Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Perfectionism in university students from Chinese and European cultural backgrounds : an investigation… Varey, Christine Anne 1999

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

Item Metadata

Download

Media
831-ubc_1999-0398.pdf [ 6.98MB ]
Metadata
JSON: 831-1.0053940.json
JSON-LD: 831-1.0053940-ld.json
RDF/XML (Pretty): 831-1.0053940-rdf.xml
RDF/JSON: 831-1.0053940-rdf.json
Turtle: 831-1.0053940-turtle.txt
N-Triples: 831-1.0053940-rdf-ntriples.txt
Original Record: 831-1.0053940-source.json
Full Text
831-1.0053940-fulltext.txt
Citation
831-1.0053940.ris

Full Text

PERFECTIONISM IN UNIVERSITY STUDENTS FROM CHINESE AND EUROPEAN CULTURAL BACKGROUNDS: AN INVESTIGATION OF CONSTRUCT VALIDITY by CHRISTINE ANNE VAREY BScfHon.), QUEEN'S UNIVERSITY, 1990 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 standard • <S • THE UNIVERSITY OF BRITISH COLUMBIA JULY 1999 © CHRISTINE VAREY, 1999 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. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract Perfectionism has been related to both adaptive functioning such as positive achievement striving, as well as to negative outcomes such as procrastination and depression. Numerous studies, using primarily Caucasian subjects, document a relationship between multidimensional perfectionism and depression. Few studies have examined perfectionism in samples from different cultural backgrounds. The present multidimensional conceptualization and operational definition of perfectionism remains to be tested in terms of meaningfulness and applicability to other cultural groups. Therefore the overall aim of the current study is to determine if the Multidimensional Perfectionism Scale (Hewitt & Flett, 1991) shows evidence of construct validity in a sample of university students from Chinese cultural backgrounds. Evidence of construct validity is obtained by investigating similarities and differences between the two cultural groups in the following: the internal structure of the measure; and the relationships between the measure and other measures (concurrent validity). Differences in mean levels of perfectionism between Chinese-Canadian and European Canadian students are also investigated. Data for 191 subjects are reported for the following measures: Multidimensional Perfectionism Scale (Hewitt & Flett, 1991); Multidimensional Perfectionism Scale (Frost et al, 1990); Beck Depression Inventory (Beck et al, 1979); Suinn-Lew Asian Self-Identity Acculturation Scale (Suinn et al, 1987); a Background Information Sheet. Evidence of construct validity and of the previously documented relationship between socially-prescribed perfectionism and depression was found for the Chinese-Canadian sample. The current study lends support for using the Multidimensional Perfectionism Scale (Hewitt & Flett, 1991) with Chinese-Canadian university students. TABLE OF CONTENTS Page ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES ix ACKNOWLEDGEMENTS xii CHAPTER I INTRODUCTION 1 Background 1 Purpose 3 Implications 4 CHAPTER II LITERATURE REVIEW 5 Evolution of the Perfectionism Construct 5 Early Theories 5 Antecedents of Perfectionism 8 Perfectionism Measures 11 Psychological Research on Perfectionism 12 Research Related to the Adaptive Elements of Perfectionism 16 Research on Perfectionism and Adjustment Difficulties in University Students 17 Research Related to Perfectionism in Asian-American College Students 23 Relevance and Rationale for Studying Perfectionism in Asian Students 24 Research Questions 26 V CHAPTER IE METHODOLOGY 28 Design 28 Sample 28 Data Collection Procedures 31 Instrumentation 33 Perfectionism Measures 33 Multidimensional Perfectionism Scale (H-MPS - Hewitt & Flett, 1991) 33 Multidimensional Perfectionism Scale (F-MPS- Frost, Marten, Lahart & Rosenblate, 1990) 35 Evidence Regarding the Use of Perfectionism Measures with a Chinese Sample 37 Rational for the Use of Two Perfectionism Measures in the Current Study 38 Beck Depression Inventory 39 Evidence Regarding the Use of the BDI with a Chinese Student Sample 39 Suinn-Lew Asian Self-Identity Acculturation Scale 40 Background Information Sheet 41 Analytic Procedures 41 Hypotheses 42 Hypothesis 1: 42 Hypothesis 2 43 Hypothesis 3: 45 CHAPTER IV RESULTS 46 Hypothesis 1 46 Hypothesis 2 52 Hypothesis 3 56 Preliminary Analyses 56 Post Hoc Analyses 59 Hypothesis 4: 65 Hypothesis 5: 67 Hypothesis 4b: 67 CHAPTER V DISCUSSION 75 Internal Evidence of Construct Validity 75 Internal Consistency Reliability 75 Subscale Intercorrelations 76 Summary of the Internal Evidence of Construct Validity 79 External Evidence of Construct Validity 79 Relationship Between Perfectionism and Depression 79 Relationship Between Two Measures of Perfectionism 83 Summary of External Evidence of Construct Validity 85 Comparisons of Mean Levels of Perfectionsim 86 Summary of Findings Related to Comparisons of Mean Perfectionism Levels 87 Post Hoc Comparisons 87 Relationship Between Socially-prescribed Perfectionism and Depression 87 Comparison of Mean Levels of Perfectionism based on Generation Status 88 vii Possible Explanations of the Differences found between First and Second Generation Chinese Samples 90 Comparison of Mean Levels of Perfectionism based on Acculturation Level 91 Summary of Findings Related to Generation Status and Acculturation Level 933 Findings Related to Socially-prescribed Perfectionism 944 Conclusions 955 Delimitations/Limitations 977 Suggestions for Future Research 988 Implications of the Findings for Counselling University Students 1011 REFERENCES 1033 APPENDIX A 113 Ethnic Origin Categories Used in the Current Study 1134 APPENDIX B 115 Characteristics of the Chinese and European Samples 1156 APPENDIX C 117 Letter to Professors 1178 Information Sheet on Research Study 1179 Poster 11721 APPENDIX D 122 List of U.B.C. Classes Entered to Recruit Participants 1233 APPENDIX E 124 Pictures of the Website Used by Participants to Complete Questionnaires 1245 APPENDIX F 126 Vlll Questionnaire Package 1267 APPENDIX G 1 4 1 Perfectionism Means Reported in the Current Study and in Previous Studies 1412 ix List of Tables Table Page 1 Characteristics of the Chinese and European samples 115 2 Comparison of internal consistency for the Chinese and European samples 48 3 Pearson-product-moment correlations among the H-MPS subscales for the Chinese and European samples 50 4 Comparisons of magnitudes of the intercorrelations among the H-MPS subscales and rank orders of the subscales for the Chinese and European samples 50 5 Pearson-product-moment correlations between the H-MPS total scale and the H-MPS subscales and comparisons of the magnitudes of the correlations between the Chinese and European samples 51 6 Pearson-product-moment correlations between the H-MPS scales and the BDI and comparisons of the magnitudes of the correlations between the Chinese and European samples 53 7 Pearson-product-moment correlations between the H-MPS scales and the F-MPS and comparisons of the magnitudes of the correlations between the Chinese and European samples 54 8 Means and standard deviations for study variables for website and hard-copy participants 57 9 Comparison of mean scores for study variables for website and hard-copy participants 57 10 Means and standard deviations for the H-MPS scales for males and females in the Chinese sample 60 11 Comparison of mean levels of perfectionism between males and . females in the Chinese sample 60 12 Means and standard deviations for the H-MPS scales for males and females in the European sample 61 X Table Page 13 Comparison of mean levels of perfectionism between males and females in the European sample 61 14 Means and standard deviations for the H-MPS scales for the Chinese and Eurpean samples 62 15 Comparison of mean levels of self-oriented, other-oriented, and total perfectionism between the Chinese and European samples 62 16 Means and standard deviations for socially-prescribed perfectionism for males and females from the European and Chinese samples 63 17 Comparison of mean levels of socially-prescribed perfectionism between males and females in the European and Chinese samples 63 18 Multiple comparisons of mean levels of socially-prescribed perfectionism between males and females in the European and Chinese samples 64 19 Pearson-product moment correlations between the socially-prescribed perfectionism subscale and the BDI for males and females from the European and Chinese samples 66 20 Comparisons of the magnitudes of the correlation between the socially-prescribed perfectionism subscale and the BDI between males and females from the European and Chinese samples 66 21 Means and standard deviations for the H-MPS scales for first and second-generation Chinese samples 69 22 Comparisons of mean levels of perfectionism between first and second-generation Chinese samples 69 23 Means and standard deviations for the socially-prescribed perfectionism subscale for males and females from first and second-generation Chinese and European samples 70 24 Comparison of means for socially-prescribed perfectionism between males and females from first-generation Chinese, second-generation Chinese and European samples 70 xi Table Page 25 Multiple comparisons of mean levels of socially-prescribed perfectionism between males and females from first-generation Chinese, second-generation Chinese and European samples 71 26 Means and standard deviation for the H-MPS scales for high and low acculturation groups from the Chinese sample 74 27 Comparison of mean levels of perfectionism between the low and high acculturation groups from the Chinese sample 74 28 Perfectionism means reported in the current study and in previous studies , 141 Xll Acknowledgements I would like to thank my supervisor, Dr. Beth Haverkamp, for her support and valuable feedback throughout the project and for facilitating my learning about the process of research. I would also like to express my appreciation of the guidance and support offered by my committee members, Dr. Bill Borgen and Dr. Macy Lai. I would also like to express my gratitude to the U.B.C. students, who kindly took time out of their busy schedules, to participate in this research study. I would also like to thank the professors and T.A.s who supported this study by providing access to their students. For the valuable guidance and feedback during the analysis stage, I wish to thank Dr. Maria Trache, Statistical Consultant in the Faculty of Education. Finally, I would like to thank my partner Tom for his encouragement and support throughout the program and for his technical assistance in setting up the study website. I would like to extend many thanks to my peers who offered continual support during the program and to Georgina and Colleen for their friendships. 1 Chapter I Introduction Background Over the past decade, both clinicians and researchers have shown much interest in the topic of perfectionism. An ever expanding body of research suggests that perfectionism is a multidimensional construct, comprised of personal and interpersonal elements (Frost, Marten, Lahart & Rosenblate, 1990; Hewitt & Flett, 1991). Hewitt and Flett (1991) propose three core dimensions of perfectionism: self-oriented; other-oriented; and socially-prescribed. Self-oriented perfectionism involves setting high and unrealistic standards for oneself and stringently evaluating oneself. Other-oriented perfectionism is self-oriented perfectionism directed towards others. Socially-prescribed perfectionism involves the perception that significant others set high standards and stringently evaluate you. Although perfectionism has been positively linked to achievement and achievement striving, it has also been linked to negative outcomes such as low self-esteem, procrastination, poor psychosocial adjustment, and shame (Flett, Hewitt & De Rosa 1996; Saddler & Sacks 1993; Sorotzin 1985). Perfectionism has also been associated with various psychopathologies such as personality disorders, depression, and anorexia nervosa (Flett, Hewitt, Blankstein & Mosher 1995; Garner, Olmstead, &Polivy 1983; Hewitt & Flett 1991;). Several researchers have identified perfectionism as a factor in the suicides of talented individuals (Blatt 1995; Delisle 1986) and Pacht (1984) indicated that perfectionism played a role in law school dropout rates. One might expect perfectionism to be more prevalent on university campuses given the concentration of 2 talented young people, the emphasis on achievement and the pressures to excel. Thus, perfectionism is an important construct to investigate in a higher education setting. However, one gap in the literature is that perfectionism has been studied in primarily middle class, Caucasian university students and little, if any, research has been undertaken using subjects from different cultural backgrounds. Therefore, the present multidimensional conceptualization and operational definition of perfectionism remains to be tested in terms of its meaningfulness and applicability to other cultural groups. Given that 44.5% of the total immigrants to British Columbia between 1991 and 1996 were of Chinese heritage from Hong Kong, the People's Republic of China and Taiwan, it is likely that there is an increased presence on campuses of first and second generation Chinese students. Although the University of British Columbia does not keep statistics on students' ethnic origins, an Undergraduate Student Demographic Survey was administered in 1997 to a sample of 847 students. The purpose of the survey was to assess the extent to which societal demographic changes are reflected on the UBC. campus. In the survey, students were asked to identify the ethnic or cultural group(s) to which their ancestors belonged. Of the 461 respondents who reported a single response, fifty-four percent of the students identified Chinese as their ethnic background, followed by East Indian (6 percent) and British Isles (5 percent). In a separate question, students were asked to identify the category within which they considered themselves to be. Of the 843 respondents, 49 percent identified themselves as White, followed by Chinese (31 percent) and South Asian (5 percent) (UBC. Office of Budget and Planning, 1998). The high percentage of Chinese students on campus, supports a decision to study perfectionism in this cultural group. 3 Studying perfectionism in a Chinese sample is also warranted because the construct is likely relevant to this group. In a review of the research literature on Asian Americans, an emphasis on education and academic achievement is salient. Sue and Zane (1985) suggest that this strong cultural emphasis may lead not only to educational achievement but also to mental health problems such as: somatic complaints and family conflict (Sue and Sue, 1974); higher levels of depression (Kuo, 1984); and other socioemotional adjustment difficulties (Sue and Zane, 1984). Purpose The likely relevance pf the perfectionism construct to Asian-university students and the increasing numbers of students from Chinese cultural backgrounds on campus at the University of British Columbia, argues for research to test the cross-cultural applicability of a perfectionism measure using this sample. Therefore, the overall aim of the current study was to determine if the Multidimensional Perfectionism Scale (H-MPS, Hewitt & Flett, 1991) shows evidence of construct validity in a sample of university students from Chinese cultural backgrounds. Construct validity was explored by investigating similarities and differences between Chinese-Canadian and European-Canadian samples in the following: the internal structure of the measure; and the relationships between the measure and other measures. In this way, both internal and external sources of validity were examined. It was decided that mean levels of perfectionism would be reported for the Chinese-Canadian sample and comparisons between the two cultural groups, with respect to mean levels of perfectionism, would be made if strong evidence of construct validity was demonstrated. 4 Implications Implications of the findings of the current study are primarily theoretical/conceptual but also practical. If evidence of construct validity is provided, it may suggest that the personal and social aspects of perfectionism may generalize across cultures, thus bolstering the construct validity. Further studies replicating these findings and investigating perfectionism in other cultural groups would be warranted. If there is evidence of reliability and validity of the H-MPS using a Chinese sample, the current study may lend support for using the measure with Chinese university students. In addition, the study may shed light on whether the previously documented relationship between multidimensional perfectionism and depression applies to a Chinese sample. If differences between the Chinese and European samples with respect to mean levels of perfectionism are not found, the stereotype of the high-achieving, hard-working, perfectionistic Asian student may be debunked. If differences in the psychometric properties of the measure and/or differences in mean levels of perfectionism are found between the two groups, further investigations using Chinese samples would be warranted to better understand the construct. 5 Chapter II Literature Review Evolution of the Perfectionism Construct Early Theories Early formulations of perfectionism were derived primarily from the clinical observations of several key individuals who laid the groundwork for the present conceptualization of perfectionism and concomitant research. Therefore, it is important to provide some background information as it sets the stage for the present research. Early definitions of perfectionism emphasized the cognitive styles of individuals that seemed to strive relentlessly towards goals that were viewed, from an observer's perspective, to be unreasonable and unattainable. Several theorists (Barrow & Moore, 1983; Burns, 1980; Sorotzin, 1985) documented the following cognitive tendencies of perfectionistic individuals: dichotomous, or "all-or-none" thinking whereby one views experiences as successes or failures with no middle ground and no flexibility; overgeneralization whereby one generalizes failure, for example, from one situation to others and "tends to jump to the dogmatic conclusion that a negative event will be repeated endlessly" (Burns, 1980 p.38); should statements, such that one can fall prey to the "tyranny of should systems" (Fforney, 1945) with statements like "I should never fail" and "I ought to do better"; and moralistic self-evaluations such that when a goal is not met to the complete satisfaction of the perfectionistic individual it is accompanied by self-criticism which may result in a failure experience and a loss of self-esteem. Based on clinical observations, it was thought that many detrimental behaviour patterns and emotional problems could be attributed to perfectionistic tendencies (Sorotzkin, 6 1985). Bums (1980) linked perfectionism to decreased productivity, health problems, low self-esteem, disturbances in personal relationships, and a vulnerability to debilitating mood disorders such as depression, performance anxiety, test anxiety, social anxiety and writer's block, as well as obsessive-compulsive disorders. The high incidence of perfectionistic attitudes in people suffering from depression was particularly striking for Burns who suggested that:"Perfectionists become trapped by non-productive, self-critical ruminations that lead to depression and an unrealistically negative self-image"(pg. 38). Asher Pacht (1984), in his Distinguished Professional Contribution address at the American Psychological Association meeting in 1983, indicated that in the literature he had reviewed, perfectionism was associated with a variety of psychopathologies. The list included: alcoholism, erectile dysfunction, Munchausen syndrome, irritable bowel syndrome, depression in children and adults, anorexia nervosa, obsessive compulsive personality disorders, writer's block, ulcerative colitis, and Type A coronary-prone behaviour (Pacht, 1984). In fact, both Burns and Pacht noted that perfectionism was a particular problem for some university students. Burns studied 25 law school students at Penn State who sought counselling in their first year because of a high degree of stress. Many expressed the urge to leave school. Among a majority of students in this group, Burns (1980) observed an entrenched perfectionistic thinking pattern. He speculated that the law students had always perceived themselves to be at the top of their class during high school and university, and were having great difficulties accepting a place somewhere in the middle of their class. As a result, they reacted with panic, frustration, depression, and ultimately with a loss of self-esteem such that they desired to withdraw from the painful circumstances. Some dropped out of law school and a few may have contemplated suicide. 