting of thirty adjectives which have been rated as 'masculine', 'feminine' or 'neutral' personality characteristics by a panel of college student judges. Responses are recorded on a 7-point Likert-type scale indicating to what extent the subject  62  Methodology  believes that each adjective describes themselves. Responses range from 'always or almost always true' to 'never or almost never true'. Average length of time to complete is ten minutes or less (Bern). 'Masculine' and 'feminine' identity scores were achieved through calculating the arithmetic average of the respondents self-ratings on items from the masculine scale and the feminine scale. Subjects with scores above a cutoff point for either scale were categorized accordingly. Respondents with scores above the cutoff score on both scales were described as 'Androgynous'. Respondents whose scores on both scales were below the cutoff score were categorized as 'Undifferentiated'. In the present study, subjects were categorized on the basis of their scores and each category was coded as follows: androgynous = 1, masculine = 2, feminine = 3 and undifferentiated = 4. Reliability and validity of the BSRI have been demonstrated in a series of studies by the test author, as well as others. Reliability coefficients on test-retest trials over a four-week span on the original normative sample of the BSRI ranged from .75 for males to .94 for females (Bern, 1974). Internal consistency coefficients ranged from .75 for females to .90 for males (Bern). Discriminant validity was reported in the BSRI manual (Bern) with coefficients ranging from -.14 to +.33 between Masculinity and Femininity scales. Construct validity of the BSRI has been demonstrated through a number of experiments in which subjects behavioral responses to a variety of tasks were correctly predicted based upon BSRI scores. Specifically, subjects scoring high in 'Androgyny' tended to respond with behaviors not typical of their own sex, in both clearly defined and in ambiguous tasks, more often than did subjects who scored high on either the Masculinity or Femininity scales (Bern, 1974b; Bern, 1976). Normative data was provided in the BSRI manual (Bem, 1974) for a range of age and ethnicity groups, including adult men and women of childbearing age. The BSRI is perhaps the most widely used measure of sex-role identity. It has clearly been well researched and tested and is easy to administer, with clear directions designed for use by individuals from secondary-school age through older adulthood. Conceptually, the BSRI rejects  63  Methodology  earlier notions of masculinity and femininity as bipolar opposites. Consistent with current sociocultural definitions of sex-roles, Bem (1974) was the first to develop an instrument adopting a two-dimensional definition of masculinity and femininity which recognizes that a single individual can endorse both masculine-valued and feminine-valued characteristics (Kelly & Worrell, 1977; Spence, Helmreich & Stapp, 1975). For individuals coping with infertility, the relative tendency to endorse characteristics valued as masculine or feminine may be related to the individual's pattern of responses to the experience. Coopersmith Self-Esteem Inventory (SEI) Adult Form The Coopersmith Self-Esteem Inventory (Coopersmith, 1981) was one of the two measures used to assess aspects of an individual's 'Psychosocial Competence' (Tyler, 1978). Included by Schlossberg (1981) as a primary trait affecting an individual's adaptation to a transition, 'psychosocial competence' refers to a collection of corresponding attitudes toward self and the world, which provide an individual with a sense of competence and confidence. Although Tyler (1978) included a variety of self-attitudes within the construct of psychosocial competence, many of these may be understood as elements of the principal attitudes of self-esteem and a sense of personal competence. Coopersmith's SEI (1981) was utilized in the current study to measure the degree of selfesteem or positive self-evaluation, held by each subject. The SEI is a brief self-report inventory consisting of 25 self-statements and taking typically 10 minutes or less to complete. Eight selfstatements reflect a positive appraisal, (ie. "My family understands me"), with the remaining 17 indicating a negative self-appraisal (ie. "I give in very easily"). Respondents indicate for each statement, whether it described how he or she usually feels, by marking one of two boxes, labelled 'Like Me' and 'Unlike Me'. Positive statements marked 'Like Me' and negative statements marked 'Unlike Me' are scored as 'correct'. Raw scores are multiplied by four to obtain the Total Self Score. High scores indicate high self-esteem. Numerous studies concerning various forms of validity and reliability of the School Form are reported in the SEI Manual (Coopersmith, 1981), with reliability coefficients ranging from .80 64  Methodology  to .92. While the majority of studies cite validations specifically relevant to children and school performance criteria, a handful of validation studies of the school form provide useful information relating to adult functioning. Coopersmith (1967) demonstrated that SEI scores tend to correlate positively with measures of creativity, resistance to group pressures, perceived reciprocal liking and willingness to express unpopular opinions. In addition, Matteson (1974), found SEI scores to be positively correlated with measures of effective communication between family members and family adjustment. The Coopersmith Adult Form is essentially identical to the school form, with several items reworded slightly to reflect age-appropriate social and work situations. This form was normed on a group of over 200 college and university volunteers, aged 16 to 34. Norms reported for subjects aged 20 to 34 (N=148) include the mean score (g=66.7± 18.8) and Cronbach alpha (a..81). Minor gender differences were reported, with females scoring slightly higher on average (p.=71.6) than males (1.58.4). No other reliability or validity data is available for the SEI Adult Form.  Rotter Personal Beliefs Ouestionnaire (I-E Scale) The Rotter Personal Beliefs Questionnaire (Rotter, 1966) is a self-report inventory designed to assess the extent to which an individual perceives the locus of control of events as internal or external. The questionnaire consists of twenty-nine pairs of statements which reflect commonly held beliefs about personal causality and responsibility. For example, item #15 reads: a) "In my case getting what I want has little or nothing to do with luck", reflecting an internal sense of control; and b) "Many times we might just as well decide what to do by flipping a coin", reflecting a belief in externally controlled events. Twenty-three statement pairs contribute to the score, with the remaining six included as 'filler' items. Respondents are required to select the one statement from each pair which they most strongly believe to be true. The order of 'internal' statements and 'external' statements are randomly alternated. Scores are calculated by a simple tally of the number of 'internal' statements. Higher scores reflect greater internal locus of control. The average time required to complete the inventory is between five and ten minutes. Reliability and validity of the I-E Scale have been demonstrated in a series of studies by the 65  Me thodolo gy  test author, as well as others (Rotter, 1966). Reliability coefficients on test-retest trials over a onemonth span ranged from .60 to .78 for males and .83 for females (Rotter). Retest scores after two months correlated .55 with original test scores, for men and women combined. Internal consistency coefficients ranged from .65 to .79 (Rotter). Item validity for the I-E Scale was established through repeated trials and refinements, in which 77 of the 100 original test items were eliminated on the basis of high social desirability scores or their lack of discriminant value. Construct validity for the remaining 23 items was demonstrated in a study in which medical patients were compared on the basis of observed passivity versus self-effort toward recovery (Seeman & Evans; as reported in Rotter, 1966). Since its publication in 1966, Rotter's instrument has been used widely in a range of experimental conditions to measure the cognitive dimension of locus of control. In the present study, Rotter's instrument is used to measure an aspect of 'psychosocial competence' (Tyler., 1978), a cognitive-emotional contruct consisting of an individual's sense of self-worth and selfefficacy. Derogatis SCL-90-R  The Derogatis SCL-90-R (Derogatis, 1975) is a widely used inventory designed to measure symptoms of psychological distress. In the present study, the SCL-90-R will be used as a dependent measure, to gauge the level of emotional and physical distress felt by subjects experiencing infertility. The SCL-90-R consists of 90 indicators of psychological distress. Respondents indicate the extent to which they have experienced each over the past week. Most respondents complete the survey in approximately 15 minutes (Derogatis, 1975). Scale range is from 0 = 'not at all' to 4 = 'extremely'. The scoring of the SCL-90-R produces nine subscales corresponding with distress symptoms often reported by infertile couples, as well as three general distress indices. The symptoms identified are a) Somatization (SOM), b) Obsessive-Compulsive (OBS), c) Interpersonal Sensitivity (INT), d) Depression (DEP), e) Anxiety (ANX), f) Hostility (HOS), g) Phobic Anxiety (PHOB), h) Paranoid Ideation (PAR), and i) Psychoticism (PSY). The three 66  Methodology  general indices are derived through combinations and summations of the symptom scores, and include; a) The General Severity Index (GSI), reflecting both number and intensity of symptoms reported; b) The Positive Symptom Total (PST), reflecting number of symptoms reported; and c) The Positive Symptom Distress Index (PSDI) reflecting intensity of symptoms reported. Due to the potential problem with regards to multicollinearity presnted by the use of additive scores in conjuction with subscales, only the scores from the nine symptom subscales were included in the statistical analysis. High correlations between subscales scores are a normal feature of this test, resulting in possibly inflated numbers of significant variates. However, the depth and breadth of information available through the subscale scores was viewed as the preferred alternative in a study of differences between individuals. Reliability and validity information reported in the SCL-90-R manual (Derogatis, 1975) suggests that the instrument is well-researched. Correlational coefficients comparing the nine symptom subscales range from .77 to .90 (Derogatis, 1975), suggesting high internal consistency. Such high coefficients, however, suggest that distinctions between subscales should be made with caution. Test-retest coefficients ranging from .78 to .94 (Derogatis, 1977; Edwards, 1978) have been reported. Concurrent validity of the SCL-90-R has been demonstrated with high reported correlations with symptom scores on the MMPI and the Middlesex Hospital Questionnaire (Boleoucky & Horvath, 1974; Derogatis, Rickels & Rock, 1976). Normative data is provided in the SCL-90-R manual (Derogatis, 1975) for both psychiatric and non-psychiatric adult and adolescent populations. The SCL-90-R has been tested extensively, in a wide range of clinical and medical research. Of interest in the present context, several studies have found the SCL-90-R to be useful in assessing changes in stress levels in populations responding to a variety of health-related difficulties. Clinical levels of distress were detected in individuals with sexual dysfunctions (Derogatis, 1976), cancer patients (Craig & Abelhoff, 1974), and chronic pain patients (Hendler, Derogatis, Avella & Long, 1977). Derogatis (1980) also utilized the SCL-90-R to develop a 67  Methodology  profile of the psychological impact of breast and gynaecological cancers on women. In these and other studies, the SCL-90-R has demonstrated a high degree of sensitivity to changes in stress levels, as well as the ability to discriminate between categories of patients, and between patient and non-patient populations (Derogatis, 1975). For the present study the demonstrated sensitivity of the SCL-90-R to stress levels is of greatest interest. It may be expected that non-psychiatric-case levels of distress may be indicated in a variety of the symptom categories, many of which are consistent with stress-related symptoms reported by infertility patients (Menning, 1980). Finally, the substantial validity and reliability data reported (Derogatis, 1975), the provision of relevant normative data, as well as the ease of administration and brevity of the SCL-90-R make it a useful instrument in the measurement of symptomatic distress levels of individuals responding to the experience of infertility. Adjustment to Infertility (AIS)  The Adjustment to Infertility Scale (Sabatelli, Meth & Gavazzi, 1988) was used to measure the level of adjustment to, or acceptance of a number of issues which have been identified as relevant to the experience of infertility. The MS consists of thirty-two 'I-statements' about feelings, perceptions and behaviors of the individual relative to the experience of infertility. Respondents indicate their degree of agreement with each statement on a 5-point Likert-type scale, ranging from 1='strongly disagree' to 5='strongly agree'. Areas of focus assessed by the AIS include the impact of infertility upon the respondents self-concept, sexual identity, marital relationship and sexual relationship. Higher scores indicate better adjustment. Completion time is approximately five minutes. Sabatelli et al. (1988) provided normative data on sample of 52 women and 29 men, all of whom were members of infertile couples. A reliability coefficient of .91 was reported (Sabatelli et al., 1988), although it is unclear how this was achieved. Moderate to high negative correlations with measures of anxiety and depression, ranging from -.29 to -.70 (Sabatelli et al., 1988), lend support to claims of construct validity. Based upon much of the literature previously cited (Frias & Wilson, 1985; Kraft et al., 1980; Menning, 1980; Sabatelli et al., 1988; Valentine, 1986; 68  Methodology  Woollett, 1985), the measure appears to address many of the issues of greatest relevance to those experiencing infertility, frequently echoing statements reported by infertile subjects. While the measure at present appears to lack sufficient supportive data to warrant its exclusive use as a criterion measure, the focus upon the primary issue of infertility provides a valuable adjunct to the more general factors assessed by the SCL-90-R (Derogatis, 1975). For this reason, as well as its ease of administration, the AIS was selected as a measure in the study of adults experiencing the infertility transition.  