7 Burns suspected that withdrawing served the purpose of protecting their identity and self-esteem from their own self-criticism. There is still some debate in the literature whether or not perfectionism is always a negative characteristic. While Hamachek (1978) differentiates normal versus neurotic perfectionism, Pacht (1984) indicated that he preferred not to use the label normal perfectionism because: "the insidious nature of perfectionism leads me to use the label only when describing a kind of psychopathology." (p.387). Burns concurred stating that when he speaks of perfectionism he dose not mean the healthy pursuit of excellence by men and women who take genuine pleasure in striving to meet high standards...the perfectionists I am talking about are those whose standards are high beyond reach or reason, people who strain compulsively and unremittingly toward impossible goals and who measure their own worth entirely in terms of productivity and accomplishment. For these people, the drive to excel can only be self-defeating (p.34). Missildine (1963) also viewed perfectionism as a negative quality that drove individuals to strive relentlessly to attain unrealistic goals. Hamachek (1978), however, described two types of perfectionists: normal and neurotic perfectionists. Normal perfectionists are able to flexibly apply perfectionistic standards, lowering their standards as the situation demands without a loss of self-esteem or feelings of failure. They also seem to establish realistic expectations for themselves and are thus more likely to experience success. However, the neurotic perfectionist seems to set unrealistically high standards and strives toward unattainable goals. Neurotic perfectionists often feel as though their efforts and performances are never quite good enough such that they rarely, if ever, feel satisfied with, 8 or derive self-esteem from, their accomplishments. Many have speculated as to the origins of perfectionism and most theorists link the development of perfectionism to early parent-child relationships and to the emotional environment of the family. Antecedents of Perfectionism Alfred Adler proposed that one of the central strivings of human beings was for perfection (1964) or superiority (1972). People strive for mastery and perfection as a compensatory reaction to basic feelings of inferiority. Striving for perfection was thus viewed as a normal part of human development that is learned early in life in the family, the primary social environment of the child. Motivated by overcoming feelings of inferiority, individuals become goal-directed and contribute to the development of the human community and thus live on what Adler called the "useful" side of life. However, a "faulty family value" of basing self-worth on performance that is communicated to children by parents, could lead to the development of an even greater sense of inferiority and one's strivings become self-focused in the name of protecting a delicate ego. According to Adler, this lifestyle no longer constitutes a healthy pursuit of excellence and one ceases to contribute to the greater common good, thus living on the "useless" side of life. Like Adler, Hamachek (1978) speculated that unhealthy perfectionism is a product of the way in which approval of children is communicated by parents. Therefore, family environments that communicate non-approval, inconsistent approval, or conditional positive approval are more likely to engender neurotic perfectionism in children. As Hamachek (1978) states: The conditional positive approval environment differs from the non-approval environment in that expressed approval is explicit when certain high standards are met. However, it is like the non-approval environment insofar as there is a 9 noticeable absence of feedback when performance is below expected standards. Since approval has been experienced under more favourable circumstances, non-approval has a greater likelihood of being interpreted as punishment. It is in ways like this that the mental connection between being perfect and approved of are linked strongly together (pg. 29). Normal perfectionism, however, develops from positive and negative role modelling. Positive modeling occurs when a child observes a parent who sets realistic expectations and goals and derives a real sense of pleasure from a job well done. On the other hand, negative role modelling occurs when children observe a parent with a proclivity for disorganization and for jumping from task to task without completion. In the latter case, rather than mimicking the parent's behaviour, a child sets out to be meticulous and organized as a reaction against the parent's frustrating behaviour. Other theorists also implicate perfectionistic parents, whose self-esteem is contingent on their children's success, in the development of perfectionism in children (Barrow & Moore, 1983; Burns, 1980; Frost, Lahart & Rosenblate, 1991; Missildine, 1963). Halgrin and Leahy (1989) offer some explanation of the mechanism of transmission: "Parents reinforce their child's excellence in academic or social worlds while they react anxiously to failure, viewing it as a poor reflection of themselves. These children, in a desperate pursuit of parental love and acceptance, strive to be flawless" (pg. 223). Burns (1980) suggested that although perfectionism is rooted in early interactions with parents, the perfectionistic "mind-set" easily perpetuates itself once it has taken hold. Burns explains: Whenever the child performs in an outstanding manner, he or she repeats an internal message of the following type: I did perfectly on this. This shows that I'm okay and 10 that I deserve to feel good... Positive feelings created by the inner dialogue powerfully reinforce the perfectionism. In contrast, when the child makes an error or goofs up, the automatic response is a succession of self-punishing negative thoughts like: / shouldn 't have made that mistake and How could I be so stupid? That kind of thinking results in guilt, anxiety, and frustration, further reinforcing the belief that mistakes are unacceptable. I believe that perfectionistic attitudes and their emotional consequences reinforce each other (pg. 41). As Barrow and Moore (1983) point out, it is likely that there are societal factors that also contribute to the development and maintenance of perfectionism in some individuals, and the educational system, with its emphasis on achievement, may be particularly guilty. Frost et al. (1991) concluded that perfectionism in parents is important to consider in understanding perfectionism in children and that the gender of the parent, and the level of parental criticism, appear to be important in the transmission of perfectionism to children. With respect to the etiology of perfectionism, much remains to be uncovered by well-designed empirical investigations as is reflected by the following statement: "Despite considerable conceptual interest in the influence of relationships on the development of perfectionism, surprisingly little research has been completed on parent-child relationship correlates or presumed causes of perfectionism" (Rice, Ashby & Preusser, 1996 p. 248). In summary, in the 1970s and early 1980s practitioners began to notice what they called perfectionism in some of their clients, including students for whom this appeared to be a particular problem. Based on clinical observations, these practitioners began to conceptualize perfectionism primarily as a cognitive style that was associated with psychopathology and adjustment difficulties. Many theories relating to the etiology and 11 potential correlates of perfectionism were theorized; however, measures of perfectionism were noticeably lacking. All of these theories still needed to stand the test of empirical research in order to have validity. Burns saw the development of a measure of perfectionism as a necessary step in understanding and delineating the construct. As a result, he developed the first instrument that measured perfectionism, the Burns Perfectionism Scale (Burns, 1980). Perfectionism Measures The Burns Perfectionism Scale (Burns, 1980) was adapted from a portion of the Dysfunctional Attitudes Scale (Weissman & Beck, 1979) to create a ten-item inventory that is heavily weighted on personal standard setting and concern over mistakes. The scale operationalizes what could be called a unidimensional conceptualization of perfectionism in that it focuses exclusively on self-directed cognitions. Most research investigating perfectionism prior to 1990 employed the use of this operational definition. A subscale of the Irrational Beliefs Test (Jones, 1968) that focuses on personal standard-setting has also been used to operationalize the construct, as has the perfectionism subscale of the Eating Disorder Inventory (Garner, Olmstead & Polivy, 1983) which emphasizes personal standard setting and parental expectations. In the current literature (eg., Frost et al., 1990; Hewitt & Flett, 1991), it appears that this unidimensional conceptualization of perfectionism (ie., focused almost exclusively on self-directed cognitions) as reflected in the above-mentioned scales, is viewed as being inadequate and a multidimensional conceptualization has been operationalized and generally accepted. With the advent of psychometrically sound scales that reflect a multidimensional 12 conceptualization of perfectionism, research on perfectionism has burgeoned over the past decade. Psychological Research on Perfectionism Hewitt and Flett (1990) contended that while beliefs about perfectionism for the self represent a central aspect of perfectionism, it is in fact a multidimensional construct, consisting of self, motivational and social components and that these elements are related to adjustment and other difficulties. They criticized past research that utilized the Burns Perfectionism Scale for its myopic focus and presented a new conceptualization of perfectionism, reflecting those dimensions that were untapped by prior research. Similarly, Frost and colleagues (1990) independently proposed a multidimensional conceptualization of perfectionism, albeit a different one from Hewitt and Flett. Both groups of researchers then set out to test their conceptualization empirically and,through programmatic research, have contributed substantially to the ever expanding empirical literature in this domain. Hewitt and Flett (1990, 1991) proposed three core dimensions of perfectionism, comprising the Multidimensional Perfectionism Scale (H-MPS): self-oriented; other-oriented; and socially prescribed perfectionism. The differences between the dimensions include to whom the perfectionism is directed (eg., the self or others), and who is setting the standards for perfection (eg., the individual or others). Self-oriented perfectionism is similar but not identical to the Burns conceptualization. Similar to Burns, Hewitt and Flett viewed self-oriented perfectionism to include behaviours such as setting exacting standards for oneself and rigidly evaluating and criticizing one's own behaviour (p.457). However, they also added a motivational component which included striving to attain 13 perfection in one's activities and striving to avoid failures. It was also hypothesized that a discrepancy between the ideal and actual self was also a part of self-oriented perfectionism. Research had already linked such a discrepancy to depression (Higgins et al. 1986; Straumann, 1989) and low self-esteem (Hoge & McCarthy, 1983; Lazzari et al, 1978). In addition, self-oriented perfectionism, as operationalized by the Burns Scale, had already been associated with indices of maladjustment, including anxiety (Flett & Hewitt, 1989) and subclinical depression (Hewitt & Dyck, 1986; Hewitt & Flett, 1990; Pirot, 1986). Thus, there was an empirical basis for Hewitt and Flett's (1990) hypothesized association between self-oriented perfectionism and adjustment difficulties, namely depression. Another key dimension of perfectionism proposed by Hewitt and Flett (1990) was other-oriented perfectionism. They drew on the work of Hollender (1965) who proposed that some individuals engage in interpersonal perfectionistic behaviour. These behaviours consisted of holding unrealistic standards for significant others, placing importance on perfection in other people, and stringently evaluating and criticising others' performances (Hewitt & Flett, 1990). This dimension is equivalent to self-oriented perfectionism that is directed towards other people. Thus, other-oriented perfectionism would be hypothesized to be associated with interpersonal problems and, potentially, with loneliness. Rationale for this proposed dimension came from research on interpersonal functioning and "other-oriented should" statements (Demaria et al, 1989; Kassinove, 1986) and research on familial aspects of levels of aspiration (Morris, 1961). Finally, the third dimension proposed, socially-prescribed perfectionism, taps into the need to attain the standards and expectations of significant others. More specifically, 14 it involves the perception that others have unrealistic and unattainable standards for them and that they are being evaluated and criticized by others and pressured to attain perfection in their endeavours. They hypothesized that socially-prescribed perfectionism should be associated with a variety of negative outcomes, including anger, anxiety and depression, given that the standards set by others are viewed as being unreasonably high and uncontrollable. Hewitt and Flett (1991) also suggested that individuals with elevated levels of socially-prescribed perfectionism who are concerned about meeting others' standards would be expected to exhibit a greater fear of negative evaluation and desire for approval from others as well as attempt to avoid disapproval. Hewitt and Flett (1991) provide rationale for this proposed dimension by citing research that linked the perception that other's have high expectations of them to relapses in schizophrenia (Vaughn & Leff 1983) and depression (Hooley & Teasdale, 1989). Another group of researchers, Frost, Marten, Lahart and Rosenblate (1990), also conceptualized perfectionism as a multidimensional construct. Frost and colleagues concurrently and independently developed a Multidimensional Perfectionism Scale (F-MPS), although this scale was based on a somewhat different conceptualization. They hypothesized six dimensions including: excessive concern over making mistakes; high personal standards; the perception of high parental expectations; the perception of high parental criticism; the doubting of the quality of one's actions; and a preference for order and organization. In their original series of studies on the F-MPS, Frost and colleagues found that, although prior definitions of perfectionism underscored the importance of setting excessively high standards, their findings suggested that concern over mistakes was the core dimension. Furthermore, they found that this dimension was most closely 15 associated with psychopathology, especially depression. They also found that two subscales, namely, high personal standards and a need for order and organization, were related to positive characteristics like positive achievement striving and work habits. It was also found that concern over mistakes, personal standards, parental expectations, parental concerns, and doubts about actions subscales were all related to the overall perfectionism score, the other subscales of the F-MPS, and other perfectionism measures. However, the organization subscale was only marginally related to both overall perfectionism and the other subscales and was deemed to be a peripheral rather than core element of perfectionism. The question remained: How are the two multidimensional conceptualizations as reflected in the two MPSs related to each other, and are they similarly related to psychopathology? Frost and colleagues (1993) investigated the relationships between the two Multidimensional Perfectionism Scales and their relationships to measures of positive and negative affect. They found that there was much overlap between the two measures and that the F-MPS correlated primarily with the self-oriented and socially-prescribed perfectionism subscales of the H-MPS. Other-oriented perfectionism was related to a much smaller degree with the F-MPS. They also found that the subscales for each measure demonstrated differential relationships with measures of positive and negative affect. It was found that the dimensions that have previously been related to indices of psychopathology, concern over mistakes (F-MPS) and socially prescribed perfectionism (H-MPS), were most closely related to negative affect, while those which have been formerly associated with more adaptive characteristics, for example, high personal standards (F-MPS) and other-oriented perfectionism (H-MPS), were associated with 16 positive affect. Frost and colleagues (1990) conducted a factor analysis of the two MPSs and identified a conceptually clean two-factor solution. The dimensions that loaded on the first factor, labelled Maladaptive Evaluation Concerns, were, concerns about making mistakes, doubt's about actions, parental expectations, parental criticism, and socially-prescribed perfectionism. This factor was significantly related to depression and negative affect. The second factor, labelled Positive Striving, consisted of high loadings for the following dimensions: personal standards, organization, self-oriented perfectionism, and other-oriented perfectionism. They found this factor to be unrelated to depression but significantly correlated with positive affect. Other investigators (Rice, Slaney & Ashby, 1998; Suddarth, 1996) have more recently confirmed the findings of Frost et al. (1993), thus providing more convincing evidence of the adaptive and maladaptive elements of perfectionism. These empirical findings are consistent with Hamacheck's (1978) conceptual distinction between normal and neurotic perfectionism. Ffamachek's theorizing and the current empirical findings, taken together, indicate that perfectionism is a very complex phenomenon (Blatt, 1995) that is associated with both adaptive and maladaptive functioning. Research Related to the Adaptive Elements of Perfectionism Some dimensions of perfectionism such as self-oriented perfectionism (H-MPS) and high personal standards and orderliness (F-MPS) appear to have adaptive potential. Self-oriented perfectionism has been related to resourcefulness and constructive striving (Flett, Hewitt, Blankstein & Mosher, 1991), high self-standards (Hewitt & Flett, 1991) and high levels of positive behavioural coping, including an action-orientation and conscientiousness (Flett, Russo & Hewitt, 1994). In their investigation of the personality 17 correlates of perfectionism using the NEO-PI, Hill, Mclntire and Bacharach (1997) found that self-oriented perfectionism was strongly associated with conscientiousness and that achievement striving and dutifulness were the two elements of conscientiousness which contributed most to self-oriented perfectionism. Frost, et al. (1990) found that high personal standards are associated with positive achievement striving and efficacy and negatively associated with the frequency of procrastination. However, when the variance in common with the efficacy scale was controlled, the personal standards subscale was significantly correlated with depression. In fact, most of the literature on perfectionism is focused on the perfectionism-depression connection. Research on Perfectionism and Adjustment Difficulties in University Students Depression can be a serious problem for university students. Loneliness, isolation, and suicide ideation and attempts have all been related to depression in this population (Hewitt, Flett & Weber, 1994; Westerfield & Furr, 1987). In his review of theoretical accounts and empirical research, Blatt (1995) concluded that there is an unequivocal link between perfectionism and depression. Individuals with high levels of self-criticism and perfectionism have been found to be vulnerable to experience failure and to react to failure with increased depression (Blatt & Zuroff, 1992). In fact, Blatt (1995) noted that self-critical, perfectionistic clients responded poorly to brief treatment, both pharmacologic and psychotherapeutic, for depression. Multidimensional perfectionism has been associated with adjustment difficulties and depression in university students (Arthur & Hayward, 1997; Flett, Hewitt, Blankstein & Mosher, 1995; Flett, Hewitt, Blankstein & Mosher, 1991; Flett, Hewitt, Blankstein & O'Brien, 1991; Flett, Hewitt & DeRosa, 1996; Hewitt & Dyck, 1986; Hewitt & Flett, 1991; 18 Hewitt & Flett 1990; Joiner & Schmidt, 1995; Preusser, Rice & Ashby, 1994; Rice, Slaney, and Ashby, 1998; Sadler & Sacks, 1993;), as well as with suicide ideation (Delisle, 1990; Hewitt, Flett & Weber, 1994). Much of the research on perfectionism has focused on the direct link between perfectionism and depression (Rice et al., 1998). Research with university students has consistently demonstrated that both self-oriented and socially-prescribed perfectionism are correlated positively with subclinical depression and are independently related to depression (Arthur & Hayward, 1997; Flett, Hewitt, Blankstein & O'Brien, 1991; Hewitt & Dyck, 1990; Hewitt & Flett, 1991; Saddler & Sacks, 1993;). Hewitt, Flett and Weber (1994) found that both self-oriented and socially-prescribed perfectionism are associated with greater suicidal ideation in university students. In fact, based on anecdotal evidence, the relationship between perfectionism and suicide ideation and intent is strongest for "gifted" individuals (Blatt, 1995; Delisle, 1986, 1990; Driscoll, 1989; Farrell, 1989; Hays & Sloat, 1989; Weisse, 1990). Socially-prescribed perfectionism, in particular, has been implicated in the psychosocial difficulties experienced by university students. High levels of socially-prescribed perfectionism have been linked to greater loneliness, shyness, fear of negative evaluation, and lower self-esteem (Flett, Hewitt & De Rosa, 1996). It has been demonstrated that high levels of socially-prescribed perfectionism are associated with lower GPAs (Arthur & Hayward, 1997) and that when combined with low levels of self-actualization predict higher levels of depression in college students (Flett, Hewitt, Blankstein & Mosher, 1991). One factor that may play a role is academic procrastination which has been linked to socially-prescribed perfectionism (Saddler & Sacks, 1993). Academic procrastination has been defined as: "The self-reported tendency to put off 19 academic tasks nearly always and to experience problematic levels of anxiety associated with it" (Rothblum, Solomon, & Murakami, 1986, pg. 387). Solomon and Rothblum (1984) found that academic procrastination was positively correlated with anxiety and depression in college students. Those who have high levels of socially-prescribed perfectionism may put off tasks that need to be done, such as writing a paper, for fear of it not being good enough and for fear of negative evaluation. If one procrastinates long enough, anxiety about the assignment sets in, and depression may ensue. However, further research investigating this complex association among perfectionism, academic procrastination and depression is required. While studies investigating the perfectionism-depression connection that employ the use of the F-MPS are appearing in the recent literature, much less research overall has been done using this scale in contrast to the H-MPS. Frost and colleagues have found that overall perfectionism is significantly related to self-critical depression (Frost et al., 1990) and that overall perfectionism and the following subscales were significantly and positively related to depression using the Beck Depression Inventory (Beck, Ward, Mendelson & Erbaugh, 1961): concern over mistakes; parental criticism; and doubts about actions (Frost et al., 1993). Interestingly, when one combines the findings of studies that utilized the F-MPS and those that used the H-MPS, a clear picture of the perfectionism-depression connection emerges. For the most part, it appears that those dimensions previously associated with maladaptive evaluation concerns (ie., they loaded on this factor) were correlated with indices of depression (socially-prescribed perfectionism, concern over mistakes, parental criticism) and those that were associated with positive achievement striving (ie., they loaded on this factor) were not correlated 20 with depression levels (other-oriented perfectionism). However, one dimension, self-oriented perfectionism, while loading on the achievement striving factor is also significantly associated with depression. In addition, it has been demonstrated that self-oriented perfectionism interacts with negative life events, namely failure experiences, to predict depression and other negative mental health outcomes which are exacerbated when the individual also has high levels of socially-prescribed perfectionism (Blatt, 1995). While Hamachek (1978) referred to individuals as normal or neurotic perfectionists, the empirical findings ostensibly point to the existence of positive achievement strivings accompanied by psychological disturbance within some individuals. For the most part, the research outlined above pertains to the investigation of the direct link between perfectionism and depression. However, the association may be much more complicated than previously thought. A few groups of researchers have proposed and empirically tested models that attempt to delineate the path from perfectionism to depression, including: a model that places self-esteem in a mediational role (Preusser et al., 1994; Rice et al., 1998); and a diathesis-stress model that postulates an interaction between perfectionism and stress or negative life events to predict depression (Flett et al, 1995; Hewitt & Dyck, 1986; Joiner & Schmidt, 1995). Preusser et al. (1994) found some support for self-esteem as a mediator of the perfectionism-depression link. Rice and colleagues (1998) further investigated the validity of the mediational model, using structural equations modeling, and found that adaptive perfectionism was not directly or indirectly (via self-esteem) related to depression. Maladaptive perfectionism, in contrast, was associated with low self-esteem and higher 21 depression. They concluded that self-esteem was a buffer of maladaptive perfectionism, rather than a mediator of the perfectionism-depression association. In testing the diathesis-stress model, Hewitt and Flett (1993), using a sample of psychiatric patients, found that self-oriented perfectionism interacted primarily with achievement stressors to predict depression, whereas socially-prescribed perfectionism interacted with both interpersonal and achievement stressors to predict depression. Flett et al. (1995) found that, for college students, high levels of self-oriented perfectionism interacts with the experience of negative life events to predict depression, over and above the main effects of self-oriented perfectionism and life stress. They offer an interpretation of these findings: Self-oriented perfectionism may be relatively adaptive in situations of low stress. It may be that self-oriented perfectionists under low stress have a sense of personal control and their need for control is not threatened...However, a substantially different picture emerges under conditions of high stress. Greater levels of depressive symptomatology are likely to result when the self-oriented perfectionist reports experiencing negative life events that have the potential to disrupt both the sense of control and the need for personal control. (Flett et al. 1995, pg. 128) Findings related to socially-prescribed perfectionism using a college sample were much less clear. While there was a significant positive association between socially-prescribed perfectionism and symptoms of depression in one of the two samples of college students, the interaction between socially-prescribed perfectionism and negative life stress did not reach significance. It remains unclear why there are such discrepancies 22 between a college and psychiatric patient sample, and much remains to be learned about the path from perfectionism to depression and about the role played by third variables such as stress, procrastination and self-esteem. However, the link between perfectionism and depression in university students is well substantiated and can have far-reaching implications for adjustment, quality of life and for the treatment of depression in this population. In summary, since 1990 the research literature on perfectionism has expanded tremendously with the advent of recently developed perfectionism scales. The empirical findings provide consistent evidence that perfectionism is in fact a multidimensional construct that can play a role in both adaptive and maladaptive functioning. Given the interest and research activity in this domain, one can speculate that the perfectionism construct will be further delineated in the decade to come. However, one serious limitation in this domain of research is the lack of attention given to investigating cultural differences in perfectionism. As Chang (1998) points out, according the Basic Behaviour Science Task Force of the National Advisory Mental Health Council (1996), the examination of cultural differences is a top priority for developing valid theories and effective interventions that can apply to the health needs of diverse groups. Given that perfectionism has been studied primarily in college students from Western-European cultural backgrounds, the nature of the perfectionism construct and the applicability of the current conceptualization of perfectionism to individuals from other cultural backgrounds remains untested and unknown. 23 Research Related to Perfectionism in Asian-American College Students There is a dearth of empirical studies examining perfectionism in a cross-cultural context. In fact, in a thorough review of the empirical literature, only one published article that addresses cultural differences in perfectionism was located. Chang (1998) investigated cultural differences between Asian American and Caucasian American college students in social problem solving, perfectionism, and suicidal risk. Using the F-MPS, Chang found that Asian Americans generally scored higher on perfectionism when compared to Caucasians and scored significantly higher on four of the subscales: concern over mistakes; parental expectations; parental concerns; and doubts about actions. Kawamura (personal communication, August, 1998) also investigated perfectionism in Asian and Caucasian college students and replicated Chang's findings. The only difference was that, while Asian students scored higher than Caucasian students on the concern over mistakes subscale, statistical significance was not reached in Kawamura's study. In addition, Kawamura found that Caucasian students scored higher on the personal standards subscale which is generally associated with positive achievement striving, or adaptive functioning, although the differences were not statistically significant. No significant gender differences in perfectionism were found for either cultural group. Chang also found that Asian students scored higher on both measures of suicidal risk and that, irrespective of culture, perfectionism added significant incremental validity in predicting responses on both scales of suicidal risk. It is particularly noteworthy that differences between the two cultural groups were generally consistent across the two studies and that Asian Americans scored significantly higher on the dimensions of perfectionism that are associated with maladaptive functioning and negative mental health outcomes. Clearly, initial evidence points to 24 potential differences in perfectionism between European and Asian university students such that further investigation into the multidimensional nature of perfectionism in the Asian population and the link to maladaptive functioning is warranted. Relevance and Rationale for Studying Perfectionism in Asian Students There is a popular image of Asian students as being diligent workers and high achievers. Statistical evidence points to the academic and educational achievements of Asian students. In the United States, Asian-American children comprise 4.4% of the "gifted" school population, while they represent only 2.2% of school-age children (Kitano, 1986). The 1980 United States Census revealed that Asian Americans have the highest level of college education of any ethnic or racial group in that country. For individuals from twenty-four to twenty-nine years of age, 60% of Chinese men and 44% of Chinese women had completed college compared to 34% and 22% respectively for the majority group. Research on Asian Americans suggests a cultural orientation that emphasizes education and academic achievement (Cordova, 1983; Kitano, 1969; Mordkowitz & Ginsburg, 1986; Pang, 1991; Santos, 1983; Sata, 1983; Schwartz, 1971; Sue & Chin, 1983). As Pang (1991) reiterated: Asian-American traditional values place great importance upon education for personal growth and family pride. The identity of many Asian Americans is tied to family pride and each member contributes to that pride. Children are encouraged to work hard in school because their academic achievement contributes to family status and honour.. Asian-American students may feel more pressure from their families, parents, and communities to excel in school, (p. 1) 25 It has been suggested that this strong cultural emphasis on education may lead to not only greater academic achievement, but also to greater mental health problems (Sue & Zane, 1985). Sue and Sue (1974) compared Japanese-American and Chinese-American college students with European-American students on the MMPI and found that the Asian sample reported more somatic complaints and greater family conflict. Kuo (1984) found Asian Americans reported greater levels of depression as compared to a Caucasian sample and concluded that, while Asian Americans have low utilization rates of mental health services, they do not have low levels of depressive disorders. In examining the academic achievements and socioemotional adjustment of Chinese undergraduate students, Sue and Zane (1984) concluded that: "The image of high-achieving, well-adjusted Chinese student is tempered when examining criteria other than grade point average" (pg. 570). Given family pressures to excel and that adjustment difficulties and affective disorders may accompany the educational achievements of some Asian students, it would be relevant and interesting to study perfectionism in this group. There is now substantial evidence that perfectionism plays a role in both the positive achievement strivings and the maladaptive evaluation concerns of university students from European cultural backgrounds. However, the applicability of a multidimensional view of perfectionism and the potential role played by perfectionism in the achievements and adjustment difficulties of Asian university students has not yet been examined, with the exception of Chang's preliminary work (1998). While Chang investigated perfectionism in Asian students as a group, he calls for norms for the various ethnic groups that comprise this categorical group. Studying perfectionism in a sample of Chinese university students at U.B.C. is important given the potential relevance of the perfectionism construct to this group, and given the large proportion of U.B.C. undergraduate students, 54 percent, who report their ethnic origins are Chinese. Therefore, the major objective of the current study was to determine if the Multidimensional Perfectionism Scale (H-MPS, Hewitt & Flett, 1991) showed internal and external evidence of construct validity. Internal evidence was assessed by comparing the internal structure of the measure for Chinese and European university students. External evidence was evaluated by comparing the relationships between the H-MPS and other measures for the two samples. Similar internal structures and relationships between the H-MPS and other measures for the two samples was interpreted as providing evidence of construct validity for a Chinese-Canadian sample of university students. The following specific research questions were addressed in the current study. Research Questions (1) Do the internal consistency reliabilities and the internal patterns of correlation of the H-MPS show evidence of construct validity for a sample of students from Chinese cultural backgrounds? 1-A Are there differences in the magnitudes of the internal consistency coefficients of the H-MPS scales between the Chinese and European samples? 1-B Are there differences in the patterns of the internal consistency coefficients for the H-MPS subscales between the Chinese and European samples? 1-C Are there differences in the magnitudes of the correlations among the H-MPS subscales between the Chinese and European samples? 27 1-D Are there differences in the patterns of correlation among the H-MPS subscales between the Chinese and European samples? (2) Does the H-MPS show external evidence of construct validity for a sample of students from Chinese cultural backgrounds? 2- A Are there differences in the magnitudes of the correlations between the H-MPS scales and a measure of depression between the two groups? 2-B Are there differences in the patterns of correlation between the H-MPS scales and a measure of depression between the Chinese and European samples? 2-C Are there differences in the magnitudes of the correlations between the H-MPS scales and another measure of perfectionism between the two groups? 2-D Are there differences in the patterns of correlation between the H-MPS scales and another measure of perfectionism between the Chinese and European samples? If strong evidence of construct validity is demonstrated in answering research question one, a preliminary investigation of the similarities and differences between European and Chinese students with respect to mean levels of perfectionism would be meaningful. (3) Are there differences in levels of perfectionism between university students from European and Chinese cultural backgrounds? Differences in levels of perfectionism will be investigated by using total and subscale means for the H-MPS. 28 Chapter III Methodology Design The major aim of the current study was to determine whether there is evidence of construct validity for the Multidimensional Perfectionism Scale (H-MPS) using a sample of Chinese undergraduate students. A series of self-report questionnaires were administered to European and Chinese undergraduate students at the University of British Columbia to obtain relevant data. Comparisons between the two samples in the internal structure of the measure and in the relationships between the H-MPS and other measures were made. Therefore the study employed a group-comparison design and used correlational data. Both descriptive and inferential statistical procedures were applied. Sample 197 undergraduate students at the University of British Columbia (U.B.C) voluntarily participated in the current study. In order to participate in the study, students had to meet the following criteria: currently registered as students at U.B.C; currently reside in the lower mainland of British Columbia; and identified that their ancestors were from Chinese or European cultural backgrounds. The ethnic origin categories established by Statistics Canada (1996 census Dictionary) were adapted and used to classify participants as European and Chinese. Appendix A provides details on the groups included in the European and Chinese samples. Of the 197 participants, 5 percent responded to a poster and 95 percent responded to in-class invitations to participate. One hundred and ninety-one students met the above criteria and were included in the analyses. Six cases were excluded for the following 29 reasons: four individuals identified that their ancestors were from Japan, Korea, or Vietnam; one individual reported that he was not a U.B.C. student; and one person indicated that his ancestors were from China and Norway. Of the 191 remaining participants, 117 (78 females and 39 males) were from Chinese cultural backgrounds and 74 (40 females and 34 males) were from European cultural backgrounds. The mean age for the Chinese sample was 20.38 years, with 80 percent of the sample between the ages of 18 and 22. Ninety-seven percent of the students were registered as full-time undergraduate students, with 50 percent in first year, 29 percent in second year, 15 percent in third year and 5 percent in fourth year. Forty-nine percent of the Chinese students identified themselves as perfectionists, while 51 percent did not. The following was the breakdown for the reported birthplaces of the Chinese students: 42 percent were born in Canada, 25 percent were born in Hong Kong, 15 percent were born in Taiwan, 8 percent were born in China and 10 percent were born elsewhere outside of Canada. With respect to first language learned, the following was the breakdown: 56 percent reported Cantonese, 21 percent reported Mandarin; 13 percent reported English; and 10 percent reported other. In summary, 58 percent of the Chinese students were born outside Canada and 77 percent learned either Cantonese or Mandarin as their first language. Appendix B lists some of the characteristics of the Chinese and European samples in the current study.. The mean age for the European sample was 20.93 years, with 80 percent of the sample between the ages of 18 and 24. Ninety-six percent of the sample was registered as full-time undergraduate students, with 38 percent in first year, 38 percent in second year, 11 percent in 3rd year, and 12 percent in fourth year. Forty-three percent of the 30 European students identified themselves as perfectionists, while 57 percent did not. The following was the breakdown for the reported birthplaces of the European students: 89 percent in Canada; and 11 percent in Britain and Continental Europe. With respect to first language learned, the following was the breakdown: 80 percent reported English; 18 percent reported other European languages: and 1 percent reported French. In summary, 89 percent of the European students were born in Canada and 80 percent learned English as their first language. Students were provided with the option of completing questionnaires in hard copy or electronic form. Twenty-eight percent of the whole sample (Chinese and European samples) completed questionnaires through the website and 72 percent used hard copy forms. For the Chinese sample, 75% used hard copy forms, while 25% used website forms. For the European sample, 68% used hard copy forms, while 32% used website forms. While 72 percent of the male participants used the website, only 11 percent of the female participants used the website. Also noteworthy was that a greater proportion of the web-site users were students in the faculties of Science and Engineering. Overall, the European and Chinese samples appear to be similar in terms of the characteristics discussed above. A t-test was performed to determine if the two samples differed in mean age and chi square tests were conducted to determine if the proportion of females to males, hard copy to website participants, and self-reported perfectionists to self-reported non-perfectionists, differed for the two samples. No significant differences were found, indicating that the two samples are similar in terms of these characteristics. In comparing the characteristics of the European and Chinese samples with the characteristics of the U.B.C. population (U.B.C. Budget and Planning Statistics, 1999), 31 the groups were similar in age and the samples used in the current study had higher proportions of students in first and second year, lower proportions in years three to five and, a higher percentage of students in Arts. Unfortunately U.B.C. does not keep statistics on the ethnic origins of the student population, therefore it could not be determined how representative the sample used in the current study is of the U.B.C. population as a whole. Data Collection Procedures Two methods of recruitment were used to obtain the study sample. Posters were placed on campus bulletin boards in the Student Union Building and in many departments. In addition, letters of information about the study along with a request to enter classes for five to ten minutes to invite students to participate were sent to professors in the following departments: anthropology and sociology; asian studies; biology; chemistry; computer science; economics, engineering; and physics. Follow-up phone calls were made approximately one week following the receipt of the letter. Appendix C contains copies of the following: letter to professors; letter of information; and poster. These departments were selected as they had large classes of first and second year students and contained a high proportion of students from various departments across campus. Between late November 1998 and early February 1999, the researcher entered sixteen classes to invite students to participate. Appendix D lists the classes entered. The same information contained in the poster was read to students and any questions students had about the study were answered. Questionnaire packages were then handed out to individuals who decided to participate and pick-up dates were arranged at that time. 32 Students were given a contact number and e-mail address to obtain further information or to sign up to participate at a later date. Students were given two options for completing the questionnaire package, in hard copy or web-site form. The latter method involved setting up a website for the current study and providing students with a URL address and a unique password. Rationale for using the website came from the U.B.C. 1997 Undergraduate Student Demographic Survey that found that 90 percent of the undergraduate population had access to non-U.B.C. computers and that 60 percent used these computers to access the web. These statistics are a conservative estimate of the percentage of U.B.C. undergraduate students who access the web given many students use U.B.C. computers to access the web and the numbers have likely increased since 1997. In addition, the website method of data collection offers certain advantages: participants submit data directly to the database which allows the researcher to bypass the data-entry stage; and participants are alerted to missing or inappropriate data and are prompted to return to those questions to complete them before the data can be submitted. In this way, the researcher ensures a complete data set while at the same time ensuring confidentiality and anonymity. Appendix E contains pictures of the website. Using the website for data collection in the current study was experimental. Unfortunately, the publisher of the BDI (Psychological Corporation, 1996) would not give permission to the researcher to post the BDI on the website. Therefore, the BDI had to be completed in hard-copy form by website participants and returned to the researcher. As a result, many students who initially opted to complete questionnaires through the website decided 33 against it in favour of the hard-copy forms since the BDI had to be returned to the researcher anyways. A series of self-report questionnaires were administered to students. These included: a Background Information Sheet; the Multidimensional Perfectionism Scale (MPS - Hewitt & Flett, 1991); the Multidimensional Perfectionism Scale (MPS - Frost, Marten, Lahart, Rosenblate, 1990); the Suinn-Lew Asian Self-Identity Acculturation Scale (SL-ASIA - Suinn, Rickard-Figueroa, Lew & Vigil, 1987) to Chinese students only; and the Beck Depression Inventory (BDI - Beck, Rush, Shaw & Emery, 1979). This package of questionnaires took approximately 20 to 30 minutes to complete. Instrumentation Copies of all measures appear in Appendix F. Perfectionism Measures Multidimensional Perfectionism Scale (H-MPS - Hewitt & Flett. 1991) The H-MPS was initially developed to extend the conceptualization of perfectionism to include both self and social components. Prior to the development of this scale and the other MPS (F-MPS, Frost et al., 1990), perfectionism was conceptualized as a unidimensional construct, including only self-directed cognitions. The scale is designed to measure perfectionism as a general trait and is comprised of three subscales which allow for the assessment of individual differences in self-oriented, socially- prescribed, and other-oriented perfectionism. Items reflecting the three dimensions were derived from case descriptions and theoretical formulations. The scale was initially constructed using large numbers of university students and psychiatric patients in Toronto, Canada. The Marlowe-Crown Social-Desirability Scale (Crowne & 34 Marlowe, 1960) was utilized in scale construction, with items being retained only if they had a correlation of less than .25 with social desirability. Respondents rate the extent of their agreement with each item on a 7-point Likert-type scale. A total perfectionism score can be obtained by summing all 45 items and several are reverse keyed. Each subscale total is calculated by summing the 15 items pertaining to it. Higher scores are indicative of higher levels of perfectionism. Both total and subscale scores will be utilized in the present study given the exploratory nature of the study using a new population. With respect to the psychometric properties of the scale, numerous studies have reported on the reliability and validity of the scale (Hewitt & Flett, 1991a, 1991b; Hewitt, Flett & Blankstein, 1991; Hewitt, Flett, Turnbull-Donovan & Mikail, 1991), and the following coefficients were drawn from these studies. Item-to-subscale total correlations reported range from: .51 - .73 for self-oriented perfectionism; .43 - .64 for other-oriented perfectionism; and .45 - .71 for socially-prescribed perfectionism. Internal consistency, as measured by chronbach alpha coefficients range from: .86 - .89 for self-oriented perfectionism; .79 - .82 for other-oriented perfectionism; and .86 - .87 for socially-prescribed perfectionism. Intercorrelations among the H-MPS subscales range from .25 -.40. While there is some degree of overlap among the three subscales, the dimensions were reported to be distinct (Hewitt & Flett, 1991). Such overlap would be expected given the three dimensions are all tapping into the same overall construct. There is also some evidence of temporal stability with test-retest coefficients reported for a three month period at: .88 for self-oriented perfectionism; .85 for other-oriented perfectionism; and .75 socially-prescribed perfectionism. Hewitt and Flett (1991) also investigated the 35 factor structure and conducted a principal components analysis on the item responses of large numbers of university students. They confirmed that three factors, accounting for 36% of the variance in perfectionism scores could be retained. Solid evidence of convergent validity of the H-MPS subscales comes from comparisons with other personality measures (Hewitt & Flett, 1991b). Mixed support for discriminant validity of the three subscales has been reported. For example, self-criticism was positively correlated with all three perfectionism dimensions, however, Hewitt & Flett (1991b) suggest that this may be a true reflection of the perfectionism construct's nomological net (p. 463). Moderate correlations have been calculated between significant other, clinician and self-reported ratings of perfectionism suggesting that perfectionistic behaviours are observable and can be reliably identified by others. Multidimensional Perfectionism Scale (F-MPS- Frost. Marten. Lahart & Rosenblate. 1990) The two MPS scales were developed concurrently, although Frost and colleagues published first in 1990. Frost et al. also speculated that perfectionism was a multidimensional construct; however, they hypothesized five relevant and somewhat separate dimensions including: personal standards; concern over mistakes; parental expectations; doubts about action; and organization. Initial item selection was based on the content of pre-existing perfectionism scales (Burns Perfectionism Scale, Burns 1980; the Eating Disorder Inventory, Garner et al, 1983; and Obsessionality, Rachman & Hodgson, 1980) and a large number of new items were generated. Initially, the scale was developed using only female university students at Smith College (Boston) in the United States. 36 In their initial study, Frost and colleagues (1990) conducted a factor analysis which yielded 6 factors, accounting for 54% of the total variance. The factors roughly corresponded to the hypothesized dimensions, with the exception of parental expectations which could be broken down into two distinct factors: parental expectations; and parental criticism. All factors were highly intercorrelated with the exception of the organization subscale. From examining the scale structure, and the relationship between the subscales and other measures of perfectionism, Frost et al (1990) concluded that the major dimension was concern over mistakes. They found that the organization subscale was least related to the other dimensions and other measures of perfectionism. Henceforward, the organization subscale, while retained and viewed as a separate factor, has been dropped from the calculation of total perfectionism. Currently, the scale is comprised of 35 items and 6 dimensions: personal standards (7 items); concern over mistakes (9 items); parental expectations (5 items); parental criticism (4 items); doubts about actions (4 items); and organization (6 items). All items are summed to obtain a total perfectionism score (5 dimensions) and items pertaining to each subscale are summed to calculate subscale scores. Clavin, Clavin, Gayton and Broida (1996) sought to extend the reliability and validity of the F-MPS to men. This study combined with the initial study by Frost and colleagues yielded the following psychometric data. Internal consistency coefficients that have been reported ranged from: .81 - .91 for total perfectionism; .88 - .91 for concern over mistakes; .81 - .83 for personal standards; .82- 84 for parental expectations; .77 - .84 for parental criticism; .77 - .79 for doubts about actions; and .93 -.94 for organization. Thus, the internal consistency coefficients indicate that the items on 37 the scale hang together quite well. No test-retest data could be found for this scale, so there is no presently available data on the temporal stability of the measure. Frost et al (1990) found some evidence of concurrent validity, reporting a correlation of .85 with the Burns Perfectionism Scale. However, it is likely that this relationship is inflated given some of the items on the F-MPS were initially derived from the Burns Perfectionism Scale. In 1993, Frost, Heimberg, Holt, Mattia and Neubauer investigated the relationship between the two MPS measures and their relationship with positive and negative affect. They found substantial overlap between Frost et al.'s total perfectionism scale and Hewitt et al.'s self-oriented and socially prescribed perfectionism scales, thus providing evidence that the two are closely related. There was a correlation between total perfectionism (F-MPS) and the other-oriented perfectionism scale (H-MPS), but it was much smaller in magnitude. The Frost measure appears to reflect a global dimension containing elements of both self-oriented and socially prescribed perfectionism. Additional evidence of construct validity comes from studies that examined the relationship between perfectionism and various measures of personality as well as psychopathology and procrastination (Frost, 1990). It was found that some of the dimensions were differentially related to symptomatology and psychopathology in the expected directions. Clavin et al (1996) reported that the F-MPS was weakly correlated with social desirability (r = -.20, p < 05). Evidence Regarding the Use of Perfectionism Measures with a Chinese Sample Although Chang (1998) investigated perfectionism and suicidal risks in Caucasian and Asian-American college students, using the F-MPS, information pertaining to the 38 psychometric properties of the F-MPS was not reported for the Asian sample. No similar data on Asian students appears to be available for the H-MPS, nor does data specifically related to Chinese students, a subgroup within the broad Asian category. Rational for the Use of Two Perfectionism Measures in the Current Study Given the psychometric evidence provided for both MPSs, there is strong evidence that the perfectionism construct is, in fact, multidimensional. Thus, measures such as the Burns Perfectionism Scale (Burns, 1980), reflecting a unidimensional conceptualization of perfectionism will not be considered for this study. Given that the construction of the H-MPS was more rigorous and the reported psychometric properties are stronger, it will be used as the primary perfectionism measure. For example, the F-MPS was developed using items from pre-existing scales reflecting a unidimensional conceptualization of perfectionism and then correlated with those same scales to present evidence of concurrent validity. In addition, no test-retest data is available, it is weakly correlated with social desirability, and the measure does not reflect an important and empirically validated dimension, that of other-oriented perfectionism, that is included in the Hewitt et al. measure. Given that the latter scale has been developed and has undergone significant validation using large numbers of both male and female university students in Canada, it is the preferred measure. However, given that the F-MPS has been used and validated on Asian Americans, unlike the H-MPS, which does not appear to be validated for Asian or Chinese Canadians, it is important to use both measures in the present study. 39 Beck Depression Inventory The Beck Depression Inventory (BDI) is a widely used self-report measure of depression (Beck, Rush, Shaw & Emery, 1979). The BDI contains 21 items that respondents rate on a Likert-type scale of 0 to 3 in terms of intensity. An overall depression severity score can be calculatated. Beck, Steer and Garbin (1988) conducted a meta-analysis of studies that were done between 1961 and 1986. They reported a mean internal consistency coefficient of .86 for a clinical sample, and .81 for a non-clinical sample. Test-retest correlations of .48 to .86 for a clinical population and .60 to .83 for a non-clinical population were reported for time intervals ranging from hours to two weeks. They also estimated concurrent validity of .60 or higher for the relationship with other assessments of depression including clinical assessments and other self-report measures of depression. Factor analytic studies have indicated that the BDI reflects a general depression syndrome comprised of three primary factors: negative attitudes; performance difficulties; and somatic complaints. Some studies, however, have found different numbers of factors. While Beck's claim of multidimensionality has been confirmed, there is still some conflicting evidence regarding the factor structure. Evidence Regarding the Use of the BDI with a Chinese Student Sample Several researchers have investigated the applicability of the BDI using non-Western subjects. Tashakkor and Barefoot (1989) have confirmed the usefulness of the BDI as a measure of depression/maladaptive functioning in college student populations of Iranian descent. The BDI has been translated into several languages including Mandarin and Cantonese. Several investigators have researched the Chinese-BDI (C-40 BDI) and found the translated version to have adequate reliability and validity using clinical and nonclinical samples (Shek, 1990; Shek, 1991; Zheng, Wei, Lianggue, Guochen & Chenggue, 1988; Chan & Tsoi, 1984; Chan, 1991). Chan (1991) compared the C-BDI and the BDI (English) using a sample of bilingual undergraduate students. With respect to the reliability and integrity of the English version for Chinese respondents, Chan reported an internal consistency coefficient of .87, a mean interitem correlation of .24, and a range of .21 - .60 for corrected item-total correlations. Chan also found some evidence of convergent validity in that he reported a correlation of .74 with another measure of depression, the CES-D. Chan (1991) also conducted an item factor analysis and found a clean two-factor solution. Factor 1 was loaded by cognitive and affective items, and Factor 2 was loaded by items pertaining to fatigue and insomnia (somatic complaints). Other studies that have investigated the factor structure of the C-BDI in China reported very similar findings (Chan & Tsoi, 1984; Shek, 1990). Overall, there appears to be sufficient evidence to use the BDI as a general measure of depression severity with 1st and 2nd generation Chinese-Canadians. Suinn-Lew Asian Self-Identity Acculturation Scale Prior to the development of this scale, no objective measure of acculturation tailored specifically for Asian individuals existed. The SL-ASIA Scale was modelled after the Acculturation Rating Scale for Mexican Americans (ARMS A - Cuellar, Harris, Jasso, 1980). The scale was developed to reflect the multidimensional nature of acculturation, taking into account the possibility of bicultural development (ie., a score of 3 or so out of 5). It is a 21-item multiple choice scale which assesses cognitive, attitudinal and behaviour domains including: language (4 items); identity (4 items); 41 friendship choice (4 items); behaviours (5 items); generation/geographic history (3 items); attitudes (1 item). The scale includes both actual behaviours and ideals or preferences. In scoring the scale, a total value is obtained by summing all item scores and dividing by the total number of items yielding a score of 1.00 - 5.00. Higher scores indicate higher levels of acculturation, or a Western identification, with lower scores indicative of lower levels of acculturation, or Asian identification. Four studies reporting on the psychometric properties of the SL-ASIA for student and non-student populations, including the factor structure, were located in the literature (Suinn et al, 1987; Atkinson & Gim, 1992; Suinn, Ahuna, & Khoo, 1992; Ownbey & Horridge, 1998). All studies used Asian-American samples. Overall, the reported internal consistency ranged from .88 to .99. With respect to concurrent validity, two studies using demographic variables produced similar findings. The demographic variables having the strongest relationships with acculturation level were: years of school attendance in the U.S. and age upon beginning school in the U.S. Additional studies have presented evidence of factorial validity (Suinn et al, 1992; Ownbey et al, 1998). Background Information Sheet The 21-item Background Information Sheet was used to obtain background information germane to the present study and demographic information relevant to the validation of the SL-ASIA scale using a Chinese-Canadian sample (See Appendix F). Analytic Procedures Comparisons of internal consistency and correlation coefficients between the Chinese and European samples were conducted using a two-tailed Fisher z prime transformation test. Comparisons between the two samples in mean levels of 42 perfectionism were conducted using 2-tailed t-tests or ANOVAs. Where multiple t-tests were performed, the Bonferroni correction was used to adjust the significance level in order to minimize the chance of making a Type I error. Given the researcher did not have sufficient grounds for stating directional hypotheses, two-tailed tests were used, allowing the researcher to determine the significance level of differences between means in either direction. The patterns and ordering of the correlation coefficients were analyzed by means of visual inspection. Hypotheses Hypothesis 1: Hypothesis 1 investigated internal evidence of construct validity of the H-MPS using a Chinese sample. 