Procedure For the period of just over one month, all eligible patients attending scheduled appointments at the the Infertility clinic in the Obstetrics and Gynaecology Department at Foothills Hospital in Calgary, Alberta were informed of the study. The assistance and cooperation of clinic physicians was enlisted in identifying patients who met the eligibility criteria. Physicians informed all eligible patients of the study and invited each to volunteer as subjects. Only one couple was reported to have declined at this point. After signing a release (Appendix B), agreeing to allow the physician to provide the researcher with their name(s), interested couples and individuals proceeded directly to the researcher, located in a spare office or examining room within the clinic. Interviews were conducted with all candidates by the principal investigator, to explain the purpose, the structure and the requirements of the study and to provide an opportunity for candidates to ask questions. In cases where both members of the couple participated, the interview was held jointly. To guide the explanation, the letter of introduction (Appendix C) was read aloud by the interviewer as volunteers read along. Volunteers were then given an opportunity to raise questions and concerns. The interviewer endeavored to respond as completely as possible, without revealing the hypotheses of the study. At this time, candidates were offered the opportunity to withdraw from participation. All volunteers interviewed agreed to proceed at this point. All participants then read and signed a consent form which articulated each subject's rights, including the right to confidentiality and the right to withdraw at any time, as well as an assurance that their participation as subjects would be entirely independent of any and all medical treatment  69  Methodology  received at the clinic (Appendix D). Addresses were collected on this form, for purposes of follow. up correspondence. Completed consent forms were held in a file separate from all response materials. Once consent forms were completed, subjects were given the P.I.S. (see Appendix A), containing questions seeking demographic and factual data. Subjects were asked to complete this survey before leaving the clinic. To facilitate this, the researcher left the room to allow for privacy. After approximately five minutes, the researcher returned to collect the completed P.I.S. forms. Each participant was then given the questionnaire package, containing a second copy of the introductory letter, the five measurement instruments and a large, stamped envelope, preaddressed to the investigator. The five instruments were counter-balanced for order of presentation, using a randomized Latin square design, to address any effects order of presentation may have on subject responses. The average time required to complete each of the five instruments was between 5 and 15 minutes. The entire questionnaire package required between 45 minutes and one hour to complete. Subjects were specifically instructed to complete the contents independent of their partner, to control for the effects of interpersonal influence on subject responses. To reduce the effects of social desirability, subjects were encouraged to respond to the measures as candidly as possible, reiterating that anonymity of responses was assured. Subjects were instructed to omit their names from all material in the questionnaire package. Individual response forms were identified only by numerical codes, to facilitate analysis of the data. Questionnaire packages were taken home to be completed and returned through the mail in the pre-addressed, stamped envelope provided. Subjects who did not return questionnaire packages within one month of the interview were sent a follow-up letter (Appendix E) encouraging participants to return the instruments as soon as possible. Participants who had not returned the completed package within one month of the distribution of the follow-up letter were identified as having withdrawn from the study and no further contact was made.  70  Methodology  Analysis of Data  The goal of the present study was to determine the degree to which the characteristics of the individual and characteristics of the particular transition were related to the individual's levels of adaptation and psychological distress in response to the experience of infertility. The question was of relatedness and a procedure describing a multi-correlational relationship was therefore required. Although several formulae are adequate to describe the relationships of the several independent variables, the inclusion of multiple dependent measures restricted the choice of appropriate analysis to Canonical correlation (Tabachnick & Fidell, 1983), a procedure ideally suited to answering the questions asked within the study. Two canonical correlations were performed, one for each of the two sets of independent variables; characteristics of the transition and characteristics of the individual. Canonical correlations produce pairs of linear combinations, or 'canonical variates', representing each significant correlation between predictor and criterion variables. The use of this procedure provides extensive descriptive data which reveals the various dimensions along which the two sets of predictor variables and criterion variables are related. Following the two canonical analyses, which constituted the primary investigation, a number of additional statistical procedures were performed to clarify various elements of the data. T-tests were performed to compare subjects by gender on the new canonical variates. In addition, a series of multivariate analysis of variance procedures were performed to compare groups by race and sex-role on the newly created canonical variables. New canonical analyses were performed with these variables removed to facilitate the comparisons. All statistical procedures were accomplished with the use of the BMDP statistical package for computers.  71  Results  CHAPTER FOUR Results  In this chapter the results of all statistical analyses performed on the collected data are presented. Descriptive statistics of the scores obtained by the sample on the standardized measures are reported, as well as the results of the canonical correlational analyses depicting the relationship between the predictor variables and the criterion variables. Assumptions and Data Handling  In this section the manipulations or deletions made to the raw data are described. Included in this description are an explanation of the handling of missing data as well as an examination of the assumptions underlying the statistical analyses and the procedures used to handle data which failed to conform to these assumptions. Cases including missing data were handled in one of two ways. On variables for which a mathematical average could be obtained (any variables scored on an interval or ordinal scale), the missing score was replaced with the average score. For those items which were measured on a nominal (categorical) scale, missing cases were left blank. An examination of the assumption of normality basic to a canonical analysis revealed abnormally high skewness and kurtosis for a number of variables. Each variable was examined individually for the effect of its non-normal distribution and was either deleted or included, according to the impact of the distribution on the overall result. The distribution of scores on one predictor variable and one criterion variable were extreme in their abnormality. As reported in Chapter 3, the Yes/No split on the dichotomous predictor variable regarding the cultural value of children was 93% to 7%, producing abnormally high skewness (2.90) and kurtosis (6.48). This imbalance rendered the effect on the canonical variates difficult to interpret, and resulted in the variable being dropped from further correlational analyses. The distribution of scores on the Phobic Anxiety' subtest of the Derogatis SCL-90-R was also strongly positively skewed (2.88) with extreme kurtosis (8.42). Examination of the raw data 72  Results  revealed that 69% of respondents scored '0.00' on this subtest. As this variable had a very low loading on the output canonical variates, it was removed from subsequent calculations to eliminate the possible negative effects of its abnormal distribution. Scores on three other criterion variables, 'Somatization', 'Anxiety' and 'Psychoticism', were sufficiently abnormally distributed to require closer scrutiny. Distributions for each of these variables were moderately positively skewed (-2.00) with moderately high kurtosis (-4.4). After removing each variable singly as well as cumulatively from successive runs of the canonical correlation, it was noted that the pattern of relative loadings of the remaining variables remained stable across trials, suggesting little negative effect of the distribution patterns. To maintain the integrity of the study, as well as to increase the interpretability of the results, these variables were included in the final analysis. Five additional predictor variables received treatment of note. Two characteristics of the individual variables (religion and previous experience) consistently exhibited extremely low correlations with significant canonical variables on successive runs of the program and were therefore dropped from subsequent runs. A third predictor variable, sex, demonstrated consistently low contributions to the variates and was removed from the pool of input variables to facilitate comparisons by sex on the outcome canonical variate scores (reported in Additional Analyses). Fourth, the number of months seeking medical assistance was highly correlated with the number of months attempting to conceive. To avoid the effects of multicollinearity, only the latter measure was used. Finally, the distribution of the sample by race or ethnic origin was moderately uneven, with a disproportionate number being grouped into a single category (skewness = 2.16; kurtosis = 3.18). Although the distribution was of some concern, the contribution of this variable to the outcome variate scores was consistently significant, although difficult to interpret. Scores for race were therefore included in the main analysis, but excluded from one run so that outcome canonical variate scores might be compared by racial category to determine the nature of their influence.  73  Results  Descriptive Data: Independent Variables  A summary of the descriptive data for all variables entered into the canonical equation as independent variables is included in Table 3. Mean scores for Self-Esteem (78.45 ± 15.68) and Locus of Control (13.65 ± 3.81) compared favorably with standardized norms for similar populations. Scores for Income level and Education level were combined arithmetically (Income x Education) to derive the new variable Socioeconomic Status (SES). Participants tended to be representative of a somewhat high socioeconomic group. Table 3: Descriptive Data for Independent Variables Characteristics of the Transition  Characteristics of the Individual Variable  Mean  SD  Self-Esteem Locus of Control Age SES Sex-Role Health Race  78.45 13.65 31.52 10.00 *2.34 *3.59 *1.36  15.68 3.81 4.87 3.25 1.13 .52 .88  Variable  Mean  SD  Diagnosis Duration (in months) Perceived Outcome Hope  *2.83 48.11 * 1.63 *2.21  1.31 28.88 .78 1.09  * See following paragraph for explanation The remaining independent variables included in the canonical analyses were categorical in nature and therefore require more complete description. Of the Characteristics of the Individual entered into the analysis, scores for the variables of Sex-Role Identity, General Health and Race reflect categorical groupings. As noted in Chapter 3, individuals were assigned to one of four sexrole categories based on their mean scores on the subscales of masculinity and femininity. The four categories were #1 Androgynous (32%), #2 Masculine (18%), #3 Feminine (30%) and #4 Undifferentiated (20%). The group mean (II = 2.34 ± 1.13) reflects the fairly even distribution of the scores. Scores on the variable Race reflect four racial and ethnic categories, including White/Anglo (1), French Canadian (2), Asian (3) and Other (4). The mean score (1.1 = 1.36 ± .88) indicates a clustering of scores in the first two categories. The mean score of the categorical variable General  74  Results  Health (II = 3.59 ± .52) reflects a clustering of scores at the higher end of the range (1 = Poor, 2 = Fair, 3 = Good, 4 = Excellent), indicating a majority of the sample reported good or excellent health. Three variables included in Characteristics of the Transition, including Diagnosis, Perceived Outcome and Hope, were also measured categorically. Diagnosis was measured in five categories, including No diagnosis (1), Unknown Cause (2), Male Factor (3), Female Factor (4) and Couple Factor (5). For the present analysis, the mean score of II. = 2.83± 1.31 is included. The mean score of perceived outcome (II = 1.63 ± .78) represents scores in three categories, Uncertain (1), Temporary (2) and Permanent (3). Finally, the individual's perceived chances of success (Hope) were measured in ordered categories with (1) representing a less than 25% chance, (2) reflecting a 25 to 50% chance, (3) indicating a 50 to 75% chance and (4) reflecting a better than 75% chance of success. Distributions for each of these variables are detailed in Chapter 3. Descriptive Data: Dependent Variables A summary of descriptive data for the scores entered into the canonical analysis as dependent variables is included in Table 4. Subject scores for symptom subscales on the SCL-90R were compared with standardized norms for men and women provided by Derogatis (1975). On average, subject means were approximately one standard deviation above non-patient population means, indicating elevated distress levels. Data for all subjects (males and females) was combined for entry into the canonical analyses. Table 4: Dependent Variables Variable Somatization Obsessive Compulsiveness Interpersonal Sensitivity Depression Anxiety Hostility Paranoid Ideation Psychoticism Adjustment to Infertility  Women Mean SD .44 .55 .57 .58 .38 .45 .40 .22 119.05  Men Mean  .36 .31 .60 .54 .40 .46 .48 .50 .45 .35 .48 .43 .54 .31 .30 .20 15.98 122.67  75  SD  Combined Mean SD  .30 .44 .44 .61 .40 .54 .45 .58 .31 .40 .45 .46 .39 .40 .34 .23 16.84 120.13  .45 .56 .49 .50 .42 .47 .52 .32 16.99  Results  Canonical Correlational Analyses  Canonical correlational analyses were performed using BMDP6M (Frane, 1988) between the set of predictor variables and the two sets of criterion variables. Separate analyses were computed for those variables included in the 'characteristics of the individual' cluster and those included in the 'characteristics of the transition' cluster. Characteristics of the Transition  Four characteristics of the transition variables were entered into a canonical correlational analysis as predictor variables. These were: Infertility diagnosis (five levels), Perception of Outcome (three levels), Perception of Chances for Success (four levels: ordinal) and Duration of Infertility (number of months trying to conceive).  