1 - A Magnitudes of internal consistency coefficients HI: Differences in the internal consistencies of the H-MPS scales (total and subscales) between the Chinese and European samples, analyzed by comparing the magnitudes of the chronbach alpha coefficients using a two-tailed Fisher z prime transformation test, will be statistically significant at p< .05. HO: Differences in the internal consistencies of the H-MPS scales (total and subscales) between the Chinese and European samples, analyzed by comparing the magnitudes of the chronbach alpha coefficients using a two-tailed Fisher z prime transformation test, will not be statistically significant at p< .05. 1-B Ordering of the internal consistency coefficients 43 HI: The ordering of the magnitude of the chronbach alpha coefficients among the H-MPS subscales, analyzed by descriptively rank-ordering them, will be different for the Chinese and European samples. HO: The ordering of the magnitude of the chronbach alpha coefficients among the H-MPS subscales, analyzed by descriptively rank-ordering them, will not be different for the Chinese and European samples. 1 -C Magnitudes of intercorrelations HI: Differences in the magnitudes of the correlations among the H-MPS subscales between the Chinese and European samples, analyzed by means of a two-tailed Fisher z prime transformation test, will be statistically significant at p< .05. HO: Differences in the magnitudes of the correlations among the H-MPS subscales between the Chinese and European samples, analyzed by means of a two-tailed Fisher z prime transformation test, will not be statistically significant at p< .05. 1-D Patterns of intercorrelations HI: Patterns of the correlations among the H-MPS subscales, analyzed by descriptively comparing the ordering and direction of the intercorrelations, will be different for the Chinese and European samples. HO: Patterns of the correlations among the H-MPS subscales, analyzed by descriptively comparing the ordering and direction of the intercorrelations, will not be different for the Chinese and European samples. Hypothesis 2 Hypothesis 2 investigated the external evidence of construct validity for the H-MPS using a Chinese sample. 44 2-A Magnitudes of correlations HI: Differences in the magnitudes of the correlations between the H-MPS scales and the BDI between the Chinese and European samples, analyzed by means of a two-tailed Fisher z prime transformation test, will be statistically significant at p< .05. HO. Differences in the magnitudes of the correlations between the H-MPS scales and the BDI for the Chinese and European samples, analyzed by means of a two-tailed Fisher z prime transformation test, will not be statistically significant at p< .05. 2-B Patterns of correlations HI: Patterns of correlations between the H-MPS scales and the BDL analyzed by descriptively comparing the ordering and the direction of the correlations, will be different for the Chinese and European samples. HO. Patterns of correlations between the H-MPS scales and the BDI, analyzed by descriptively comparing the ordering and direction of the correlations, will not be different for the Chinese and European samples. 2-C Magnitudes of correlations HI: Differences in the magnitudes of the correlations between the H-MPS scales and the F-MPS between the Chinese and European samples, analyzed by means of a two-tailed Fisher z prime transformation test, will be statistically significant at p< .05. HO. Differences in the magnitudes of the correlations between the H-MPS scales and the F-MPS between the Chinese and European samples, analyzed by means of a two-tailed Fisher z prime transformation test, will not be statistically significant at p< .05. 2-D Patterns of correlations HI: Patterns of correlations between the H-MPS scales and the F-MPS (total scale), analyzed by descriptively comparing the ordering and direction of the correlations, will be different for the Chinese and European samples. HO: Patterns of correlations between the H-MPS scales and the F-MPS (total scale), analyzed by descriptively comparing the ordering and direction of the correlations , will not be different for the Chinese and European samples. Hypothesis 3: Hypothesis 3 investigated differences in mean levels of perfectionism between Chinese and European samples. HI: Differences in mean levels of perfectionism using the H-MPS scales (total and subscales) between the Chinese and European samples, analyzed by means of two-tailed t-tests, will be statistically significant at p< .05. HO: Differences in mean levels of perfectionism using the H-MPS scales (total and subscales) for the Chinese and European samples, analyzed by means of two-tailed t-tests, will not be statistically significant at p< .05. 46 Chapter IV Results Results of the analyses are presented for each hypothesis. SPSS 8.0 for Windows was used to calculate the reliability coefficients, the Pearson-product-moment correlations, and the t-tests and ANOVAs comparing mean levels of perfectionism for the Chinese and European samples. The z-tests comparing correlation coefficients were calculated manually, and the patterns and ordering of the correlation coefficients were analyzed by means of visual inspection. Hypothesis 1 Hypothesis 1 investigated the internal evidence of construct validity for the H-MPS. Hypotheses 1-A and 1-B addressed the comparisons between the Chinese and European samples with respect to the magnitudes (1-A) and ordering of the magnitudes (1-B) of the internal consistency coefficients using the total and subscale scores of the Multidimensional Perfectionism Scale (H-MPS, Hewitt & Flett, 1991). Table 2 reports these results. Upon visual inspection of the chronbach alpha coefficients, it was found that all of the coefficients for both the Chinese and European samples were larger than .80, and some were larger than .90. This suggests that the H-MPS demonstrated a satisfactory level of internal consistency for both groups. When the magnitudes of the chronbach alpha coefficients were compared, using a two-tailed Fisher z prime transformation test, none of the differences reached statistical significance. Given the lack of significance, no 47 differences in the internal consistency of the H-MPS were indicated between the two groups. When comparing the rank ordering of the magnitudes of the alpha coefficients for the three subscales within the Chinese and European samples, no differences were identified. Therefore, the patterns of internal consistency were similar for the two samples. Given that no differences were found between the Chinese and European samples in the internal consistency of the H-MPS or the ordering of the internal consistency coefficients, a similarity of internal structure is indicated for the two groups. Further comparisons of internal structure were investigated in Hypotheses 1-C and 1-D.. Hypotheses 1-C and 1-D addressed the comparisons between the Chinese and European samples with respect to the magnitudes (1-C) and patterns (1-D) of the correlations among the H-MPS subscales. In examining the Pearson correlation coefficients among the H-MPS subscales, it was found that all of the coefficients were positive and significant at p< .01 for both the Chinese and European samples. Table 3 lists the Pearson correlation coefficients for the two samples. It was also noted that, qualitatively, all of the intercorrelations were larger in magnitude for the European group, with a range of .43 - .55 and.28 - .43 for the European and Chinese samples, respectively. When the magnitudes of the correlation coefficients were compared, using a two-tailed Fisher z prime transformation test, none of these differences were statistically significant. That means that observed differences between the groups in the magnitudes of the intercorrelations among the H-MPS subscales cannot be attributed to real differences and may be the result of sampling error 48 Table 2 Comoarison of internal consistency for the Chinese and European Samples Scale Chinese Alpha coefficient N European Alpha coefficient N z P H-MPS Total 0.91 111 0.93 71 -0.84 0.4 Self-oriented perfectionism 0.92 113 0.91 73 0.4 1.31 Other-oriented perfectionism 0.81 114 0.82 73 -0.2 0.84 Socially-prescribed perfectionism 0.83 111 0.87 71 -1.24 0.22 * p< .05, 2 tailed for z value =1.96 49 However, in comparing the magnitudes of the correlation coefficients for self-oriented and socially-prescribed perfectionism subscales, the z value of -1.692 is associated with a p value of .05 for a one-tailed test. Table 4 reports these results and lists the rank ordering of the intercorrelations among the H-MPS subscales for the two samples. In comparing the rank ordering of the magnitudes of the correlations among the H-MPS subscales within the Chinese and European samples, differences emerged. Self-oriented perfectionism and other-oriented perfectionism were most highly correlated for both groups, followed by self-oriented and socially-prescribed perfectionism, and then by other-oriented and socially-prescribed perfectionism for the Chinese sample. For the two latter correlations the reverse was true for the European sample. While finding no real differences in the magnitudes, direction, and levels of significance of the intercorrelations among the H-MPS subscales indicate a similarity of internal structure, differences in the rank order of these intercorrelations, namely ranks 2 and 3, suggest that the structures may be slightly different. Given the above findings suggest that the structure of the perfectionism construct may be somewhat different for the Chinese sample, correlations between the total perfectionism scale and the subscales of the H-MPS were compared for the two samples, using a two-tailed Fisher z prime transformation test. Table 5 reports these results. The magnitudes of the correlation between the H-MPS total scale and the socially-prescribed perfectionism subscale were found to be statistically significant, with the Chinese group demonstrating a weaker correlation. This indicates that the socially-50 Table 3 Pearson-product-moment correlations among the H-MPS subscales for the Chinese and European samples Perfectionism scale Self Other Social Chinese Self 0.55** 0.50** Other 0.43** 0.43** Social 0.28** 0.32** European **correlations significant at p< .01 Table 4 Comparisons of the magnitudes of the intercorrelations among the H-MPS subscales and rank orders of the subscales for the Chinese and European samples Rank Order Paired Subscales z p Chinese European Self-Other -1.08 0.28 1 1 Self-Social -1.69 0.09 2 3 Other-Social -0.81 0.42 3 2 Note abbreviations: Self, Self-oriented perfectionism; Other, Other-oriented perfectionism, Social, Socially-prescribed perfectionism All differences were not significant at p< .05, 2 tailed for z value =1.96 51 Table 5 Pearson-product-moment correlations between the H-MPS total scale and the H-MPS subscales and comparisons of the magnitudes of the correlations between the Chinese and European samples Perfectionism subscale Chinese (N= 116) European (N=74) z p H-MPS Total H-MPS Total Self 0.81** 0.85** -0.85 0.4 Other . 0.75** 0.78** -0.48 0.64 Social 0.68** 0.81** -1.97 0.04* ** correlations reached significance at p<01 *p<05, 2-tailed for z value 1.96 52 prescribed perfectionism subscale is less related to the overall construct for the Chinese sample as compared to the European sample. Hypothesis 2 Hypothesis 2 investigated the external evidence of construct validity for the H-MPS. Hypotheses 2-A and 2-B addressed the comparisons between the Chinese and European samples with respect to the magnitudes (2-A) and patterns (2-B) of correlations between the H-MPS scales and the Beck Depression Inventory (BDI, Beck et. al, 1961). Table 6 reports these results. Upon visual inspection of all the Pearson correlation coefficients between the H-MPS scales and the BDI, all correlations were positive and larger in magnitude for the European group as compared to the corresponding coefficients for the Chinese group. Only the correlations between socially-prescribed perfectionism and depression and between total perfectionism and depression reached significance for both samples. When the magnitudes of the correlation coefficients were compared for the two samples, using a two-tailed Fisher z prime transformation test, statistically significant differences were found for the correlation between socially-prescribed perfectionism and depression and for the correlation between total perfectionism and depression. In both cases, the magnitude of the correlation was significantly larger for the European sample, which indicates there is a stronger relationship with depression for the European sample as compared to the Chinese sample. When comparing the ordering of the magnitudes of the correlations between the four perfectionism scales and depression within each sample, by rank ordering them, no differences were identified. While no differences between the Chinese and European 53 Table 6 Pearson-product-moment correlations between the H-MPS scales and the BDI and comparisons of the magnitudes of the correlations between the Chinese and European samples Perfectionism Scale Chinese (N= 110) European (N=65) z £ BDI BDI Self 0.11 0.17 -0.39 0.7 Other 0.07 0.09 -0.16 0.87 Social 0.27** 0.53** -1.96 0.05*** Total 0.19* 0.34** -2.1 0Q4*** Note abbreviations: Self, self-oriented perfectionism; Other, other-oriented perfectionism; Social, socially-prescribed •correlations reached significance at p<05 ••correlations reached significance at p<01 ***p<05,2-tailed for z value 1.96 54 Table 7 Pearson-product-moment correlations between the H-MPS scales and the F-MPS and comparisons of the magnitudes of the correlations between the Chinese and European samples Perfectionism Scale Chinese (N= 116) European (N=74) z p F-MPS F-MPS Self 0.58** 0.67** -0.98 0.33 Other 0.40** 0.46** -0.44 0.66 Social 0.69** 0.77** -1.12 0.26 Total 0.74** 0.79** -0.8 0.42 •correlations reached significance at p< 01 ***p <05, 2-tailed for z value 1.96 55 samples were found in the patterns of correlation between the perfectionism scales and depression, significant differences were found in the magnitudes of the correlation between socially prescribed perfectionism and depression and between total perfectionism and depression. Hypotheses 2-C and 2-D also investigated external evidence of construct validity for the H-MPS. Hypothesis 2-C and 2-D addressed the comparisons between the Chinese and European samples with respect the magnitudes (2-C) and patterns (2-D) of correlation between the H-MPS scales and a second perfectionism measure, the Multidimensional Perfectionism Scale (F-MPS, Frost et. al, 1990). Table 7 reports these results. Upon visual inspection of the Pearson correlation coefficients between the H-MPS and the F-MPS, all coefficients were positive and significant at p< .01 for both samples. The magnitudes of all the correlation coefficients were larger for the European group, as compared to the corresponding coefficient for the Chinese group. When the magnitudes of the correlation coefficients were compared for the two groups, using a two-tailed Fisher z prime transformation test, none of these differences reached statistical significance. That means that observed differences in the magnitudes of the correlation coefficients cannot be attributed to real differences and may be an artifact of sampling error. In comparing the rank ordering of the magnitudes of the correlation coefficients within each group, no differences were identified. Therefore, no differences in the patterns or magnitudes of the correlations between the H-MPS and the F-MPS were found between the Chinese and European samples. 56 Hypothesis 3. The third hypothesis addressed whether mean levels of perfectionism (total and subscales) differed for the European and Chinese samples. Prior to testing the third hypothesis, it was necessary to undertake two preliminary analyses. Preliminary Analyses. It was important to determine if participants who completed questionnaires using the web-site differed from those who completed the questionnaires using hard copy forms. No differences would indicate that data for the two groups could be analyzed together, rather than undertaking separate analyses for each group. Two-tailed t-tests for independent groups were used to determine if the web-site and hard-copy groups differed on the following study variables: self-oriented perfectionism, socially-prescribed perfectionism, other-oriented perfectionism, total perfectionism (H-MPS), total perfectionism (F-MPS), and depression. Table 8 lists the means and standard deviations for the above-mentioned study variables. When mean scores for each variable were compared for the two groups, no statistically significant differences were found. Table 9 reports these results. Therefore, data for the two groups were analyzed together for hypothesis 3. Prior to addressing the second hypothesis, it was also important to test whether mean levels of perfectionism differed for males and females within each cultural group. No differences would indicate that data for males and females within each group could be analyzed together rather than separately. Table 10 lists the means and standard deviations for each of the three perfectionism subscales and for the total perfectionism scale for males and females within the Chinese sample. 57 Table 8 Means and standard deviations for study variables for website and hard-copy participants Hard Copy (N=l 38) Website (N=53) Variable Mean SD Mean SD Self 4.7 1.01 4.64 1.09 Other 3.76 0.82 3.8 0.82 Social 3.6 0.88 3.57 0.9 H-MPS Total 4.02 0.7 4 0.74 F-MPS Total 3.08 0.57 3.04 0.6 BDI 0.63 0.53 0.6 0.33 Table 9 Comparison of mean scores for study variables for website and hard-copy participants Variable _ t df p Self 0.33 189 0.74 Other -0.33 189 0.74 Social 0.24 189 0.81 H-MPS Total 0.13 189 0.9 F-MPS Total 0.43 189 0.67 BDI 0.34 173 0.74 p< .05, 2-tailed 58 When mean levels of perfectionism were compared for males and females within the Chinese group, using two-tailed t-tests, no statistically significant differences were found. Table 11 reports these results. Therefore, data for males and females in the Chinese sample were analyzed together. When mean levels of perfectionism were compared for males and females within the European group, using two-tailed t-tests, a statistically significant difference was found for the socially-prescribed perfectionism subscale, with females scoring significantly higher than males. No significant differences were found for the other perfectionism scales. Table 12 lists the means and standard deviations for the perfectionism scales for males and females within the European sample. Table 13 reports these results. Therefore, data for males and females in the European sample were analyzed separately for the socially prescribed perfectionism subscale, and data were analyzed together for the remaining perfectionism scales. When mean levels of self-oriented, other-oriented and total perfectionism were compared for the Chinese and European samples, using two-tailed t-tests, no statistically significant results were found. Table 14 lists the means and standard deviations for the European and Chinese samples and Table 15 reports these results. Therefore, no real differences in self-oriented perfectionism, other-oriented perfectionism, and total perfectionism between the Chinese and European groups were identified in the current study. Given that gender differences in mean levels of perfectionism were found for socially- prescribed perfectionism within the European group, mean levels of socially-prescribed perfectionism were compared for males and females in the Chinese and 59 European groups using a one-way ANOVA for four groups. Table 16 lists the means and standard deviations for the socially-prescribed perfectionism subscale. In comparing the four groups, statistical significance was reached, clearly indicating a difference among the means. Table 17 reports these results. In order to identify which pairs of means differed significantly, multiple t-tests were conducted using the Bonferroni correction to adjust the significance level in order to minimize the chance of making a Type I error. Differences in mean levels of socially-prescribed perfectionism reached statistical significance only for males and females within the European group. However, differences in the means for European males and Chinese females were associated with a p value of .06. Table 18 reports these results. No differences in mean levels of perfectionism were identified between the Chinese and European samples in the current study. While no gender differences were found within the Chinese sample across any of the perfectionism scales, European women were found to have significantly higher mean levels of socially prescribed perfectionism as compared to their male counterparts. Post Hoc Analyses Hypothesis 2-A addressed the comparisons between the Chinese and European samples with respect to the magnitudes of the correlation between the H-MPS scales and the BDI. Given that males in the European group demonstrated significantly lower mean levels of socially-prescribed perfectionism than females, hypothesis 2-A was reanalyzed by splitting the two samples by gender and comparing the means of these groups. 60 Table 10 Means and standard deviations for the H-MPS scales for males and females in the Chinese sample Female Male Perfectionism Scale Mean SD Mean SD Self 4.6 0.97 4.91 1.19 Other 3.63 0.75 3.77 0.96 Social 3.68 0.75 3.5 0.98 Total 3.97 0.6 4.06 0.82 Table 11 Comparison of mean levels of perfectionism between males and females in the Chinese sample Perfectionism Scale t df p Self 1.53 115 0.13 Other 0.86 115 0.39 Social -1.07 115 0.29 H-MPS Total 0.7 115 0.48 p < .05, 2-tailed 61 Table 12 Means and standard deviations for the H-MPS scales for males and females in the European sample Female (N=40) Male (N=34) Perfectionism Scale Mean SD Mean SD Self 4.74 0.95 4.55 1.06 Other 3.97 0.68 3.84 0.92 Social 3.82 0.99 3.23 0.83 H-MPS Total 4.17 0.69 3.87 0.79 Table 13 Perfectionism Scale sample t df P Self -0.79 72 0.43 Other -0.69 72 0.49 Social -2.8 72 0.01** H-MPS Total -1.75 72 0.09 **p<01, 2tailed 62 Table 14 Means and standard deviations for the MPS scales for the Chinese and European samples Perfectionism Scale Group Mean SD N Self Chinese 4.7 1.05 117 European 4.65 1 74 Other Chinese 3.68 0.82 117 European 3.91 0.8 74 Social Chinese 3.62 0.83 117 European 3.55 0.96 74 H-MPS Total Chinese 4 0.68 117 European 4.03 0.75 74 Table 15 Comparisons of mean levels of self-oriented, other-oriented, and total perfectionism between the Chinese and European samples Perfectionism Scale t df p Self 0.35 189 0.73 Other -1.9 189 0.06 H-MPS Total -0.32 189 0.75 p<05, 2tailed 63 Table 16 Means and standard deviations for socially-prescribed perfectionism for males and females from the European and Chinese samples Group Mean SD N European males 3.23 0.83 34 European females 3.82 0.99 40 Chinese males 3.5 0.98 39 Chinese females 3.68 0.75 78 Table 17 Comparison of mean levels of socially-prescribed perfectionism between males and females in the European and Chinese samples Socially-prescribed perfectionism F df p Between Groups 3_3 3 .02* * significant at p< .05 64 Table 18 Multiple comparisons of mean levels of socially-prescribed perfectionism between males and females in the European and Chinese sample Groups p European males European females 0.02* Chinese males 0.53 Chinese females 0.06 European females Chinese males 0.37 Chinese females 0.84 Chinese males Chinese females 0.73 *mean difference is significant at p< .05 65 When the magnitudes of the correlation coefficients were compared for the four groups, using two-tailed Fisher z prime transformation tests, statistically significant differences were not found. However, the difference between European and Chinese females was associated with a z value of -1.83 which is associated with a p value of .03 for a one tail test. Tables 19 and 20 report these results. Therefore, observed differences between the groups in the relationship between socially-prescribed perfectionism and depression cannot be attributed to real differences and may be due to sampling error. Given no statistically significant differences in mean levels of perfectionism were found between the Chinese and European samples in the current study, the researcher was curious whether potential differences may have been masked by the heterogeneity of the two samples. Since data on generation status and level of acculturation were collected for the Chinese sample, comparisons of mean levels of perfectionism between samples of first and second generation Chinese students and between high and low acculturation groups within the Chinese sample were made. The following hypotheses were added to the study: Hypothesis 4; HI: Differences in mean levels of perfectionism between first and second generation Chinese samples , analyzed by means of a two-tailed t-test, will be statistically significant at p< .05. HO: Differences in mean levels of perfectionism between first and second generation Chinese samples, analyzed by means of a two-tailed t-test, will not be statistically significant at p< .05. 66 Table 19 Pearson-product moment correlations between the socially-prescribed perfectionism subscale and the BDI for males and females from the European and Chinese samples European Males European Females Chinese Males Chinese Females Perfectionism Scale BDI (N=28) BDI (N=37) BDI (N=35) BDI (N=75) Socially-prescribed 0.43* 0.52** 0.42* 0.19 Perfectionism ** correlations reached significance at p<01 *correlations reached significance at p< .05 Table 20 Comparisons of the magnitudes of the correlation between the socially-prescribed perfectionism subscale and the BDI between males and females from the European and Chinese samples Groups z p European males European females Chinese males Chinese females -0.44 -0.04 -1.14 0.66 0.96 0.25 European females Chinese males Chinese females -0.52 -1.83 0.61 0.07 Chinese males Chinese females -1.18 0.24 p < .05, 2-tailed 67 Hypothesis 5: HI: Differences in mean levels of perfectionism between low and high acculturation groups from the Chinese sample, analyzed by means of a two-tailed t-test, will be statistically significant at p < 05. HO: Differences in mean levels of perfectionism between low and high acculturation groups from the Chinese sample, analyzed by means of a two-tailed t-test, will not be statistically significant at p <05. Hypothesis 4 addressed whether mean levels of perfectionism differed for first and second generation groups within the Chinese sample. When mean levels of self-oriented, other-oriented, socially-prescribed and total perfectionism were compared for the two Chinese samples, using two-tailed t-tests, statistically significant differences were found only for the socially-prescribed perfectionism means, with the second-generation group scoring higher than the first-generation group. Table 21 lists the perfectionism means and standard deviations for both groups and Table 22 reports the results. The researcher deemed it was important to contextualize these findings by comparing the means with the European sample for the socially-prescribed perfectionism subscale. Therefore, an additional hypothesis was included in the study, hypothesis 4b. Since gender differences in mean levels of socially-prescribed perfectionism were found for the European sample, differences were analyzed by gender. Hypothesis 4b: HI: Differences in mean levels of perfectionism between males and females from first generation Chinese, second generation Chinese and European samples, analyzed by 68 means of a one-way ANOVA, will be statistically significant at p< .05. HO: Differences in mean levels of perfectionism between males and females from first generation Chinese, second generation Chinese and European samples, analyzed by means of a one-way ANOVA, will not be statistically significant at p< .05. When mean levels of socially-prescribed perfectionism were compared for males and females from first generation Chinese, second generation Chinese and European samples, using a one-way ANOVA, statistical significance was reached, clearly indicating a difference among the means. Table 23 lists the means and standard deviations and Table 24 reports these results. In order to identify which pairs of means differed significantly multiple t-tests using the Bonferroni correction were conducted. Significant differences in mean levels of socially prescribed perfectionism were found between European males and European females and between European males and second-generation Chinese females, with European males scoring significantly lower in both cases. Table 25 reports these results. Hypothesis 5 addressed whether mean levels of perfectionism differed for the high and low acculturation groups from the Chinese sample. Prior to testing this hypothesis, two preliminary analyses were required. Given the SL-ASIA scale was developed and validated using samples of Asian-Americans, it was important to establish that the measure demonstrated adequate internal consistency for the current sample of Chinese-Canadians. Chronabah alpha coefficients were calculated to check the internal consistency. A high internal consistency coefficient, r = .89, was found for the Chinese-Canadian sample (N= 117) in the current study, suggesting the measure can be used with sufficient reliability. In order to justify splitting the Chinese sample into high and low 69 Table 21 Means and standard deviations for the MPS scales for first and second generation Chinese samples Perfectionism Scale Generation Mean SD N Self 1 4.75 1.1 64 2 4.61 0.99 47 Other 1 3.63 0.91 64 2 3.73 0.68 47 Social 1 3.44 0.83 64 2 3.82 0.76 47 Total 1 3.94 0.71 64 2 4.05 0.59 47 Table 22 Comparisons of mean levels of perfectionism between first and second generation Perfectionism Scale t df P Self 0.73 109 0.47 Other -0.63 109 0.51 Social -2.48 109 0.02* Total -0.88 109 0.38 *p< .05,2-tailed Table 23 Means and standard deviations for the socially prescribed for males and females from first generation Chinese, second-generation Chinese and European samples Group Mean SD N European males 3.2 0.85 32 European females 3.81 0.97 42 First generation Chinesse males 3.2 0.97 21 First generation Chinese females 3.55 0.74 43 Second generation Chinese males 3.7 0.81 14 Second generation Chinese females 3.86 0.73 34 Table 24 Comparison of means for socially-prescribed perfectionism between males and females from first generation Chinese, second generation Chinese and European samples Socially-prescribed perfectionism F df p Between Groups 2.96 6 0.01** **mean difference is significant at p< .01 71 Table 25 Multiple comparisons of mean levels of socially-prescribed perfectionism between males and females from first generation Chinese, second generation Chinese and European samples Groups p European males European females 0.04* First generation Chinese males 0.99 First generation Chinese females 0.59 Second generation Chinese males 0.53 Second generation Chinese females 0.03* European females First generation Chinese males 0.11 First generation Chinese females 0.8 Second generation Chinese males 0.99 Second generation Chinese females 0.99 First generation Chinese males First generation Chinese females 0.73 Second generation Chinese males 0.62 Second generation Chinese females 0.08 First generation Chinese females Second generation Chinese males 0.99 Second generation Chinese females 0.68 Second generation Chinese males Second generation Chinese females 0.99 *mean difference is significant at p< .05 72 acculturation groups using the SL-ASIA, the distribution was checked for normality by calculating the ratio of the skewness and kurtosis statistics to the standard error. Normality is rejected if this ratio is less than -2 or greater than +2. Given the skewness for the current sample was found to be .06 and the kurtosis was found to be -.892, normality of the distribution was assumed. Therefore, the distribution was split equally into three groups, with the bottom third reflecting low acculturation or "Asian identification" (Suinn et al, 1987) and the top third reflecting high acculturation or "Western identification" (Suinn et al, 1987). The middle third of the distribution, reflecting a "bicultural identification" (Suinn et al, 1987) was excluded from the analysis of Hypothesis 5. When mean levels of perfectionism (total and subscales) were compared for the low and high acculturation groups from the Chinese sample, using two-tailed t-tests, no statistically significant differences were found. Table 26 lists the means and standard deviations and Table 27 reports these results. Given no differences in self-oriented, other-oriented, socially-prescribed or total perfectionism between low and high acculturation groups were identified in the current study, comparisons with the European group were not warranted. The results of the current study can be summarized as follows: no significant differences in the internal consistency of the H-MPS were found between the two groups; no significant differences in the magnitudes of the intercorrelations among the three H-MPS subscales were identified between the two groups, although a slight difference in the rank order of the intercorrelations was identified (ranks 2 and 3); the magnitudes of the correlations between socially-prescribed perfectionism and depression and between 73 total perfectionism and depression were found to be significantly larger for the European group, although no differences in the ordering of the correlations were identified; no significant differences in the magnitudes or patterns of the correlations between the H-MPS and the F-MPS were found between the Chinese and European samples; and no significant differences in mean levels of perfectionism were found between the European and Chinese samples, although European men were found to have significantly lower mean levels of socially prescribed perfectionism as compared to European women. The post hoc results can be summarized as follows: differences in the magnitudes of the correlations between socially-prescribed and depression between males and females from the European and Chinese samples, were not identified; differences in mean levels of socially-prescribed perfectionism were found between first and second generation groups from the Chinese sample, with the second generation group scoring higher; differences in mean levels of socially-prescribed perfectionism were found between European men and second generation Chinese females, with European men scoring lower; and differences in mean levels of perfectionism were not identified between high and low acculturation groups from the Chinese sample. 74 Table 26 Means and standard deviations for the MPS scales for high and low acculturation groups from the Chinese sample Perfectionism Scale Group Mean SD N Self Other Social Total Chinese (low) Chinese (high) Chinese (low) Chinese (high) Chinese (low) Chinese (high) Chinese (low) Chinese (high) 4.7 4.57 3.68 3.74 3.44 3.73 3.94 4.01 1.08 1.16 0.92 0.74 0.83 0.8 0.74 0.7 41 40 41 40 41 40 41 40 Table 27 Comparison of mean levels of perfectionism between the low and high acculturation groups from the Chinese sample Perfectionism Scale t df p Self Other Social 0.53 -0.33 • 1.56 79 79 79 0.59 0.74 0.12 Total -0.44 79 0.66 p< .05, 2-tailed 75 Chapter V Discussion The aim of the current study was to determine if the Multidimensional Perfectionism Scale (H-MPS) showed evidence of construct validity in a sample of Chinese-Canadian university students. Evidence of construct validity was obtained by investigating similarities and differences between Chinese-Canadian and European-Canadian samples in the internal structure of the measure and in the relationships between the H-MPS scales and other measures. Comparisons in mean levels of perfectionism were also made on a preliminary basis. A discussion of the results is organized using the following headings: internal evidence of construct validity; external evidence of construct validity; comparisons of mean levels of perfectionism; post hoc comparisons; findings related to socially-prescribed perfectionism. Limitations of the current study, suggestions for future research, and implications of the findings for counselling university students are also discussed. Internal Evidence of Construct Validity Internal Consistency Reliability Two sources of internal evidence of construct validity were examined and compared for the Chinese and European samples; internal consistency and subscale intercorrelations. Internal consistency reliability was assessed by calculating chronbach alpha coefficients for each of the H-MPS scales for the European and Chinese samples. The magnitudes and the rank ordering of the correlations were then compared for the two samples. Significant differences in the magnitude and rank ordering of the internal 76 consistency coefficients, between the two samples, were not identified. These findings demonstrate adequate internal consistency for the Chinese sample given that the magnitudes of the coefficients exceeded .80. This means that the items constituting each subscale are generally highly correlated with each other, indicating that the items are tapping into the same characteristic. The similarity of internal structure for the two samples provides preliminary evidence of construct validity for a Chinese-Canadian sample. In addition, the magnitudes of the internal consistency coefficients reported for the Chinese sample; .92 self-oriented perfectionism, 81 other-oriented perfectionism, 87 socially-prescribed perfectionism, are comparable to those reported in the initial validation study (Hewitt & Flett, 1991) using a Canadian university sample; .86 self-oriented perfectionism, .82 other-oriented perfectionism, .87 socially-prescribed perfectionism. These findings provide further evidence of construct validity for the H-MPS using a Chinese-Canadian sample. Given the results for the Chinese sample were interpreted by means of comparison with the European sample, it is also important to assess reliability and validity for the European sample. The internal consistency coefficients reported for the European sample; .91 self-oriented perfectionism, .82 other-oriented perfectionism, .87 socially-prescribed perfectionism, are also comparable to the respective coefficients reported by Hewitt and Flett (1991). Therefore construct validity is also supported for the European-Canadian sample in the current study. Subscale Intercorrelations Correlations among the H-MPS subscales were calculated for the Chinese and European samples to assess the degree of relationship among the three dimensions of 77 perfectionism. The magnitudes and rank ordering of the correlations were compared for the two samples. No significant differences in the magnitudes of the subscale intercorrelations were found between the Chinese and European samples, indicating a similarity of internal structure. However, differences in the rank order of the intercorrelations were found, with the correlation between the self-oriented and socially-prescribed perfectionism subscales larger in magnitude than that between the other-oriented and socially-prescribed perfectionism subscales for the Chinese sample. The reverse was true for the European sample. While Hewitt and Flett (1991) do not report specific coefficients for the different combinations of subscales, theoretically one would expect the self-oriented and socially-prescribed perfectionism subscales to be least correlated as was found for the European sample. Self-oriented perfectionism involves perfectionism directed towards the self and has been related to adaptive functioning in university students, for example personal control (Flett et al, 1991), and achievement striving and conscientiousness (Hill et al, 1998) whereas socially-prescribed perfectionism is an interpersonal or social dimension that has been linked to maladaptive functioning in university students, for example depressed mood (Flett et al, 1991, Frost et al, 1990)and low self-esteem(Arthur & Hayward, 1997;Flett et al, 1996). It was expected that self-oriented and other-oriented perfectionism would be more highly correlated than the latter subscales given they have been theoretically and empirically linked. Both subscales have been positively correlated with self-esteem and adaptive functioning (Flett et al, 1991) and both have been related to narcissism, dominance and authoritarianism (Hewitt & Flett, 1991). It was also expected that other-oriented and socially-prescribed perfectionism would be more highly 78 correlated than self-oriented and socially-prescribed perfectionism as they are both interpersonal dimensions of perfectionism. Thus, it was unexpected that the self-oriented and socially-prescribed perfectionism subscales were more closely related than the other-oriented and socially-prescribed subscales for the Chinese sample. One may speculate that for some university students from Chinese cultural backgrounds, perfectionism directed toward the self is a strong reflection of the introjection of perfectionistic expectations significant others, namely parents, have for them. An equally plausible explanation of the observed differences is that the items were interpreted somewhat differently by some of the Chinese-Canadian students and did not differentiate the perfectionism dimensions as well. A visual inspection of the magnitudes of the correlations for the two samples also supports this explanation. Invariably, the Chinese sample demonstrated larger coefficients which is indicative of a greater degree of overlap, meaning that the perfectionism dimensions may be somewhat less distinct for the Chinese sample. Consistent with this explanation, Hewitt and Flett (1991) found mixed support for the discriminant validity of the subscales. They found that self-criticism was positively related to all three dimensions and it may be that self-criticism or a construct like self-criticism is more salient for university students from Chinese cultural backgrounds and accounts for the higher level of shared variance among the subscales. Future studies corroborating these findings would be necessary in order to confirm and better understand this difference in the rank order of the intercorrelations among the H-MPS subscales between the European and Chinese samples in the current study. 