Eight symptom subscale scores as well as an  Adjustment to Infertility score were entered into the correlation as criterion variables. Hypothesis One:  The first research hypothesis predicted that there would be no statistically significant canonical correlation at the  a = .05 level on the mean scores on a weighted linear composite of  scores for the dependent variables of Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Psychoticism and Adjustment to Infertility for the independent variables of Diagnosis, Duration, Perception of Outcome and Perception of Success and the joint effects of these factors. The results of the first canonical correlational analysis supported the null hypothesis. The first canonical correlation was .58 (X 2 = 49.04; df = 36; a = .07). Although it was not significant at the a = .05 level, the correlation coefficient suggested that the two sets of variables were correlated at approximately r =.6, indicating that some relationship existed between the dependent and independent variables (see Figure 1). Canonical loadings, which are the correlations between the new canonical variables and the original variables, are presented in Table 5. An examination of these statistics revealed that, of the independent variables, Perception of Outcome (r = .5) and Duration (r = .48) appeared to  76  Results  contribute most strongly to this relationship. Canonical loadings of the dependent variables indicated that 'Interpersonal Sensitivity' was the single strongest contributor to the correlation. The direction of these scores relative to each other appeared to suggest that symptoms of Interpersonal Sensitivity tended to decrease as the duration of infertility increased and were also less for individuals who perceived their infertility to be a permanent condition. Figure 1  Linearity of Correlation Between First Canonical Variates Characteristics of the Transition  1  1  2  1  1  1  1  1  1 1  1  11  11  1 1 1  11 1 12 12 1  11 1  1  1 1 1 1  11  1 1  1  1  1 1  1  1  11 11 11  1  11 1  11  -1  -  1  +  1  1  1  1  1  1  1 2  1 1  +  +  +  -1.8 -2.4  -1.2  + -.60  + .60  +  + 1.2  0.0  CNVRS1  77  + 1.8  + 2.4  Results  Table 5: Loadings for First Canonical Variable (Correlations of Canonical Variables with Original Variables) Independent Variables  Dependent Variables  Variable  Variable  CNVRF1  Diagnosis Duration Outcome Chances  -.392 .477 .500 .037  Somatization Obsessive Compulsiveness Interpersonal Sensitivity Depression Anxiety Hostility Paranoid Ideation Psychoticism Adjustment to Infertility  CNVRS1 .065 .013 -.403 -.154 .244 .312 .114 .016 -.148  Characteristics of the Individual  Eight 'characteristics of the individual' variables were entered into a canonical correlational analysis as predictor variables. These were: Self-Esteem, Locus of Control, Sex-Role (four categories), General Health (four levels), Age, Race (four categories), Income (four levels) and Education (four levels). The latter two variables were arithmetically combined using a transform paragraph to create a new variable SES (socioeconomic status). Eight symptom subscale scores as well as an Adjustment to Infertility score were entered into the correlation as criterion variables. Hypothesis Two:  The second research hypothesis predicted that there would be no statistically significant canonical correlation at the a = .05 level on the mean scores on a weighted linear composite of scores for the dependent variables of Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Psychoticism and Adjustment to Infertility for the independent variables of Self-Esteem, Locus of Control, Sex-Role, General Health, Age, Race and SES and the joint effects of these factors. The canonical correlations and the results of the Chi-square test for statistical significance of the canonical correlations are displayed in Table 6. The first canonical correlation was .70 and was significant at the a = .01 level. The second canonical correlation was .59 and was significant at the a = .05 level. The third canonical correlation was .49 and was not significant. 78    Results  These results indicated that two significant and non-orthogonal links exist between the two sets of variables (See Figures 2 & 3) and required a rejection of the null hypothesis for this set of variables.  Table 6: Results of Canonical Correlations Canonical Con.  Eigenvalue  Chi-Square  df  110.25 67.99  63 48  .70 .59  .497 .351  Probability .00 .03  Figure 2	Linearity of Correlation Between First Canonical Variates Characteristics of the Individual  2 12 11  1 2 1 2  1 1 1 1 11  + 1  11 1 1111 1	 1 2 1  1 1  1 2  1 1  11	 1 1 2  1  -  2  11  1 1  1  1  1  1  1  2 -2  -.50  -1.5  -2.5 -3.0  -2.0  0.0  -1.0 CNN/RS 1  79    1.5  50 1.0  Results  Linearity of Correlation Between Second Canonical Variates Characteristics of the Individual  Figure 3  3  1  2  1  1 1 1  2 1  +  2 1  11 1  1  1 11  1 2  1  2 1	 1 1 1 112 1 1	 11 11 1 1 12	 1	 2  1 1  1  1  1 2  1  1  1 1 1  1 1  1 1 1  1  1 1  4  -1.8  1.8  0.0  4.5  2.7  .90  -.90  3.6  CNVRS2  A summary of the canonical loadings of the independent variables on the first two canonical variates is presented in Table 7. An examination of these statistics revealed that five independent variables contributed to the first canonical variable. Self-esteem (r = .57), SES (r = .53), and Race (r = .51) all contributed quite strongly to the new variable, with self-esteem contributing the strongest effect. Locus of control (r = .47) and Age (r = .43) also contributed at moderate but significant levels. As self-esteem and locus of control appeared to play a significant part in this  80  Results  variable, as well as high socioeconomic status, the underlying dimension linking the variables appeared to be related to a sense of well-being or physical and psychological 'Security'. t This may involve what Schlossberg (1981) refers to as 'Psycholosocial Competence'. The influence of Race was not easily understood within this context and was addressed separately in a later section. Table 7: Loadings for First and Second Canonical Variables (Correlations of Canonical Variables with Original Variables) Independent Variables Self-Esteem Locus of Control Sex-Role Health Age Race SES  CNVRF1  CNVRF2  .57 .47 -.16 -.06 .43 .51 .53  -.59 -.19 .41 -.21 .69 .26 -.11  Canonical loadings of the first two canonical variables for the second set (dependent) of variables are presented in Table 8. A number of the original variables contributed at a significant level to the new variables, with Hostility (r = -.78) and Somatization (r = -.7) exhibiting very strong correlations and Interpersonal Sensitivity (r = -.57), Paranoid Ideation (r = -.5), Depression (r = -.5) and Obsessive-Compulsiveness (r = -.5) exhibiting a moderately strong effect. Anxiety (r = -.44) and Adjustment to Infertility (r = .45) also contributed to the overall correlation. The dimension underlying this first canonical variate appeared to be related to the effects of irritability and anger, both internalized and externalized . By examining the content and the direction of the loadings on first canonical variable, it was possible to suggest the nature of the relationship between the two sets of original variables. The results obtained suggested that those individuals who reported a high level of psychosocial competence, or security with themselves and their surroundings were less likely to respond to infertility with symptoms of tension and anger and were somewhat more likely to report higher levels of adjustment to infertility, than those individuals who indicated less physical or  81  Results  psychosocial security. Table 8: Loadings of First and Second Canonical Variates (Correlations of Canonical Variables with Original Variables) Dependent Variables Somatization Obsessive Compulsiveness Interpersonal Sensitivity Depression Anxiety Hostility Paranoid Ideation Psychoticism Adjustment to Infertility  CNVRS1 -.70 -.50 -.57 -.51 -.44 -.78 -.51 -.32 .45  CNVRS2 .32 .70 .51 .60 .80 .19 .33 .55 -.09  Examination of the canonical loadings for the second new canonical variable revealed a new set of links between the two sets of original variables. The second canonical variable was independent of the first, as it represented the linkages which exist between the two sets of variables after the effects of the first canonical variable were removed. Loadings of the predictor variables into the second canonical variate indicated that three of the independent variables played a significant role in the relationship. In this second new variable, the effect of Age (r = .69) appeared to be the strongest contributor, with Self-Esteem (r = -.59) contributing a significant negative effect. Sex-Role also appeared to be correlated moderately strongly (r = .40) to this new variable, although it showed only a negligible effect on the first canonical variate. The structural dimension represented by this variable was more difficult to define than was the first. It was clear that higher age and lower self-esteem appeared to be clustered. The influence of Sex-Role was less clear and required some interpretation. Higher scores on the categorical variable of sex-role were achieved by individuals who identified with predominantly feminine characteristics (category 3) or who showed low identification with descriptors of either gender-type (undifferentiated: category 4). Both categories are distinguished by low identification with (masculine) traits which reflect a sense of self-directedness or personal efficacy. Individuals in the latter category, undifferentiated, have also failed to identify with 82  Results  (feminine) traits reflecting qualities such as emotional warmth and nurturance. Taken together, the masculine and feminine character traits included in the Bern (1974) instrument constitute a fairly comprehensive listing of those personality traits considered valued in our society. Individuals who deny having any of these traits may be indicating an image of themselves as lacking in many personality traits. Thus the higher sex-role scores loading on this variate appeared to indicate a group of individuals who characterize themselves as lacking in those aspects of personality most valued by society. In combination with the low self-esteem and higher age reported by these individuals, it appeared that the construct informing this second canonical variate was a measure reflective of a poorly defined and negative self-image. On the other side of this second new variate, Anxiety (r = .8) and ObsessiveCompulsiveness (r = .7) exhibited very strong correlations, with Depression (r =.6), Psychoticism (r = .55) and Interpersonal Sensitivity (r = .5) all contributing moderately strongly. Taken together, it appeared that the underlying dimension measured by this canonical variate related to the feelings and behavior symptomatic of anxiety or nervousness. In examining the structure of the canonical loadings for the second canonical variable, it was possible to determine the nature and shape of the relationship identified between the two sets of predictor and criterion variables. Unlike the first canonical variate-pair, which primarily measured the positive impact of individual strengths on levels of distress experienced by the participants, the second variable appeared to measure the effect of individual weaknesses. The results suggested that those infertile individuals who were older and who had low self-esteem and an undifferentiated gender identity also tended to exhibit elevated levels of anxiety and nervousness, expressed as increased emotional and behavioral symptoms of distress. Additional Analyses In an effort to further clarify the pattern of results obtained through the canonical correlational analyses, a series of comparison tests were performed. Using individual scores on the newly created canonical variates as raw data, subjects were compared on the basis of gender, race and sex-role. 83  Results  Comparison on Canonical Variates by Gender  Using BMDP3D (Sockne & Forsythe, 1988) independent t-tests were performed to compare males and females on the two canonical variate pairs. The results are presented in Table 9. As indicated, no significant differences were found between males and females on either of the canonical variables.  Table 9: Variable CONVRF1 CONVRF2 CONVRS1 CONVRS2  Comparison of Males & Females on First and Second Canonical Variate Pairs Men T Score Women SD 11	SD I.1 -.12 -.05 -.10 -.07  .97 .97 1.03 1.13  .21 .08 .15 .11  1.03 1.06 .95 .78  -1.35 -.53 -1.07 -.82  df  P Value  55.1 54.2 60.9 68.7  .18 .60 .29 .42  Comparison of Canonical Variables by Race  As previously reported (see Table 7), scores on the variable Race showed a strong positive correlation (r = .51) with the first new canonical variable (CNVRF1). The nature or the meaning of this relationship was difficult to interpret within the context of the underlying dimension of psychosocial competence or security. To clarify the role played by race, a new canonical analysis was performed. For this run, scores for race were removed from the pool of predictor variables so that the outcome canonical variables could be compared by racial group. Using the BMDP6M (Frane 1988), a canonical correlational analysis was performed with race excluded. Input variables in the first set included: Self-Esteem, Locus of Control, Sex-role, Health, Age and SES. Input variables for the second set were: Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Paranoid Ideation, Psychoticism and Adjustment to Infertility. The results of this new canonical correlation (See Table 10) reveal that the percentage of variance, as indicated by the Eigenvalue, dropped from almost 50% of the variance accounted for in the first analysis, to 39% of the variance accounted for  84  Results  when race was removed. However, this appears to be the only change. There are still two strong links between the two sets of variables, with the first canonical link having a correlation of r = .62 and the second also showing a strong correlation of r = .58.  Table 10: Canonical Correlations - Race Excluded  Eigenvalue Canonical Chi-Square df Correlation .621 .582  .386 .339  88.33 58.05  54 40  Probability .002 .032  Furthermore, as can be seen in Table 11, the pattern of correlations between the original variables and the new canonical variables (canonical loadings) remains essentially unchanged. The contributions of each of the variables, relative to one another, is virtually the same as was seen in Tables 5 and 6. Table 11:	 Loadings for First and Second Canonical Variates (Correlation of Canonical Variables with Original Variables) Dependent Variables  Independent Variables Variable  CNVRF1  Self-Esteem .873 Locus of Control	 .600 Sex-Role -.145 .054 Health .123 Age SES .639  CNVRF2 -.190 .113 .481 -.196 .828 .240  Variable  CNVRS1  Somatization Obsessive-Compulsive Interpersonal Sensitivity Depression Anxiety Hostility Paranoid Ideation Psychoticism Adjustment to Infertility  85  -.667 -.815 -.800 -.746 -.748 -.718 -.728 -.663 .567  CNVRS2 .070 .351 .150 .296 .525 -.112 -.046 .228 .240  Results  Using BMDP7D (Dixon, Sampson & Mindle, 1988) to conduct a one-way analysis of variance, subject scores were compared by racial group (White/Anglo, French Canadian, Asian & Other) on the four new outcome variables, CNVRF1, CNVRF2, CNVRS 1 and CNVRS2. Separate analyses were performed for each of the outcome variables. Descriptive data and probability values are reported in Table 12. As can be seen, no significant differences were shown between groups.  