79 Summary of the Internal Evidence of Construct Validity Overall, investigations of the internal structure of the H-MPS in Chinese and European samples provide preliminary evidence of construct validity for Chinese-Canadian university students given the similarity of internal structure as compared with a European sample, the high internal consistency coefficients (>80), and given the findings are comparable to those of the initial validation study (Hewitt & Flett, 1991). The observed differences in the rank order of the intercorrelations between the two samples may suggest that the measure is operating in a slightly different fashion for Chinese-Canadian university students, although sampling error could be an equally plausible explanation of the findings. Construct validity is also supported for the European sample in the current study. External Evidence of Construct Validity Relationship Between Perfectionism and Depression Construct validation is an ongoing process of collecting evidence about the meaning of a test, by demonstrating relationships between a test and other measures (Kaplan & Sacuzzo, 1997). In the current study, the relationships between the H-MPS and the Multidimensional Perfectionism Scale (F-MPS, Frost et. al., 1990) and the Beck Depression Inventory (BDI, Beck et. al., 1979) were examined and compared for the Chinese and European samples. Both the magnitudes and rank ordering of the correlations were compared for the two samples. The socially-prescribed perfectionism and total perfectionism scales demonstrated significant correlations with depression for both the Chinese and European samples. No differences in the rank order of the correlations were found. 80 Overall, the similar pattern of correlations for the two samples indicates that the subscales, or perfectionism dimensions, are related to depression in the same way, thus providing evidence of construct validity for a Chinese-Canadian sample. Given the subscales were differentially related to depression, these findings also constitute evidence of the dimensionality of the construct for a Chinese sample. The most important finding related to the relationship between perfectionism and depression was that the previously documented link between socially-prescribed perfectionism and depression was extended to a Chinese sample. This association was expected given the culture's emphasis on educational achievement and family pressures to excel (Cordova, 1983; Pang, 1991; Sue & Zane, 1984 and 1985), and given Chang's preliminary findings that Asian-American college students scored significantly higher than Caucasian college students on the dimensions of perfectionism that are associated with adjustment difficulties. However, the current findings also suggest a somewhat weaker relationship between socially-prescribed perfectionism and depression for a Chinese sample given the magnitude of the correlation was found to be significantly smaller for the Chinese sample, as compared with the European sample. The perception that others hold perfectionistic expectations and stringently evaluate one may be associated with fewer adjustment difficulties for some Chinese-Canadian university students. This may be related to cultural norms, meaning that some children in Chinese-Canadian families may perceive parents to have high expectations and to be critical of them, yet may not be related to the same level of helplessness and low self-esteem that it would be in a more individualistic culture. 81 Another possible explanation of the weaker association between socially-prescribed perfectionism and depression may be related to previous findings in the current study, that is, the greater association between self-oriented and socially-prescribed perfectionism for the Chinese sample. Self-oriented perfectionism has been related to having high personal standards (Hewitt & Flett, 1991; Hill et al, 1998), conscientiousness ( Hill et al, 1998), and to the perception of personal control (Flett et al, 1995). It may be that the association between socially-prescribed perfectionism and adjustment difficulties is attenuated for Chinese-Canadian students because socially-prescribed perfectionism is also related to some adaptive elements of perfectionism like personal control. One can only speculate as to why these differences in the degree of relationship between socially-prescribed perfectionism and depression emerged for the two samples. Contextualizing these findings in previous research findings sheds more light on these differences that were found. All of the previous findings in this area (Flett et al, 1995; Frost et al., 1990) demonstrated significant correlations between socially-prescribed perfectionism scores and BDI scores, ranging from .31 - .44. The correlations reported in the current study, for both groups, are very close to this range. It appears that the European sample in the current study had a correlation that was somewhat larger in magnitude and the Chinese sample had a correlation that was slightly smaller in magnitude compared to the range of magnitudes reported above. In fact, the magnitude of the correlation for the Chinese sample is closer to the range reported above. It may be that sampling error can account for the significant differences found in the association between socially-prescribed perfectionism and depression for the Chinese 82 and European samples in the current study, given the non-random sampling methods, low numbers, gender imbalance and positively skewed distribution for depression for both samples. It may be that the samples in the current study are not representative and the findings must be interpreted with caution. In addition, the current study did not control for third variables such as self-esteem, which has been demonstrated to be a buffer of maladaptive perfectionism (Preusser, et. al, 1994). It may be that there were differences between the samples in level of self-esteem or another third variable that could account for these differences. The difference in the magnitude of the correlations between total perfectionism and depression between the Chinese and European samples that was identified in the current study is difficult to interpret given other researchers did not assess this relationship. Hewitt and Flett (1991, 1994, 1995) consistently view perfectionism as a true multidimensional construct and do not include analyses using the total scale of the H-MPS. However, given the socially-prescribed perfectionism subscale comprises one third of the total scale, one might speculate that the relationship between socially-prescribed perfectionism and depression accounts for this relationship. It is speculative as to why a weaker relationship was observed for the Chinese group, and the findings would have to be replicated in future studies before meaning is attached to them. The findings may reflect real cultural differences in the perfectionism-depression link and/or in the construct itself, or the findings may be an artifact of sampling error. Overall, given the similarity in the pattern of the correlations between the perfectionism dimensions and depression for the Chinese and European samples and 83 given the findings are comparable to previous findings, the current study provides solid external evidence of construct validity of the H-MPS for a Chinese-Canadian sample. Relationship Between Two Measures of Perfectionism Another source of external validity was investigated in the current study, that is, the relationship between the H-MPS and another test purporting to measure the same construct, the F-MPS. The magnitudes and rank ordering of the correlations between the two perfectionism measures were examined and compared for the European and Chinese samples. All correlations were found to be positive, moderate and significant at p< .01. No significant differences between the Chinese and European samples in the magnitudes and patterns of the correlations were found, indicating a similar relationship between the two measures for the two samples. These findings provide strong evidence of construct validity for a Chinese-Canadian sample. The findings with respect to the patterns of correlation are consistent with the previous findings of Frost and colleagues (1990) who also reported the following rank order (descending): socially-prescribed perfectionism; self-oriented perfectionism; and other-oriented perfectionism. The findings with respect to the magnitudes of the correlations are somewhat different, with the current study reporting larger correlations for both the Chinese and European samples. Frost (1990) reported the following correlations: .49 for self-oriented perfectionism; .28 for other-oriented perfectionism; and .57 for socially-prescribed perfectionism whereas the current study reported the following correlations for the Chinese and European samples, respectively: .58 and .67 self-oriented perfectionism; .40 and .46 other-oriented perfectionism; and .69 and .77 socially-prescribed perfectionism. 84 It is difficult and speculative to say why differences in the magnitudes of the correlation between the samples in the current study and the samples in Frost's study were found; however, differences related to the composition of the samples may be implicated. The Frost sample consisted of 553 introductory psychology students at a large Eastern U.S. college that was balanced for gender and the students received course credit for their participation. In contrast, the samples in the current study were Canadian university students on the West Coast and were much smaller in size, more diverse in terms of their major and year of study, and consisted of more women than men. Whatever the explanation of the differences between the current sample and that of Frost and colleagues (1990), the important point is that the relationship between the H-MPS scales and the F-MPS was similar for the Chinese and European samples, providing strong evidence of construct validity for the Chinese sample. Frost and colleagues (1990) did not report correlations between the H-MPS total scale and the F-MPS, making it difficult to compare the current findings. Another means of checking the validity of the H-MPS scale for a Chinese sample is to compare the correlations between the three H-MPS subcales and the total scale (H-MPS) with those between the three H-MPS subscales and the total scale (F-MPS). One would expect the H-MPS subscales to be more highly correlated with their own scale than with an external perfectionism measure (F-MPS), for validity to be supported. The above correlations were calculated and compared for the European and Chinese samples. It was found that all of the subscales for both samples were more highly correlated with the total score on its own measure (H-MPS) than with the total score on the F-MPS, with the exception of the socially-prescribed perfectionism subscale for the Chinese sample. 85 For the Chinese sample, the socially-prescribed perfectionism subscale demonstrated a correlation of .68 with the total scale on its own measure and a correlation of .69 with the total scale on the Frost measure. In addition, statistically significant differences in the correlation between the socially-prescribed perfectionism subscale and the total scale (PI-MPS) were demonstrated between the Chinese and European samples, with the Chinese group showing a weaker correlation. This indicates that the socially-prescribed perfectionism subscale is less related to the overall measure, the H-MPS, for the Chinese sample and may suggest that this subscale is acting somewhat differently for the Chinese sample. Summary of External Evidence of Construct Validity The following findings have provided solid evidence of construct validity for the H-MPS using a Chinese-Canadian sample: a similarity in the pattern of correlations between the perfectionism dimensions and depression for the European and Chinese samples; a confirmation of the previously documented link between socially-prescribed perfectionism and depression for a Chinese sample; and no significant differences in the magnitudes and rank ordering of the correlations between the H-MPS and the F-MPS for the Chinese and European samples. However, a few noteworthy differences between the two samples, with repect to external evidence, were identified. While the link between socially-prescribed perfectionism and depression was confirmed for the Chinese sample, the relationship was significantly weaker. In addition, the socially-prescribed perfectionism subscale was significantly less related to the overall measure for the Chinese sample. 86 This initial investigation of construct validity of the H-MPS for Chinese-Canadian university students provided solid internal and external evidence of validity.The expected relationships were also confirmed for the European sample, thus providing evidence of sonstruct validity for a European-Canadian sample. The question of whether differences between the two samples, with respect to the socially-prescribed perfectionism subscale, can be attributed to real cultural differences or are simply due to sampling error remains, and the findings must be replicated by future investigations using Chinese-Canadian samples. However, the researcher determined that sufficient evidence of construct validity was provided, to justify a preliminary investigation of mean levels of perfectionism for the Chinese sample. Comparisons of Mean Levels of Perfectionsim Mean levels of perfectionism were examined and compared for the Chinese and European samples in the current study. Comparisons were made using both the total and subscale means. Significant differences in mean levels of perfectionism were not found between the Chinese and European samples in the current study, although European women scored significantly higher on the socially-prescribed perfectionism subscale than the European men. Overall, these findings can be interpreted as providing further evidence of construct validity of the H-MPS for Chinese-Canadian university students. However, gender differences in mean levels of socially-prescribed perfectionism have not been reported elsewhere and may be the result of sampling error. Therefore, the representativeness of the European sample is called into question, suggesting that the sample may not be the best "control group" for interpreting the findings for the Chinese sample. 87 Given that numerous studies using Canadian university students have reported mean levels of perfectionism, the means reported in the current study were compared to those reported in other investigations (Flett et al. 1995; Hewitt & Flett, 1991). Appendix G lists these means. All of the means reported for both the Chinese and European samples in the current study fall within the range of means reported in the other investigations, with the exception of the self-oriented perfectionism mean reported for Chinese males which was slightly elevated. Summary of Findings Related to Comparisons of Mean Perfectionism Levels Given no differences in mean levels of perfectionism between the European and Chinese samples were identified in the current study, and given the means reported for the Chinese sample fall within the range reported elsewhere, construct validity of the H-MPS for a Chinese sample is further bolstered. The current findings do not support the establishment of separate norms for Chinese-Canadian university students, although future investigations confirming these findings would be warranted. The findings suggest that the personal and social aspects of perfectionism generalize across cultures, thus bolstering the construct validity of perfectionism. Future studies investigating perfectionism in other cultural groups are necessary. Post Hoc Comparisons Relationship Between Socially-prescribed Perfectionism and Depression Given European women scored significantly higher on the socially-prescribed perfectionism subscale than European men, the relationship between socially-prescribed perfectionism and depression was re-examined by taking gender into account. It was thought that gender differences in this relationship could account for the differences that 88 were found between the two cultural groups, with the Chinese sample demonstrating a weaker relationship. Comparisons of the correlation between socially-prescribed perfectionism and depression were made between Chinese females, Chinese males, European females and European males. Interestingly, when gender was considered, significant differences in the relationship between socially-prescribed perfectionism and depression were not identified. However, the correlation for Chinese women was not found to be significant in contrast to the other groups and upon visual inspection, the magnitude of the correlation was smaller for Chinese women. The magnitude of the correlation for Chinese and European men was almost identical. Thus, the weaker relationship between socially-prescribed perfectionism and depression that was found for the Chinese sample may be explained by the much lower correlation that was found for Chinese women relative to European women. Perhaps significant differences between Chinese and European women would have been detected had the samples been somewhat larger. One might speculate that differences in the socialization and expectations of women in the two cultures account for these differences, and may serve to protect some Chinese-Canadian women from the maladaptive elements of perfectionism. Future investigations, using larger samples, would be required before any firm conclusions can be drawn. Comparison of Mean Levels of Perfectionism based on Generation Status Given differences in mean levels of perfectionism between Chinese and European samples were not identified, the researcher was curious whether potential differences may have been masked by the heterogeneity of the two samples. Therefore, some post hoc analyses were performed to investigate whether differences would be identified when 89 generation status and level of acculturation were considered for the Chinese sample. Acculturation level and generation status, are very important dimensions to consider when investigating cultural differences. There are implications for the current study, for example, if the sample of Chinese-Canadian students in the current study is highly acculturated and most individuals hold a Western identification (Suinn, 1987), differences between the European and Chinese samples would not be expected. Mean levels of perfectionism, using both subcale and total scores, were compared for first and second-generation Chinese samples. Differences between these two groups were found only for the socially-prescribed perfectionism subscale, with the second-generation Chinese sample scoring significantly higher than the first-generation sample. This may indicate that second-generation Chinese university students feel more pressure from parents and significant others to excel and strive for perfection. Given socially-prescribed perfectionism is the element of perfectionism most closely linked to adjustment difficulties, these findings could have important implications for the mental health of second-generation students. In order to shed light on these findings, comparisons in mean levels of socially-prescribed perfectionism were made between the European sample and the two generation groups. It was found that second-generation Chinese females scored significantly higher than European males and that second-generation Chinese females scored slightly higher on socially-prescribed perfectionism than samples from previous studies. This same pattern was also found for European women, indicating that cultural differences in socially-prescribed perfectionism may be real and may only be detected when generation status and gender are considered. While second-generation Chinese 90 women and European women tend to score higher on the socially-prescribed