Table 12: Comparison of Group Means by Race Variable CNVRF1 CNVRF2 CNVRS1 CNVRS2  Anglo 1.1. SD .06 ± .9 -.04 ± 1.0 -.02 ± 1.0 -.06 ± 1.0  Fr. Can. pt  SD  .44 ± .5 .51 ± .4 .76 ± .3 .09 ± .4  Asian p.  Other SD  -.17 ± 1.4 -.34 ±	 .4 -.40 ± 2.0 .29 ± 1.1  pi.  df  F Value P-Value  SD  -.82 ±	 .7 .45 ± 1.5 .09 ±	 .8 .57 ± 1.4  3 3 3 3  1.46 .81 .80 .72  .233 .495 .499 .545  Group sizes were dramatically disproportionate however, as White/Anglos outnumbered all the other groups combined by a factor of five to one. Although statistical corrections for this imbalance are provided with the BMDP program, the extremely small numbers in the other representative groups (French Canadian, Asian and Other) renders this comparison relatively meaningless. As no group differences could be identified on the variable of race, it is suggested that the high loading observed on this variable in the first canonical variate may have been merely a factor of the skewness found in these scores. Comparison of Canonical Variates by Sex-Role  Although the variable sex-role was not significantly related to the first new canonical variable formed, it played an important role in defining the underlying dimension of the second canonical variable (See Table 7). In an effort to further clarify the relationship between sex-role identity and the new canonical variables formed, a new canonical analysis was performed in which sex-role was removed from the pool of original predictor variables, so that the outcome canonical 86  Results  variables could be compared by sex-role category. Using the BMDP6M (Frane, 1988), a canonical correlational analysis was performed with sex-role excluded. Input variables in the first set included: Self-Esteem, Locus of Control, Health, Age, Race and SES. Input variables for the second set included: Somatization, Obsessive Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Paranoid Ideation, Psychoticism and Adjustment to Infertility. The results of this new canonical correlation (See Table 13) revealed two strong links between the two sets of variables. The first canonical link  a < .001 level. The second canonical correlation was also high (r = .58), and was significant (a =.05). showed a strong correlation (r = .70) and was significant at the  Table 13: Canonical Correlations - Sex-Role Excluded Eigenvalue Canonical Correlation .496 .342  Chi-Square  Probability  54 40  98.25 55.73  .705 .585  df  .00 .05  As can be seen in Table 14, the pattern of correlations between the original variables and the new canonical variables (canonical loadings) remained essentially unchanged. The contributions of each of the variables, relative to one another, were virtually the same as was seen in Tables 7 and 8. Table 14: Loadings for the First and Second Canonical Variables (Correlations of Canonical Variables with Original Variables) Dependent Variables  Independent Variables Variable Self-Esteem Locus of Control Health Age Race SES  CNVRF1 CNVRF2 .56 .47 -.06 .44 .50 .54  -.65 -.24 -.21 .65 .35 -.19  Variable Somatization Obsessive-Compulsiveness Interpersonal Sensitivity Depression Anxiety Hostility Paranoid Ideation Psychoticism Adjustment to Infertility 87  CNVRS 1 CNVRS2 -.69 -.50 -.57 -.50 -.42 -.77 -.51 -.32 .46  .31 .74 -.55 .62 .80 .18 .44 .62 -.20  Results  Using BMDP7D (Dixon et al, 1988) to conduct a one-way analysis of variance, subject scores were compared by sex-role category (Androgynous, Masculine, Feminine &Undifferentiated) on the four new outcome variates, CNVRF1, CNVRF2, CNVRS 1 and CNVRS2. Separate analyses were performed for each of the outcome variables. Descriptive data and probability values are reported in Table 15.  Table 15: Comparison of Group Means by Sex-Role Variable  Androgynous A SD  CNVRF1 CNVRF2 CNVRS1  .08 ± 1.0 -.19 ± .8 .14 ± 1.0  CNVRS2  -.16 ± .9  Masculine IA  SD  .21 ± 1.0 -.06 ±	 .8 -.13 ±	 .9 .02 ±	 .7  Feminine g  SD  -.07 ± 1.0 .04 ± 1.1 .04 ±	 .9 -.06 ±	 .9  Undifferentiated df F Value P-Value g  SD  -.22 ± 1.1 .33 ± 1.3 -.20 ± 1.2  3 3 3  .46 .76 .40  .39 ± 1.6  3  .87  .714 .520 .750 .459  No statistically significant group differences were found. However, the pattern of group means suggested some interesting trends. Undifferentiated subjects appeared to score somewhat higher than the other three groups on CNVRF2 and CNVRS2, the two variables which suggested a relationship between a negative self-image and high anxiety. Although these differences were not significant, the trend appeared to support the inclusion of undifferentiated sex-role as an identifying characteristic of the group represented by the canonical link. Conversely, these same subjects appeared to score lower than the other three groups on CNVRF1 and CNVRS1, the linked correlates which reflect low distress and high psychosocial security, suggesting the possibility that undifferentiated indiviudals may tend to report less psychosocial security and more hostility and somatization. Scores for high masculine subjects also appeared to indicate a possible trend. High masculine individuals scored quite high on CNVRF1, but scored very low on the outcome variable (CNVRS 1). This trend appeared to indicate the possibility that, although high masculine subjects  88  Results  tended to indicate a high level of psychosocial security, they may have actually exhibited heightened levels of hostility and somatization as compared with feminine and androgynous subjects. Summary  In summary, a series of statistical analyses were performed with the data. Two hypotheses were offered regarding the relationships between a group of dependent variables and the two groups of independent variables. No significant relationship was detected for the first set of independent variables (characteristics of the transition) and the dependent variables (responses to infertility) and the null hypothesis was therefore accepted. For the second hypothesis, however, the null was rejected as two distinct correlational links were shown between the set of independent variables (characteristics of the individual) and the dependent variables (responses to infertility). The first link illuminated a relationship between physical and psychological security and a decreased tendency to respond to infertility with anger. The second link was characterized by a tendency for older individuals with low self-esteem and an undifferentiated sex-role to demonstrate elevated levels of anxiety and tension in response to infertility. Additional analyses were conducted to clarify the role of individual variables. Comparisons by gender indicated no significant differences between men and women on any of the canonical outcome variables. While race appeared to exhibit a strong influence on the first significant canonical link, subsequent analyses suggested that this effect may have been largely an artifact of the disproportionate weighting of a single racial category (White/Anglo). Finally, subsequent analyses to isolate the effects of sex-role revealed that, although no significant differences between sex-role types was detected, there was a tendency for undifferentiated individuals to score somewhat higher on the variates reflecting high anxiety and low-self-esteem and for masculine individuals to score somewhat lower on the variate measuring emotional calmness.  89  Discussion Conclusions  CLIAMEME Discussion  In this final chapter a discussion of the results is presented, including an attempt to interpret their meaning and relevance, as well as a brief commentary on the limitations of the study and recommendations for future directions in research. In addition, some considerations for counsellors working with infertile couples and individuals will be addressed.  Discussion of the Findings The explicit purpose of the present study has been to identify the characteristics of the individual and of the specifics of the transition, which may help counsellors and other helpers to predict how individuals and couples may respond to infertility. The implicit purpose of the study has been to provide counsellors with information which may help to identify and arrange appropriate interventions for those who are at high risk of experiencing serious difficulties in coping with the experience of infertility. While the focus has been on individual differences, it must be noted that the results indicated that, as a whole, the sample of infertile subjects exhibited moderately high levels of distress on all symptom subscales, as compared with standardized norms. Although elevated scores were noted on all symptom subscales, of particular note were elevated scores on measures of obsessive-compulsiveness, interpersonal sensitivity, depression, hostility and psychotic symptoms, all characterized by physical and emotional agitation. The distress scores achieved by the infertile sample were similar to those achieved by a sample of patients with chronic medical illness, but were well below scores achieved by a psychiatric population (Derogatis, 1975). Thus, infertility may be comparable in emotional impact to conditions such as cancer or kidney disease. While it must be recognized that the impact of infertility was felt by all the subjects, the purpose of this study was to identify individual differences in response. A series of canonical correlational analyses were employed because of the ability of this procedure to describe the number of different ways in which two sets of variables relate to one another. The analysis 90  Discussion Conclusions  produced a series of independent, non-orthogonal canonical variates which represented the unique ways in which the predictor variables, the characteristics of the individual and the characteristics of the transition, were related to the criterion variables, the measures of psychological distress and adjustment to infertility. The fact that more than one link between the two sets of variables was discovered is important because, instead of observing how infertile people as an aggregate group respond to infertility, the results allowed for an understanding of how different individuals respond to the experience. Characteristics of the Transition  The first analysis was performed to determine the link between specific characteristics of the transition, including the duration of infertility and the diagnosis, as well as the individual's perceptions regarding the outcome and the chances for success. Although no significant relationships were detected at the  a =.05 level between any of the characteristics of the transition  and the outcome indices of distress and adjustment, the high correlation (r = .6) was significant at  a = .07, and does suggest that the two sets of variables are related in some way. An examination of the data revealed that symptoms of interpersonal sensitivity, including self-consciousness and feelings of not being understood appeared to decrease somewhat as the duration of infertility increased. While these results may not be used for predictive purposes, they do lend support to the findings of Daniluk (1988) and others (Sabatelli et al., 1988), who found that distress levels tended to decrease as the length of time men and women lived with their infertility increased. Within the same canonical link, subject's perception of outcome also appeared to contribute strongly. In this case, it appeared that individuals who perceived their infertility to be permanent, rather than uncertain or temporary, reported somewhat fewer interpersonal distress symptoms. Again, this result is not significant at a = .05, but it does tend to support previous findings. Feuer (1983) reported that men who had received a diagnosis of azoospermia, a permanent and final condition, reported significantly lower distress levels than did men whose less final diagnosis left the outcome uncertain. The stress of uncertainty also appeared to play a role in Daniluk's (1988) findings that couples who had not yet received any diagnosis, or whose diagnosis was unclear or 91  Discussion Conclusions  ambiguous, demonstrated significantly higher levels of distress than couples who had received a diagnosis, even if that diagnosis was of permanent infertility. Thus, although the relationship shown by these data between decreased interpersonal sensitivity and the perception of infertility as permanent was not statistically significant, the findings appeared to corroborate a relationship demonstrated elsewhere. The remaining characteristics of the transition, diagnosis received and perceived chances of successful pregnancy, did not appear to contribute in any meaningful way to the link. Nor did any symptoms of distress other than interpersonal sensitivity appear to play any part in the relationship. The poor performance of diagnosis as a factor predicting distress levels may be understood as an artifact of the research methodology, providing an opportunity to address certain concerns regarding the choice of approach. The use of the general correlational procedure of canonical analysis restricted the ability to make fine distinctions regarding individual factors. In the case of diagnosis, subjects were not matched to diagnosis (ie. own factor/ partner's factor) nor was a comparison made on the basis of which partner was identified as being the cause of infertility. Rather, diagnosis by category was one of several possible factors within the pool. While this procedure may well have served to dilute or obscure the effects of individual factors, such as diagnosis, it did allow the researcher to examine a wide range of possible factors and to observe those factors which appeared to have the strongest predictive value. Thus, while there may or may not be differences in the way members of infertile couples respond on the basis of who is diagnosed as infertile, no effect was shown for this factor in the present study. What did appear to be evident, however, was a trend in which individuals who have been enduring infertility for an extended period of time, and who perceive infertility to be a permanent condition, demonstrated a decrease in self-consciousness and in the tendency to be sensitive or reactive to other people's inability to understand or empathize with their situation. It appears that the longer an individual lives with infertility, the more he or she becomes accustomed to being infertile. As time goes on, the insensitive and personal questions, and the illogical suggestions that infertile people are frequently subjected to (Sandelowski & Jones, 1986) become a 92  Discussion Conclusions  familiar and predictable occurrence, no longer engendering the feelings of pain, humiliation and anger they once did. With familiarity often comes a sense of comfort or acceptance of infertility. For those individuals and couples for whom infertility is permanent and final, acceptance of infertility may lead to the inclusion of infertility as a part of identity, an essential and final step in the process of successful adaptation to a transition (Schlossberg, 1981). It may be suggested that, for these individuals, the process of adaptation may be somewhat more rapid than for those individuals whose outcome is uncertain or ambiguous. Characteristics of the Individual  A majority of the sample scored high on the first canonical variable (See Fig. 2). The characteristics this group of individuals have in common may be described as 'psychosocial security', as reflected in the tendency for high self-esteem, an internal locus of control and higher SES to covary. It appeared that these men and women had a fairly solid and secure sense of themselves and their abilities in the world as they tended to indicate a positive evaluation of themselves and their abilities to effect change. By and large, they were mature in age and had a high level of education and a comfortable standard of living. Although group means for the sample indicated elevated levels of distress in general, the group whose scores contributed to the first canonical variable tended to exhibit an 'emotional calm' as indicated by the relatively low incidence of symptoms of distress overall, reflected in the moderate to high negative correlations of the symptom subscales with the new variable. In particular, this group exhibited fewer symptoms of anger, as measured by the symptom subscales of somatization and hostility, and of irritability or self-consciousness, as reflected by scores on the interpersonal sensitivity subscale. The positive effects of self-esteem on the individual's ability to cope effectively with infertility have been demonstrated repeatedly (Adler & Boxley, 1985; Chiappone, 1984; Fouad & Fahje, 1989; Murray & Callan, 1988; Sabatelli, Meth & Gavazzi, 1988). In accordance with these findings, self-esteem scores provided the single largest contribution to this first canonical 93  Discussion Conclusions  variate, accounting for the majority of the variance in the outcome variate representing lower levels of anger and interpersonal sensitivity. It is interesting to note that age contributed to this first canonical variate, indicating that higher age is moderately correlated with lower distress. This concurs with previous findings, (McEwan, et al., 1987; Sabatelli et al., 1988) which showed older women tending to report lower levels of distress. These fmdings appear to challenge the popular notion that, because age is inversely related to a woman's childbearing capability, increased age should lead to increased feelings of distress in response to infertility. However, age is only moderately correlated with the new canonical variate (r = .43). It may be that although these women and men are somewhat older than the mean, they are not the oldest in the sample. In fact, age is correlated much more strongly with the second canonical variate (r= .69), suggesting that those men and women contributing to the second link were, on average, an older group than those represented here. The meaning of age in this canonical variable may best be understood in conjunction with the covariates, higher SES, higher self-esteem and higher internal locus of control. The men and women who scored high on this variable were those individuals who have achieved a level of maturity, stability and material comfort, and who were not yet at the 'critical stage' when the time left on the woman's 'biological clock' would make further attempts to become pregnant highly risky and increasingly futile. As this is obviously an issue of greater salience to women than men, the mean scores on this variate were compared for the two sexes. A very slight nonsignificant difference was found between males and females for these paired canonical variates, with men scoring marginally higher than women on both, indicating little advantage for men over women in this group in terms of psychosocial security and emotional calmness. Individual scores on the first pair of canonical variates were compared for the four groups of sex-role types (androgynous, masculine, feminine and undifferentiated). Although the differences between group means were not statistically significant, the pattern of scores for the different groups were interesting. Undifferentiated individuals scored markedly low on the 'psychosocial security' variable. This observation was consistent with Adler and Boxley's (1985) 94  Discussion Conclusions  finding that undifferentiated sex-role types scored low on measures of self-esteem and body image. These results suggest that an undifferentiated sex-role identity may be indicative of a poorly defined sense of self overall, characterized by a lack of identification with positively valued characteristics of either men or women. These non-significant trends suggested that undifferentiated men and women also appeared to indicate a high level of anger, irritability and somatic symptoms. This trend may be reflective of a general pattern of vulnerability to distress which may, in turn, reflect poor coping skills. In fact, undifferentiated individuals also appear to display heightened levels of a variety of distress symptoms. This pattern may be more fully clarified with the results of the second canonical link. The second pair of canonical variates represented a relatively small but distinct group of subjects (Fig. 3). The dominant characteristics linking the group of people who scored high on the first (predictor) variable of the pair were low self-esteem, an undifferentiated sex-role identity and higher age. These characteristics were linked with an outcome variable comprised of high levels of emotional distress symptoms, marked primarily by heightened symptoms of anxiety, including nervousness, fearfulness and obsessive behaviors. As with the first variate, self-esteem scores contributed strongly to the new predictor variable. In this case, however, a strong negative correlation with self-esteem scores indicated a negative identity as a predominant defining feature. It appeared that the relationship demonstrated by this link is one between low self-esteem or a poorly defined self-image and a pattern of high levels of distress symptoms. This result supports the numerous studies which cite low self-esteem as a reliable predictor of poor coping efficacy (Chiappone, 1984; Fouad & Fahje, 1989; Sabatelli et al., 1988). Two other scores contributed strongly to the 'negative identity' variable. Among these was an undifferentiated sex-role identity, as suggested by a moderately high positive loading of sex-role scores. As the categorical nature of the sex-role scores made this result somewhat ambiguous, a further statistical analysis was conducted in which sex-role scores were removed from the pool of input variables and comparisons by sex-role score made on these new variables. Undifferentiated 95  Discussion Conclusions  individuals scored very high on both the new predictor variate (negative identity) and on the new outcome variate (high anxiety), in comparison to the mean scores for the other three sex-role groups. Although the differences observed were not statistically significant, this procedure served to clarify the relationship between sex-role scores and the new variable. It appeared that an undifferentiated sex-role score, indicating low identification with a variety of masculine and feminine characteristics, was linked with low self-esteem in predicting elevated feelings of anxiety in response to infertility. These findings concurred with the results of Adler and Boxley (1985), as cited above, that undifferentiated sex-role scores are related to low scores on measures of selfesteem and body image and to higher levels of distress in response to infertility. In addition to low self-esteem and an undifferentiated sex-role identity, age contributed very strongly to the 'negative self-image' variable, and in fact represented the strongest factor. This suggested that these highly anxious subjects were among the oldest of the sample. This result appeared to be in contradiction with earlier results in which age contributed to the 'psychosocial security' variate, as well as with the findings of McEwan et al. (1987) and Sabatelli et al. (1988), that older women reported lower levels of distress. In light of this apparent conflict, this finding required further investigation and clarification. Individual canonical variable scores were examined to identify those individuals who scored high (> .9) on the 'negative self-image' and 'high anxiety' variates. Of this group, over half were age 37 or older. The average age for men in this group was 39.0 years. The average age of women was 34.3 years, but this included one 26 year old. The median age for women in this group was 37 years. These results were compared with age means for high scorers (> .9) on the first variate pair linking 'psychosocial security' with 'emotional calmness' (32.0 years for women; 35.7 years for men) as well as with the age means for the sample as a whole (30.7 years for women; 32.7 years for men). Thus, it appeared that, although the men and women who scored high on the first variate pair were somewhat older than the sample mean, many of the men and women who scored high on negative identity and high anxiety were, in fact, the oldest members of the sample. Thus, these results would appear to indicate that the relationship between age and 96  Discussion Conclusions  emotional distress levels for infertile men and women may not be linear, as suggested by earlier results (McEwan et al., 1987; Sabatelli et al., 1988). Instead, these results suggest that there may actually be a curvilinear relationship between aging and distress levels for infertile men and women, with coping efficacy increasing with maturity up to a point, after which increased age appears to be related to an increase in feelings of anxiety and fearfulness for infertile men and women. The point at which the positive effects of age on responses to infertility cease and the negative effects begin cannot be identified with the data available. As such a relationship has not been reported elsewhere, further research would be required to validate and support this observation. In seeking to explain the reasons for the observed results regarding age, a number of possibilities may be examined. The most readily apparent theory would explain the results in terms of a race against the 'biological clock', suggesting that a woman experiencing infertility would develop increased symptoms of anxiety and distress as she nears the end of her natural reproductive years and the time remaining to seek a solution diminishes. This theory, however, is unsupported by other research. Moreover, it fails to adequately explain the fact that distress levels appear also to increase for men in the older age group. While it may be argued that older men are more likely to be married to older women who are facing a reproductive deadline, this argument does not satisfactorily explain the observed increase in older men's distress levels. As was the case with the first variate of psychosocial security, the influence of higher age on increased levels of distress may best be understood when examined in conjunction with the two other factors contributing to the variable of 'negative identity', low self-esteem and an undifferentiated sex-role. Taken together, these three factors appear to suggest a possible profile of individuals who have progressed through adulthood without having developed a definite sense of who they are or what they can accomplish. These men and women appear to have defined themselves in negative terms, and appear to be able to identify what they are not, but not what they are. They may have passively 'drifted', neglecting career, educational and personal development, with perhaps a long-held belief, that whatever else fails, they may always become parents. Faced 97  Discussion Conclusions  with infertility when there is little time left to realize this option, these individuals may be facing a profound crisis of identity. Schlossberg (1984) notes that identity issues are fundamental to the individual's ability to adjust to transitions. Individuals with a negative identity, who are unable to identify their strengths and resources, may be ill-prepared to face a transition if they see themselves as ill-equipped to cope. Such individuals may find the loss of fertility particularly overwhelming as it may in fact represent the perceived loss of the 'last chance' to develop a sense of purpose or meaning in life. A similar but distinctly different profile may also be derived of individuals who have actively postponed or eschewed self-development because they have long expected to assume the identity of parenthood. Matthews and Matthews (1986) have suggested that couples and individuals whose commitment to a 'parenthood identity' is high, and who fail to fulfil the expectations of that identity, may suffer an 'identity shock'. They note that "the greater the commitment to biological parenthood, the greater will be the identity shock brought about by infertility and involuntary childlessness, and the more negative the affect (sic) on self-esteem"(p. 646). Schlossberg (1984) also suggests that an individual whose sense of identity is derived primarily from a role, or the assumption of such a role, will find "the very foundations of (his or) her life shaken" (p. 22) when faced with the loss of that role. The levels of distress reported by the group of subjects identified as having a 'negative identity' may certainly be described as symptoms of a profound disruption of identity. These men and women scored high on measures of anxiety, obsessive-compulsiveness, depression, psychoticism and interpersonal sensitivity, indicating signs of an acute state of mental and emotional disorientation, irrational thoughts and behavior, and agitation. Summary  The results of the correlational analyses conducted appear to describe a number of factors which may be related to an individual's pattern of responses to infertility. Men and women who reported positive evaluations of themselves and who appeared to have higher levels of lifestyle security and comfort appeared also to report fewer symptoms of distress, especially symptoms of 98  Discussion Conclusions  anger. Men and women with undifferentiated sex-role identities, who have low self-esteem and who are considerably older than the average first-time parent appear to be at the greatest risk of feelings of distress and anxiety in response to infertility and may experience infertility as a crisis of identity. In addition, although the relationship was not observed at a statistically significant level, actual and perceived duration of infertility appeared to be strongly related to feelings of distress. Men and women who had been infertile for longer periods of time and who perceived their infertility to be permanent appeared to report fewer symptoms of distress than did those men and women who had been infertile for less time or who perceived the duration of their infertility to be temporary or uncertain. Implications for Counsellors  In discussing the implications and recommendations for counsellors, the purpose is to present a profile of those individuals who may be most at risk of experiencing difficulties in response to the experience of infertility. In this context, two risk groups may be identified and discussed. The specific counselling needs of each group will be outlined with recommendations for appropriate counselling approaches. Recently Infertile: The first group identified as being at risk of emotional distress in  response to infertility may be described as the 'newly infertile'. Although the results of the present study failed to show the link between duration of infertility and levels of distress at significant levels, the trend observed herein has been demonstrated elsewhere. Daniluk (1988), as well as Sabatelli et 0.(1988) observed that distress levels tended to decrease with time spent dealing with infertility. The implicit inference is that couples and individuals who have only recently sought medical attention regarding an inability to conceive, or who have just received a diagnosis of infertile, may be at high risk of experiencing feelings of disorientation, confusion and anger (Menning, 1980). It is during the initial days and weeks of the infertility investigation that couples may be most in need of supportive counselling. Menning (1980) has noted that, in response to the news of infertility, many people feel a profound sense of shock and surprise, often followed by denial 99  Discussion Conclusions  and anger. Daniluk (1988) has reported that infertile couples identified the first few weeks of the infertility investigation as the time when counselling would be most useful. The primary goals of counselling for those who are dealing with recent news of infertility would be to provide a safe environment, to validate the couple's feelings and to provide information. Many adults may have difficulty acknowledging the often overwhelming feelings of hurt, frustration, anger, guilt and sadness engendered by infertility, as they "may be so painful that (couples) must deny the reality of these feelings" (Wiehe, 1976, p.29). The most important task is to validate the client's experience. By encouraging the expression of feelings and demonstrating non-judgmental acceptance and reflection of emergent emotions, the counsellor provides the necessary environment in which the infertile couple can begin to recognize and express the extent to which infertility has affected their lives (Menning, 1980; Sabatelli et al, 1988; Sawatzky, 1981). Another important goal of counselling for this group is to provide information for the couple or individual about infertility and its impact. It is important to normalize the experience by explaining that feelings of anger, hurt, guilt, depression and blame are normal and predictable responses to both the fact of infertility and to the infertility investigation (Born, 1989; Menning, 1980). In addition, information regarding the incidence of infertility may help couples to normalize their experience by realizing that they are not alone, that in fact infertility affects as many as one in six couples. In addition, information about the causes of infertility, as well as details of the diagnostic and treatment procedures involved in the infertility investigation, are very important to many couples in providing a sense of control as they work through their feelings. If the counsellor is not connected with an infertility clinic he or she is unlikely to be well-versed in this information and can be most helpful by providing material or references for clients to read. Books by Menning (1977) and by Pfeffer and Woollett (1983), which explain the procedures in clear, non-technical language are useful resources for recommended reading. Counselling for this group of recently infertile men and women may be most effective 100  Discussion Conclusions  when both partners attend. By modelling acceptance and validating the experience of the individual, the counsellor's acknowledgement of different ways of responding can help to facilitate acceptance by each member of the couple of the validity of their partner's experience (Born, 1989; Hendricks, 1985). Older/ Negative Self-Image: The second identifiable risk group may be described by the  factors which covaried to create a profile of individuals at risk of experiencing high levels of anxiety, obsessive-compulsiveness and depression. The most striking feature of this group is their advanced age. More than half of the men and women contributing to this profile were over the age of 37. This fact alone may represent a significant risk factor if the primary cause for distress is the perception of diminishing time left to successfully reproduce. Two additional factors contributed strongly to the profile, however, providing a more complete picture of the individual at risk. The men and women in this group reported very negative self-appraisals. In addition to scoring low in self-esteem, a majority of these people rated themselves very low on a wide range of personality characteristics typically described as masculine or feminine. In other words, these men and women appeared reluctant or unable to identify themselves as possessing any gender-identity related strengths or weaknesses, as being essentially lacking in salient gender traits. As with age, a poorly defined self-image and low self-esteem may be sufficient criteria to place these infertile individuals at risk of having difficulties coping with infertility. The results suggest, however, that these factors tend to covary and therefore may be related in some meaningful way. Whether these are individuals who have passively 'drifted' through adulthood, or who have actively postponed their own development in favor of the expected parenthood role, they appear to lack a positive sense of identity. The intense feelings of anxiety, agitation and depression these individuals exhibit may be a result of having arrived near the end of their reproductive years threatened with the loss of what may be perceived as the only role still available to them. For these people, the perceived loss of the ability to become parents may represent, in a very real sense, a complete loss of identity. 101  Discussion Conclusions  The goals of the counsellor working with such clients would be complex and multi-level. In most cases, individual therapy would be recommended, although work including the partner or other family members may be useful at various point of the therapy. Initially, therapy may resemble the supportive counselling described above, as the counsellor builds a relationship, validates the client's experience, and facilitates the expression of feelings of anger and grief. As therapy progresses, however, it will be necessary for the counsellor to become more active and to focus more on helping the client to build self-esteem and a realistic self-image. This may involve an accurate assessment of the individual's strengths and capabilities as well as an examination of his or her needs, hopes and dreams. The use of exercises and techniques which promote selfawareness and challenge negative self-appraisals may be useful. As the individual begins to identify his or her unique strengths and capabilities, the counsellor may begin to encourage exploration of the motivations for becoming a parent. Motives for parenthood may be simple or complex and may include the individual's desire to meet his or her own needs for ego expansion, dependency, affiliation, conformity, status, occupation, virility or a sense of purpose, as well as more selfless motives as the desire to care for and protect a helpless infant or a desire to create something worthwhile and permanent (Pohlman, 1970; Turner & Helms, 1983; Sawatzky, 1981). An examination of motives for parenthood may be extremely painful for those individuals who have low self-esteem and a poorly defined self-image, and should therefore be approached with extreme care. This task is likely to raise some of the most challenging issues regarding identity, prompting questions such as "If I can't be a mother/father, who or what am I?" As difficult as it is, the examination of motives is an essential step in identifying the individual's needs and is a prerequisite to the next step of generating alternative ways of meeting those needs (Daniluk, 1991; Mahlstedt, 1987). Dependent on the outcome of the process of evaluation, individuals may need to examine a number of options which meet their needs to a lesser or greater extent. At this point in therapy, it may be desirable to include the partner, as many of the available options may require cooperation or accommodations on the part of either or both partners. 102  Discussion Conclusions  For example, if an individual decides that his/her desire to have a child is primarily motivated by the need for ego expansion, creativity, control or occupation, he/she may choose to find alternative means for meeting these needs, through increased pursuit of career or educational goals, changes in occupation or otherwise. Some may discover that their desire is primarily a response to social pressures, and may choose to live childfree. Many, however, will determine that their needs can best be met through becoming a parent and will choose to pursue adoption. For older couples, however, this option may also be fraught with difficulties. Although often seen as the end of the infertility saga, the decision to adopt is often just the beginning of a new episode of the experience. At the minimum, the initial decision to adopt leads to a variety of further considerations, including whether the age, sex, race and health of the adopted child are important, whether to adopt privately or through public agencies, and the nature and extent of the relationship desired with the birth mother. In addition, many experience the adoption process as highly stressful and experience a resurgence of feelings of grief as the finality of the infertility is underscored. For those who have not effectively resolved feelings regarding their identity as an infertile person, adoption may be seen as a confirmation of their inadequacy and failure. The process of adoption also contains reflections of many of the issues raised by infertility, and couples often find feelings of anger and loss of control being reactivated in response to the probings and scrutiny of the adoption worker (Hendricks, 1985). The adoption investigation, conducted with the purpose of ascertaining the appropriateness of the couple as adoptive parents, is often referred to as intrusive and insulting, as the adoption worker attempts to determine whether the couple have effectively resolved their feelings about infertility. As offensive as this indignity may seem, the purpose may be a valid one. There is some evidence that ineffective or incomplete resolution of feelings regarding infertility may have detrimental effects on future parenting. Burns (1990) found that parents who had been treated for infertility were significantly more likely to exhibit disturbed patterns of parenting, including overprotectiveness, abuse and neglect. These parents included those who had adopted as well as those who had achieved pregnancies through IVF or AI. 103  Discussion Conclusions  The counsellor's role in this final phase of therapy is multidimensional and idiosyncratic. For those clients who have resolved their feelings surrounding infertility and parenthood more or less effectively, therapy will consist primarily of facilitating the decision-making process. This may be achieved by using exercises and counselling techniques which help couples to generate options, evaluate them in light of available resources, prioritize and select those which have the greatest potential to both meet their needs and be successful. Those who continue to struggle with issues regarding infertility and identity will nevertheless need to make decisions. For these individuals, the process of decision-making will be most effective if it is integrated with repeated returns to a focus on self-esteem and identity development.  Implications for Researchers In this final section, a discussion of implications and recommendations for researchers is presented. The purpose of this section is to present a brief review of some of the materials and methodology utilized in the present study, with a view to evaluating the efficacy and usefulness of each. In addition, the author presents a brief discussion regarding recommendations for possible directions for future investigations in the study of individual responses to infertility.  Review of Methods and Materials The present investigation was designed to examine how personality, demographic and situational characteristics of the individual may affect the nature of his or her responses to the experience of infertility. Despite, or perhaps because of, an ambitious research design, including a large number of elements for consideration, the findings represent only a glimpse of the "extraordinarily complex reality that accompanies and defines the capacity of human beings to cope with change in their lives" (Schlossberg, 1981, p.3). In this context, the present study may be regarded as exploratory and, it is hoped, as a contribution to a growing body of research designed to clarify and describe the complex nature of how personality and situational factors define the capacity of men and women to cope with infertility. In evaluating the present attempt, an important point of concern may be raised regarding the 104  Discussion Conclusions  use of a single trial measure of a transition which may take many months or years to complete. As a one-time-only sampling, the study essentially represented a 'snapshot' of the state and functioning of the subjects at a particular point in time. The primary focus of the present study, however, was on the characteristics of the individual such as gender, self-esteem and sex-role identity, as well as characteristics of the infertility which may affect the way individuals respond to infertility. There is reason to assume that the measures used to assess individual characteristics are stable, (ie. that Subject X would score the same or very similar on, for example, the measure of self-esteem, if taken one week, one month or one year from the original test date). Repeated use of the standardized measures used in the study has indicated fairly high reliability (Bern, 1974; Coopersmith, 1981; Rotter, 1966). An important consideration, however, is the apparent assumption of stability of the levels of distress measured. Within the context of the infertility transition, an experience which has been repeatedly described as one with many fluctuations in mood and affect, it is a potential point of argument whether the results achieved are more representative of the global functioning level of the individuals surveyed, or of a temporary state. To achieve a more comprehensive understanding of the mediating effects of personality and situational variables on individuals responses to infertility, it is suggested that future research studies make greater efforts to control for the changeable nature of the infertility transition. This may be done in one of two ways. One possible design is to selectively sample individuals exclusively at a specific step of the infertility work-up, (ie. immediately post-diagnosis). Such a design would likely necessitate a lengthy period of data collection, as it may take several months to identify and contact a sufficiently large sample. In the event that such a time commitment is possible, it is arguable that time and efforts may be more profitably spent in employing a longitudinal or repeated measures approach, such as was employed by Daniluk (1988) or Edelman, Connolly and Robson (1989). In such a design, infertile individuals would be surveyed at specified intervals throughout the infertility work-up. In this way, the assumed stability of the individual's personal and situational factors may be confirmed or disconfirmed. Such a design 105  Discussion Conclusions  may provide more complete information regarding who may be most at risk of emotional distress in response to infertility, as well as whether different types of individuals are more vulnerable at different stages of the transition. Finally, comment must be made regarding the use of the Adjustment to Infertility Scale (Sabatelli, et al., 1988) as an outcome measure. This instrument was obtained with permission from the designer prior to having been substantiated through use in published research other than by its author. As such, use of the instrument in the present study may be considered a trial, and therefore requires comment regarding its validity as well as its applicability as a research tool. The validity of the instrument may be gauged, in part, by the degree to which it correlates with measures designed to assess constructs of a similar nature. Within this context, it may be expected that higher levels of adjustment to infertility may be related to decreased evidence of symptomatic distress in response to infertility. To this end, correlations with SCL-90-R scores, for both males and females, were examined. No relationship was seen between AIS scores and global SCL scores for men. For women, however, a mild negative correlation (r= -.35) was shown between AIS scores and SCL global scores, indicating a slight tendency for women who indicated lower global distress levels to report higher levels of adjustment to infertility. An examination of correlations between AIS scores and scores for individual subscales of the SCL-90-R suggested a pattern of symptoms which may be related to adjustment to infertility. For women, mild negative correlations were observed between MS scores and scores on the SCL subscales of Interpersonal Sensitivity ( r = -.42), Depression (r = -.39) and Phobic Anxiety (r = -.47), suggesting that women who reported higher levels of adjustment to infertility also indicated somewhat lower levels of interpersonal sensitivity, depression and anxiety. For men, the only relationship observed between AIS scores and SCL subscale scores was for Depression (r= .3), suggesting that men who reported high adjustment to infertility appeared to also report slightly lower levels of depression. These were only mild correlations, however, suggesting that, although there may be some connection between the levels of distress symptoms and individual's adjustment to infertility, the AIS did not appear to be measuring distress levels. 106  Discussion Conclusions  Of more immediate salience to the needs of the present study was the ability of the AIS to contribute to the picture of how men and women respond to infertility. AIS scores were included in each of the three Canonical analyses, but in each case failed to contribute to the pattern of responses to infertility. This suggests that the AIS may lack the sensitivity to discriminate in a discernible way between individuals who are at risk of poor adjustment to infertility and those who are adjusting well. On the other hand, it may be inappropriate to apply the concept of 'adjustment' in a study examining individuals who are in the process of undergoing the infertility work-up. It may be that adjustment is a state only achievable by individuals who have terminated the infertility investigation and who are in the process of evaluating and considering alternatives to biological parenthood. In this context, it may be suggested that the Adjustment to Infertility Scale may best be utilized as an assessment tool in working with subjects seeking adoption or other alternatives. Thus, although the AIS may be of use for research assessing responses to infertility, it was not found to provide and useful information in the present study. It is suggested however, that the AIS may be useful both as a clinical and a research tool for working with individuals preparing to move through the final stages of the infertility transition, to gauge the success with which the transition was made.  Directions for Future Research While the present study demonstrated two clear links between characteristics of the individual and responses to infertility, the data also provided suggestive evidence of a handful of relationships which may be indicative of possible directions for further study. Chief among the relationships suggested by the non-significant findings was the observed correlation between actual and perceived duration of infertility and distress levels. As demonstrated by Daniluk (1988) and Sabatelli et al. (1988), increased duration of infertility appears to be related to fewer symptoms of distress reported. The failure of this relationship to be shown significantly in the present study may indicate that duration of infertility is less useful in predicting distress levels than are other variables. However, the fact that duration has been demonstrated as a significant predictor of  107  Discussion Conclusions  distress in other studies, but not here, suggests that the relationship between these variables may be moderated by a third, as yet unknown, factor. To clarify this relationship, these questions warrant further attention. A second potential area for future investigations was suggested by the observed pattern of scores for individuals in the different sex-role categories. When scores on the new canonical variates of psychosocial security, emotional calmness, negative self-image and anxiety were compared, the data suggested that men and women with masculine sex-role identities, as well as those people with undifferentiated sex-role identifies, appeared to score markedly low on the outcome variable representing emotional calmness. This non-significant trend appeared to suggest the possibility that men and women who identify with masculine traits may have actually exhibited heightened levels of anger and irritability as well as increased incidence of somatic symptoms such as headaches, chest pains, nausea, dizziness and muscle aches. These trends may reflect social role expectations for masculinity. O'Neil (1981) notes that men are traditionally socialized to restrict the direct expression of grief and sadness, which may be seen as signs of 'emotional weakness'. In their place, many men are taught that expressions of aggressiveness, including anger and hostility, are the legitimate outlet for the masculine male's feelings (O'Neil). Therefore, men and women who identify with the masculine ideal, as suggested by high masculinity scores, may be more likely to respond to the stress of infertility with expressions of anger and hostility, rather than with sadness or grief. Such a restriction or redirection of emotion may result in a buildup of tension which could lead to somatic symptoms, such as headaches, muscle and chest pains, nausea and dizziness. The evidence for a relationship between masculine sex-role identity and heightened levels of anger and irritability in response to infertility is only suggestive in the present data. However, if such a relationship between sex-role identity and responses to infertility can be substantiated, its implications are many. In addition to increasing the counsellor's ability to identify men and women who are at risk of experiencing difficulty with infertility, the identification of such a relationship may serve to further our understanding of men's and women's social roles. 108  Discussion Conclusions  Moreover, it may be useful to discover whether responses such as hostility and somatization are specific to infertility, or whether these symptoms represent a typical response pattern for masculine individuals. Further investigation will be required to answer these and other questions. Conclusion  The purpose of the present study was to identify characteristics of the individual and characteristics of the infertility transition which may help counsellors and other helping professionals to understand the differences in individual responses to infertility, as well as to identify those individuals who may potentially be at risk of experiencing heightened distress and crisis feelings in response to infertility. In this regard, the study has achieved its goals. Two significant links were observed between the characteristics of the individual and the measures of distress symptoms in response to infertility. These links appear to indicate that individuals with a positive sense of self and a secure and comfortable lifestyle respond with fewer symptoms of distress, whereas men and women with low self-esteem and a poorly defined self-image appear to have more difficulty coping with infertility.  109  References  REFERENCES  Adams, J., Hayes, J. & Hopson, B. (1976). Transition: Understanding and managi_ng personal change. London: Martin Robinson. Adler, J.D. & Boxley, R.L. (1985). The psychological reaction to infertility: Sex-roles and coping styles, Sex Roles, 12, 271-279. Bell, S. (1981). Psychological problems among patients attending an infertility clinic, Journal of Psychosomatic Research, 25, 1-3. Bern, S.L. (1974). Bem sex-role inventory: Professional manual. Palo Alto: Consulting Psychologists Press, Inc. Bern, S.L. (1974b). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162. Bern, S.L. (1976). Testing the promise of androgyny. In, A. Kaplan & J.P. Bean (Eds) Beyond gsisatolgt x-i . amizacacuzugaKcapsydiduy_sgandrsayiv. Boston: Little, Brown & Co. Berger, D.M. (1980). Impotence following the discovery of azoospermia. Fertility and Sterility, 24, 154-156. Born, R.R. (1989). Adlerian perspectives on counseling the infertile couple. Individual Psychology, 4, 447-458. Bowers, N.A. (1985). Early pregnancy loss in the infertile couple. Journal of Obstetric, Gynaecological and Neo-Natal Nursing, (Nov/Dec Supplement). 55s-57s. Brammer, L.M. (1991). How to cope with life transitions: The challenge of personal change. New York: Hemisphere Publishing Corporation. Brand, H.J. (1989). The influence of sex differences on the acceptance of infertility. Journal of Reproductive and Infant Psychology, 2, 129-131.  Bryant, H. (1990). it -	 - • I I 1 UAL. Canadian Advisory Council on the Status of Women. .  -  .  II  n. Ottawa:  Burns, L. H. (1990). An exploratory study of perceptions of parenting after infertility. Family Systems Medicine, a, 177-189. Callan, V.J. & Hennessey, J.F. (1988). The psychological adjustment of women experiencing infertility. British Journal of Medical Psychology, 6.1, 137-140. Chiappone, J.M. (1984). Infertility as a nonevent: Impact. coping and differences between men and women. University of Maryland: Unpublished Doctoral Dissertation. Coopersmith, S. (1981). SEI: Self-esteem inventories. Palo Alto, California: Consulting Psychologists Press.  110  References  Daniels, K.R., Gunby, J., Legge, M., Williams, T.H. & Wynn-Williams, D.B. (1984). Issues and problems for the infertile couple. New Zealand Medical Journal, Mar 28, 1984, 185187. Daniluk, J.C. (1991). Strategies for counseling infertile couples. Journal of Counseling & Development, fa, 317-320. Daniluk, J.C. (1988). Infertility: Intrapersonal and interpersonal impact. Fertility and Sterility, 42, 1-9. Daniluk, J.C., Leader, A. & Taylor, P.J. (1987). Psychological and relationship changes of couples undergoing and infertility investigation: Some implications for counsellors. Dritish Journal of Guidance and Counsellin_g,15, 29-36. Davis, D.C. (1987). A conceptual framework for infertility. Journal of Obstetrics, Gynaecological and Neo-Natal Nursing, (Jan/Feb) 30-35. Debrovner, C.H. & Shubin-Stein, R. (1976). Sexual problems associated with infertility. Medical Aspects of Human Sexuality, (Mar). 161-162. Derogatis, L.R. (1975). The SCL-90-R. Baltimore: Clinical Psychometrics Research. Derogatis, L.R. (1983). Description and bibliography for the SCL-90-R. Baltimore: Johns Hopkins University School of Medicine. Deutsch, H. (1947). The psychology of women: A psychoanalytic perspective. Vol 2: Motherhood. London: Research Books. Dixon, W.J., Sampson, P. & Mindle, P. (1988). One- and two-way analysis of variance with data screening. In, W.J. Dixon, M.B. Brown, L. Engelman, M.A. Hill & R.I. Jennrich (Eds). BMDP statistical software manual. Berkeley: University of California Press. Domar, A.D., Seibel, M.M. & Benson, H. (1990). The mind/body program for infertility: A new behavioral treatment approach for women with infertility. Fertility and Sterility, 53, 246-249. Drake, T.S. & Grunert, G.M. (1979). A cyclic pattern of sexual dysfunction in the infertility investigation. Fertility and Sterility, 32, 542-545. Edelmann, R.J., Connolly, K.J. & Robson, J. (1989). The impact of infertility and infertility investigations: Four case illustrations. Journal of Reproductive and Infant Psychology, 1, 113-119. Eisner, B.G. (1963). Some psychological differences between fertile and infertile women. Journal of Clinical Psychology, 12, 391-392. Elstein, M. (1975). Effect of infertility on psychosexual function. British Medical Journal, 2, 296299. Feuer, G.S. (1983). The psychological impact of infertility on the lives of men. University of Pennsylvania: Unpublished Doctoral Dissertation. Fouad, N. A. & Fahje, K.K. (1989). An exploratory study of the psychological correlates of infertility on women. Journal of Counseling and Development, h.a, 97-101. 111  References  Frane, J. (1988). Canonical correlational analysis. In, W.J. Dixon, M.B. Brown, L. Engelman, M.A. Hill & R.I. Jennrich (Eds), BMDP statistical software manual, Vol 2. Berkeley: University of California Press. Frias, A. & Wilson, S. (1985). When biological childlessness is inevitable. Medical Aspects of Human Sexuality, 12, 43-51. Graham, M.A., Thompson, S.C., Estrada, M. & Yonekura, M.L. (1987). Factors affecting psychological adjustment to a fetal death. American Journal of Obstetrics and Gynaecology, 151, 254-257. Halpern, S. (1989). Infertility: Playing the odds. Ms. Magazine, (Jan/Feb) 147-152. Harris, A.B. (1989). Examining infertility as a crisis: A comparison between two groups of infertile women based on treatment status. Unpublished Doctoral Dissertation, The Fielding Institute. Hendler, N., Derogatis, L.R., Avella, J. & Long, D. (1977). EMG Biofeedback in patients with chronic pain. Diseases of the Nervous System, 31, 505-509. Hendricks, M. C. (1985). Feminist therapy with women and couples who are infertile. In, L.B. Rosewater & L.E.A. Walker (Eds), Handbook of feminist therapy. New York: Springer. Hoff, L. A. (1989). pggple in crisis: Understanding and helping (3rd ed.). Redwood City: Addison-Wesley Publishing Company. Hopson, B. (1981). Response to the papers by Schlossberg, Brammer and Abrego. :the Counselling Psychologist, E 36-39. Humphrey, M. (1977). Sex differences in attitudes to parenthood. Human Relations, 31, 737749. Kedem, P., Mikulincer, M. & Nathanson, Y.E. (1990). Psychological aspects of male infertility. British Journal of Medical Psychology, 	73-80. Kelly, J.A. & Worell, J. (1977). New formulations of sex roles and androgyny: A critical review. Journal of Consulting and Clinical Psychology, 4, 1101-1115. Kraft, A.D., Palombo, J., Mitchell, D., Dean, C., Meyers, S. & Wright Schmidt, A. (1980). The psychological dimensions of infertility. American Journal of Orthopsychiatry, 14, 618628. Laos, A., Lalos, 0., Jacobsson, L. & VonSchoultz, B. (1986). Depression, guilt and isolation among infertile women and their partners. Journal of Psychosomatic Obstetrics and Gynaecology, 5., 197-206. Leader, A. & Taylor, P.J. (1984). Diagnosis and medical management of infertility. Alberta: Unpublished Paper, University of Calgary. Leader, A.; Taylor, P.J. & Daniluk, J.C. (1984). Infertility: Clinical and psychological aspects. Psychiatric Annals, 14, 461-467.  112  References  Mahlstedt, P. (1985). The psychological component of infertility. Fertility and Sterility, 4, 335346. Matthews, R. & Matthews, A. (1986). Infertility and involuntary childlessness: The transition to nonparenthood. journal of Marriage and the Family, /a, 641-649. MacNab, R.T. (1984). Infertility and men: A study of change and adaptive choices in the lives of involuntarily childless men. California: Unpublished Doctoral Dissertation, The Fielding Institute. Maccoby, E.E. & Jacklin, C.N. (1974). The psychology of sex differences. California: Stanford University Press. Matthews, R. & Matthews A.M. (1986). Infertility and involuntary childlessness: The transition to nonparenthood. Journal of Marriage and the Family, zia, 641-649. Mazor, M. (1979). Barren couples. Psychology today, (Feb). 101-108. McBain, J.C. & Pepperell, R.J. (1987). Unexplained infertility. In R.J. Pepperell, B. Hudson & C. Wood (Eds), The infertile couple (2nd ed.),(p. 208-232), Melbourne: Churchill Livingstone. McCormick, T.M. (1980). Out of control: one aspect of infertility. Journal of Obstetric, Gynecologic and Neonatal Nursing, 2 205-206. McEwan, K.L., Costello, C.G. & Taylor, P.J. (1987). Adjustment to infertility. journal of Abnormal Psychology, 2.6, 108-116. Menning, B.E. (1977). Infertility: A guide for the childless couple. New Jersey: Prentice Hall. Menning, B.E. (1980). The emotional needs of infertile couples. Fertility and Sterility, 31, 313319. Menning, B.E. (1982). The psychological impact of infertility. Nursing Clinics of North America,12, 155-163. Morse, C. & Dennerstein, L. (1985). Infertile couples entering an in vitro fertilization programme: A preliminary survey. journal of Psychosomatic Obstetrics and Gynaecology, 4, 207-219. Murray, J. & Callan, V.J. (1988). Predicting adjustment to perinatal death. British Journal of Medical Psychology, 61, 237-244. Noyes, R. & Chapnick, E. (1964). Literature on psychology and infertility. Fertility and Sten4v, 543-558. Olshansky, E.F. (1987). Infertility and its influence on women's career identities. Health Care for Women International, a, 185-196. Olshansky, E.F. (1987b). Identity of self as infertile: An example of theory-generating research. Advances in Nursing Science, E 54-63. O'Neil, J. (1981). Male sex role conflicts, sexism, and masuclinity: Psychologial implications for men, women, and the counseling psychologist. The Counseling Psychologist, E 6180. 113  References  Pantesco, V. (1986). Nonorganic infertility: some research and treatment problems. Psychological Reports, a, 731-737. Pohlman, E. (1970). Childlessness, intentional and unintentional: Psychological and social aspects. The Journal of Nervous and Mental Disease, 151, 2-12. Reading, A.E., Chang, L.C. & Kerin, J.F. (1989). Psychological state and coping styles across an IVF treatment cycle. Journal of Reproductive and Infant Psychology, 1, 95-103. Rotter, J.B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: CjmgAILs:L1212ggi lrilA , $Q, 1-27. Russo, N.F. (1979). Overview: Sex roles, fertility and the motherhood mandate. Psychology of Women Quarterly, 4, 7-15. Sabatelli, R.M., Meth, R.L. & Gavazzi, S.M. (1988). Factors mediating the adjustment to involuntary childlessness. Family Relations, 31, 338-343. Sandelowski, M. & Jones, L.C. (1986). Social exchanges of infertile women. Issues in Mental Health Nursing, 173-189. Sandelowski, M. & Pollock, C. (1986). Women's experiences of infertility. Image: Journal of Nursing Scholarship, la, 140-144. Sawatzky, M. (1981). Tasks of infertile couples. Journal of Obstetrics. Gynaecological and NeoNatal Nursing, (Mar/Apr), 132-133. Schlossberg, N.K. (1981). A model analyzing human adaptation to transition. The Counselling Psychologist, E 2-18. Schlossberg, N.K. (1981b) Reactions to reactions. The Counselling Psychologist, E 49-50. Schlossberg, N.K. (1984). Counseling Adults in Transition: Linking Practice with Theory, New York: Springer Publishing Company. Seibel, M.M. & Taymor, M.L. (1982). Emotional aspects of infertility. Fertility and Sterility, 12, 137-145. Short, J. (1989). Infertility tales. Self Magazine, (May), 188-192. Sockne, D. & Forsythe, A. (1988). T-tests. In, W.J. Dixon, M.B. Brown, L. Engelman, M.A. Hill & R.I. Jennrich (Eds). BMDP statistical software manual. Berkeley: University of California Press. Snarey, J., Son, L., Kuehne, V.S., Hauser, S. & Vaillant, G. (1987). The role of parenting in men's psychosocial development: A longitudinal study of early adulthood infertility and midlife generativity. Developmental Psychology, 22, 593-603. Snarey, J. (1988). Men without children. Psychology Today, (Mar), 61-62. Spence, J.T., Helmreich, R. & Stapp, J. (1975). Ratings of self and peers on sex role attributes and their relation to self-esteem and conceptions of masculinity and femininity, Journal of Personality and Social Psychology, 31, 29-39.  114  References  Tabachnick, B.G. & Fidel!, L.S. (1983). Using multivariate statistics. New York: Harper & Row. Trepanier, K. (1985). Infertile couples: Alone in a crowd. The Canadian Nurse, al, 42-45. Turner, J.S. & Helms, D.B. (1983). Lifespan development (2nd ed.), New York: Holt, Rinehart & Winston. Tyler, F. (1978). Individual psychosocial competence: A personality configuration. Educational and Psychological Measurement, 31, 309-323. Valentine, D.P. (1986). Psychological impact of infertility: Identifying issues and needs. Social Work in Health Care, 11, 61-69. Veevers, J.E. (1980). Childless by choice. Toronto: Butterworths. Walker, H.E. (1978). Psychiatric aspects of infertility. Urologic Clinics of North America, 1, 481-488. Walker, H.E. (1978b). Sexual problems and infertility. Psychosomatics, 12, 477-484. Wiehe, V.R. (1976). Psychological reaction to infertiltiy. Psychological Reports, ,'i$, 863-866. Wiehe, V.R. (1976b). Psychological reactions to infertility: implications for nursing in resolving feelings of disappointment and inadequacy. journal of Obstetric. Gynecologic and Neonatal Nursing, ¢, 28-32. Woollett, A. (1985). Childlessness: Strategies for coping with infertility. International Journal of Behavioral Development, a, 473-482. Woollett, A. & Pfeffer, N. (1983). The experience of infertility. London: Virago Press.  115  APPENDIX A  PERSONAL INFORMATION SURVEY The following questions address some facts about you. Please select the answer which most closely applies to you. Indicate your answers in the spaces provided, or by circling the appropriate letter. 1. Age:  2. Sex: a) Male  b) Female  3. What is your primary ethnic/cultural background?  3a. Is having children an important value in your family/ cultural background? a) Yes b) No 4. Do you subscribe to the values of a particular faith/religion? a) Yes b) No If yes, please specify 5. Family's yearly income:  a) Up to $15000/year b) Between $15000 - $25000/year c) Between $25000 - $40000/year d) Over $40000/year  6. Education completed:  a) elementary school b) high school c) college or university d) post-graduate education  116  7. General physical health:  a) Excellent b) Good c) Fair d) Poor  8. How long (in months) have you and your partner been attempting to conceive a child? 9. How long ago (in months) did you first choose to seek medical attention for your fertility concerns? 10. Have you at present received a diagnosis regarding your fertility concerns?  11. If yes, is your diagnosis:  a) Yes  b) No  a) Male factor b) Female factor c) Couple factor d) Unknown  12. Do you see your present fertility status as: a) a temporary condition b) a permanent condition c) very uncertain 13. In your own assessment, how do you rate your chances of achieving a successful pregnancy? a) less than 25% Chance b) between 25% and 50% chance c) between 50% and 75% chance d) better than 75% chance  117  14. Prior to this time, have you been faced with any life experiences of a similar nature? (eg. termination of a pregnancy, significant personal loss, serious illness, etc.) a) Yes b) No 14a. If yes, please briefly describe:  14b. How would you rate the long term effects of this previous experience on your life? a) completely or mostly positive b) both positive and negative c) completely or mostly negative  118  APPENDIX B  Research Referral  The following patients have expressed an interest in participating in the research project "Individual Responses to Infertility", conducted by Stephanie Koropatnick.  Referring Physician  Date  119  APPENDIX C  (Form A) Individual Responses  Infertility  Thank-you for agreeing to participate in our research project. In this study, we hope to learn more about how different types of people respond to the experience of infertility. The information you provide us with will help counsellors and otner helping professionals to better understand the many emotions associated with infertility. In turn, counsellors will be better equipped to help others who face the experience you are currently dealing with. You are asked to complete the five questionnaires included in this envelope in the order in which they are presented to you. There are no right or wrong answers, so please feel free to respond in a straightforward and spontaneous manner. Same of the questionnaires may continue onto the reverse side, so please be certain to complete both sides of the forms, where applicable. Also, it is important that you do not consult with your partner or compare answers while completing the questionnaires, although you may feel free to discuss your impressions of the questions after the questionnaires have been completed and returned. Each questionnaire will take between five and fifteen minutes to complete. In total, your participation in this study will take approximately one hour. When the five questionnaires have been completed, please place all of the forms (excluding Forms A & B) into the stamped, pre-addressed envelope provided, and return this at your earliest possible convenience. All information, including your responses on all questionnaires will be kept in the strictest confidence. In fact, we ask that you not place your name on any of your response forms. Your answers will be identified only by a code number. Please remember that you are free to refuse to participate, or to withdraw your consent at any time, without jeopardizing your medical treatment. If you have any further questions regarding the study, please feel free to contact the researcher, Stephanie Koropatnick at (604) 7 33-2334, or Dr. Judith Daniluk (604-228-5768). Thank-you again for your participation. Sincerel  Stephan oropa c/o Dept. of Comselling Psychology, University of B.C.  120  APPENDIX D  (Form B) Consent Form   , voluntarily agree to participate in the study "Individual Responses to Infertility", being conducted through the Dept of Counselling Psychology at U.B.C. I further agree to grant the researcher, Stephanie Koropatnick, confidential access to information contained in my medical records maintained at the Fertility clinic. I have been informed of the purposes of the study, and have been given an opportunity to raise questions regarding it. I have received copies of Form A, explaining the study, and of Form B, granting my informed consent. I understand that I will be asked to complete six brief questionnaires and that no further requests will be made of me at this time. I have been assured that all information, including my questionnaire responses, will be kept in the strictest confidence. I understand that my participation in this study will in no way jeopardize the course of my medical treatment and that, in fact, no medical personnel at the clinic are directly involved in conducting this research project. In addition, I recognize that I may choose to withdraw my participation at any point without prejudicing my present or future medical treatment.  (Signature)  (Name - please print) (Address)  (Telephone)  121  APPENDIX E  September 29, 1989  Dear  In reviewing our files, it has come to my attention that it has been one month since you first received the questionnaire package for the research project "Individual Responses to Infertility". We value your input into this important issue and we are awaiting your response. If you have not already done so, please complete the queStionnaires and return them in the envelope provided at your earliest possible convenience. If you have already completed the questionnaire package and returned it, please disregard this notification. In any event, I wish to take this opportunity again to thank-you for your participation in this research project. Your assistance is a valuable contribution to the increased understanding of the experience of infertility.  Yours very truly,  Stephanie Koropatnick Dept. of Counselling Psychology University of B.C. 5780 Toronto Road, Vancouver, B.C. V6T 1L2  122  Appendix F (Schlossberg, 1981, p.5)  Schlossberg's Model of Adaptation to Transition TRANSITION  Event or nonevent resulting in change or assumption Change of social networks Resulting in growth or deterioration  PERCEPTION OF THE PARTICULAR TRANSITION  CHARACTERISTICS OF PRE-TRANSITION & POST-TRANSITION ENVIRONMENTS  CHARACTERISTICS OF THE INDIVIDUAL  Role Change: gain or loss Aspect: positive or negative Source:internal or external Timing:on-time or off-time Onset: gradual or sudden Duration: permanent, temporary or uncertain  Internal Support Systems Intimate relationships Family unit Network of Friends Institutional Supports Physical Setting  Psychosocial Competence Sex and Sex-Role Identification Age (and Life Stage) State of Health Race/Ethnicity Socioeconomic Status Value Orientation (religiosity) Previous Experience with a transition of a similar nature  44 ADAPTATION Movement through phases following transition: pervasiveness through reorganization Depends on: 1) Balance of individual's resources and deficits 2) Differences in pre- and post transition environments re perception, supports and individual  123  BIOGRAPHICAL INFORMATION  NAME:  Stephanie Jean Konopatnick  MAILING ADDRESS: 3167 West 5th Avenue  Vancouver, B.C. V6K 1V1 PLACE AND DATE OF BIRTH: New Weztminatek, B.C. Aptit 26, 1957 EDUCATION (Colleges and Universities attended, dates, and degrees):  University o4 B.C. 1987 - 1991: Mcatek of Ants Simon Fnasen Univensity, 1980 1984: Bachetok o4 Dougta6 Cottege, 1976 - 1978 -  POSITIONS HELD:  reacher: Vancouver Schoot Board  PUBLICATIONS (if necessary, use a second sheet):  AWARDS:  Complete one biographical form for each copy of a thesis presented to the Special Collections Division, University Library. DE•5  Ants  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items