perfectionism subscale, the link between socially-prescribed perfectionism and depression may be somewhat weaker for the Chinese women. It would be important for future investigations to unpack the influence of gender, generation status and culture on socially-prescribed perfectionism scores and to investigate the potential impact on the relationship between socially-prescribed perfectionism and depression. Possible Explanations of the Differences found between First and Second Generation Chinese Samples Gender differences within each group were explored as a possible explanation for these findings. A similar ratio of women to men within each generation group was found, and no significant differences in mean levels of socially-prescribed perfectionism were found between females and males in either generation group. Therefore, gender differences cannot account for the differences in mean levels of socially-prescribed perfectionism that were found between first and second-generation Chinese samples. It was thought that demographic characteristics may account for these differences. The second-generation sample was significantly younger than the first-generation sample, having mean ages of 19.02 and 21.33 years, respectively. One may speculate that, for younger students, parental expectations and standards may be more important given, they may be less autonomous, developmentally, and more dependent on parents. These findings are consistent with the findings of Saddler and Sacks (1993), who reported that students at lower grade classifications tended to score higher on all three perfectionism dimensions. Thus, age or developmental differences may potentially explain differences in socially-prescribed perfectionism that were found between first and second-generation 91 Chinese samples. Alternatively, the findings may be an artifact of sampling error, or the differences may reflect real differences in socially-prescribed perfectionism between the generations. Differences in level of acculturation may be another possible explanation of the differences in socially-prescribed perfectionism between the two generations, given the first generation sample scored significantly lower than the second generation in level of acculturation. If the differences in socially-prescribed perfectionism between the generations can be attributed to acculturation level, one would expect high and low acculturation groups from the Chinese sample to differ in mean levels of socially-prescribed perfectionism. Overall, an explanation of the differences in socially-prescribed perfectionism between the two generations is confounded by differences in age and acculturation that were identified, rendering these findings very difficult to interpret. Comparison of Mean Levels of Perfectionism based on Acculturation Level The Chinese sample was split into high and low acculturation groups, based on scores on the SL-ASIA. When high and low acculturation groups were compared across all of the perfectionism dimensions, statistically significant differences were not found. In addition, there did not appear to be a relationship between perfectionism and acculturation level given the correlations between acculturation level and the perfectionism scales were neglible. Finding differences based on generation status as opposed to acculturation level was unexpected. Generation status is a demographic variable that was measured objectively, whereas level of acculturation, as measured by the SL-ASIA scale in the 92 current study, included items relating to an individual's behaviours and self-identity. It was expected that acculturation level would differentiate individuals in the Chinese group better than generation status and potentially account for any cultural differences that were found in the current study. However, the latter expectation presupposed that the SL-ASIA scale measured what it purported to, and the measure has not undergone validation procedures using Chinese-Canadian samples. Although the SL-ASIA scale demonstrated a strong internal consistency coefficient of 0.88, indicating that the items on the scale are tapping into the same characteristic, the scale may be assessing superficial behaviours rather than internal identity. Kwan and Sodowsky (1997) make the distinction between external and internal ethnic identity. External identity refers to "observable social and cultural behaviours.. .that manifest in the areas of ethnic language usage, ethnic group friendship, participation in ethnic-group functions and activities, ethnic media preference, and maintaining ethnic traditions" (pg. 53). In contrast, internal ethnic identity refers to cognitive, moral and affective dimensions relating to an individual's perceptions and feelings about, and commitment to, one's ethnic group. Kwan & Sodowsky (1997) found that internal ethnic identity as opposed to external ethnic identity predicted the salience of ethnicity in Chinese-American immigrants. Upon visually inspecting the original 21 items of the SL-ASIA scale, a majority of the items appear to measure external identity, or behaviours, while only a few items seem to be tapping into internal identity. Therefore, the SL-ASIA scale may not have adequately differentiated between individuals in the current study who possess the behavioural competencies to fit into Canadian culture, but retain an internal Chinese 93 identity and those that possess the behavioural competencies to fit in and are Western identified. Future studies, employing the use of a valid measure of acculturation or internal identity, are necessary to determine if acculturation plays a role in development of perfectionism. Summary of Findings Related to Generation Status and Acculturation Level Differences in mean levels of socially-prescribed perfectionism between first and second-generation Chinese university students were identified in the current study. Second-generation students were found to score higher on this dimension of perfectionism. These findings are difficult to interpret and are confounded by differences related to age and level of acculturation that cannot be disentangled. The findings may reflect real differences in socially-prescribed perfectionism between the two generations, or may be attributed to sampling error. There is a need for future studies to clarify the influence of generation status on socially-prescribed perfectionism scores for Chinese-Canadians. In addition, cultural differences in socially-prescribed perfectionism were identified, with second-generation Chinese females scoring higher than European men. Given that European women also scored higher than European men and scored similarly to second-generation Chinese women on socially-prescribed perfectionism, these findings point to potential differences in socially-prescribed perfectionism when generation status and gender are considered. Differences in mean levels of perfectionism were not identified between low and high acculturation groups in the Chinese sample. 94 Findings Related to Socially-prescribed Perfectionism One theme that has emerged in examining all of the findings of the current study is that the socially-prescribed perfectionism subscale was implicated in all of the differences that were identified between groups, both within and between cultural groups. While sampling error may account for these differences, one possible explanation of these findings is that the trait, socially-prescribed perfectionism, is not as stable as the other dimensions. In reviewing the test-retest data that was reported for a three month period by Hewitt and Flett (1991), the socially-prescribed subscale did, in fact, demonstrate the weakest reliability coefficient at 0.75, whereas the self-oriented and other-oriented perfectionism subscales had coefficients of 0.88 and 0.85, respectively. Although a coefficient of 0.75 is adequate when considering temporal or trait stability, it is not ideal and may affect the reliability and hence the validity of the socially-prescribed perfectionism subscale and may account for some of the findings of the current study. Test-retest data is not currently available for a Chinese-Canadian sample, rendering it difficult to know if the trait is more or less stable for this group. It would be important for future studies to further investigate the stability of the socially-prescribed perfectionism subscale. Another possible explanation for the findings related to the socially-prescribed perfectionism subscale is that gender differences are significant for this dimension. Although gender differences have not been reported elsewhere, the current study found that European females and second generation Chinese females scored significantly higher than European males. In addition, upon visual inspection, females in the current study invariably scored higher than their male counterparts (see Appendix G). It would be important for future investigations into the construct of perfectionism to test for gender 95 differences on the socially-prescribed perfectionism subscale in order to clarify the role of gender. Conclusions The major aim of the current study was to determine if the H-MPS showed internal and external evidence of construct validity in a sample of university students from Chinese cultural backgrounds, as assessed by making comparisons with a European sample of university students. A similarity of internal structure and comparable relationships between the H-MPS and other measures for the two samples, would suggest construct validity for the Chinese sample. A secondary objective, contingent upon finding solid evidence of validity, was to provide an initial investigation of mean levels of perfectionism in a Chinese-Canadian sample. The H-MPS was found to be internally consistent for both the Chinese and European samples, given the high alpha coefficients reported. A similarity of internal structure between the two samples was demonstrated given that significant differences in the magnitudes of the internal consistency coefficients and of the subscale intercorrelations were not found, lending support for the dimensionality of the construct in a Chinese-Canadian sample. With respect to external evidence, the relationship between the H-MPS and the F-MPS was similar for the two samples, providing evidence of construct validity. However, differences between the Chinese and European samples were found in the relationship between the H-MPS and the BDI for the socially-prescribed perfectionism subscale, with a weaker relationship demonstrated for the Chinese sample. This may indicate a cultural difference in the relationship between socially-prescribed perfectionism and adjustment difficulties. However, when this relationship was further examined by splitting the two groups by gender, no significant differences emerged. Interestingly, the correlation for the Chinese women was not found to be significant, which is contrary to previous findings. These findings, while preliminary, suggest that socially-prescribed perfectionism may not be associated with the same degree of adjustment difficulties or perceived helplessness for Chinese-Canadian women. Overall, the researcher concluded that sufficient evidence of construct validity was obtained to justify an investigation of mean levels of perfectionism in the Chinese sample. Significant differences in mean levels of socially-prescribed perfectionism were found between European males and European females and between European males and second-generation Chinese females, with the female groups scoring higher than the European males. However, all of the means reported in the current study for the European and Chinese samples fall within, or are very close to, the range of means reported elsewhere which provides further support for construct validity for a Chinese sample. Therefore, the findings of the current study do not support the establishment of separate norms for perfectionism for Chinese-Canadian university students. Differences in mean levels of perfectionism were not found between high and low acculturation groups from the Chinese sample. The construct validity of the H-MPS for both Chinese-Canadian and European-Canadian university students was supported in the current study. However, there were some anomalies associated with the socially-prescribed perfectionism subscale, which was implicated in all of the differences that were found. This calls into question the stability of the socially-prescribed perfectionism subscale and suggests that individuals 97 may score differently on this subscale, depending on their gender. While it is possible that the gender differences in socially-prescribed perfectionism that were reported in the current study reflect real differences, it is equally possible that sampling error can account for these findings. The findings of the current study would have to be replicated by future investigations in order to draw any solid conclusions. Delimitations/Limitations Limitations of the current study are primarily related to the sample used and the research design employed. The findings of the current study cannot be generalized to Chinese-Canadian and European-Canadian university students at U.B.C. because samples of convenience rather than non-random samples were used. In comparing the European and Chinese samples with the U.B.C. population data, females, students in lower grade classifications, and students in the Faculty of Arts were over-represented in the current study. More important, the representativeness of the two samples used in the current study is impossible to assess given that U.B.C. does not keep statistics on the ethnic origins of the student population, further limiting any conclusions about generalizability. In addition, the categories used to classify participants as European or Chinese were simplistic, and the complexity of cultural background was not fully considered, in order to establish two clean categories. Those individuals who did not fit into one of these categories were excluded from participating. Thus, the current sample did not represent the true diversity of cultural backgrounds that comprise the U.B.C. population. Another limitation of the study relates to the correlational design. Cause was not assessed and third variables such as self-esteem or socieconomic status could account for the findings. Woolfolk (1995) stated that: "social class is a powerful dimension of 98 cultural differences, often overpowering other differences such as ethnicity or culture" (p. 160). However, given perfectionism was studied in university students as opposed to a community sample, one may speculate that the European and Chinese samples in the current study are from similar socioeconomic backgrounds, thus holding socioeconomic status constant. Another limitation related to the design of the study is that only self-report techniques were used. It would be important to determine if individual differences in perfectionism reflect real differences rather than self-report biases; including other techniques such as observer ratings would overcome this limitation. Finally, the statistical tools used determine the quality and reliability of the results obtained. In investigating the internal structure of a measure, factor analysis is a more rigorous statistical procedure than correlation, the procedure that was used in the current study. A confirmatory factor analysis would be helpful in confirming the presence of three factors, corresponding to the three perfectionism dimensions, and would provide stronger internal evidence of construct validity for a Chinese-Canadian sample. Suggestions for Future Research The current study investigated the construct validity of the H-MPS for a sample of Chinese-Canadian university students. Given this study is the only known investigation of the construct of perfectionism in Chinese-Canadians, it would be important for future investigations of the perfectionism construct to include samples of Chinese-Canadians, in order to confirm or disconfirm the findings of the current study. While the results clearly support the construct validity of the H-MPS for a Chinese sample, there were some interesting findings that could be explored further in future investigations. 99 While the previously documented link between socially-prescribed perfectionism and depression was confirmed for the Chinese sample, the relationship was found to be significantly weaker, particularly for Chinese women, as compared to the European sample. Chinese women appeared to have some protection from the vulnerability to depression that is normally associated with high scores on socially-prescribed perfectionism. It would be important to further clarify this relationship and to identify the characteristics or factors that may serve as a buffer of maladaptive perfectionism in this group. It was striking that all of the differences identified in the current study, both within and between cultural groups, were related to the socially-prescribed perfectionism subscale. One suggestion for future studies would be to include test-retest data on the socially-prescribed perfectionism subscale, given these findings call into question the stability of the trait. With respect to cultural differences in socially-prescribed perfectionism, the subscale was found to be more highly related to self-oriented perfectionism in the Chinese sample as compared to the European sample. It may be that the link between socially-prescribed perfectionism and depression is weaker for a Chinese sample because of the strong link to self-oriented perfectionism, which has been associated with adaptive qualities. This relationship needs to be further explored and delineated. In addition, socially-prescribed perfectionism was less related to overall perfectionism in the Chinese sample. This may indicate that the construct is somewhat different for Chinese-Canadians. Examining the structure of perfectionism in a Chinese-Canadian sample, by using confirmatory factor analysis, or obtaining qualitative feedback from a Chinese 100 sample on the socially-prescribed subscale items may be helpful to better understand how this subscale is operating. Another potential area to examine would be gender differences in mean levels of socially-prescribed perfectionism. European women scored significantly higher than European men, and all groups of women in the current study, irrespective of culture, scored higher than their male counterparts. While gender differences in mean levels of socially-prescribed perfectionism have not been identified elsewhere, gender difference may be real, particularly in some cultures such as the European cultures. These differences may not have been identified in previous studies because perfectionism has not been studied from a cultural perspective. The potential impact of acculturation level and generation status on perfectionism scores needs to be further explored in a Chinese sample. While the current study found differences related to generation status and not to acculturation level, the reverse was expected. The measure of acculturation that was used in the current study, the SL-ASIA, has not been validated on a Chinese sample here in Canada and has only been used with Asian-Americans. It would be important for future investigations into the construct of perfectionism in Chinese-Canadians to include a valid measure of acculturation, as potential differences may not have been identified due to the limitations of the measure. In addition, the differences that were found between the two generations in the Chinese sample were difficult to interpret given the results were confounded by differences in age and acculturation level. Therefore, it is important to explore and attempt to disentangle the influence of generation status, age, gender, and level of acculturation on 101 perfectionism scores and on the manifestation of adjustment difficulties related to perfectionism. Finally, it would be interesting to investigate the construct validity of the H-MPS using other cultural groups, to determine if the personal and social elements of perfectionism generalize to other cultural groups comprising our multicultural population in British Columbia. In order to better meet the health needs of a diverse population, it is important to test the cross-cultural applicability of constructs, like perfectionism, that have important psychological implications. Implications of the Findings for Counselling University Students There are several theoretical/conceptual implications of the findings of the current study, which suggest some implications for counselling university students from Chinese cultural backgrounds. Given the solid psychometric data reported for the H-MPS using Chinese-Canadian and European-Canadian university students, the current study lends support for using the measure with both populations. The H-MPS could be used in both individual and group therapy to screen for the level, or severity, of perfectionism experienced by clients from these groups. The measure could also be used to assist the client and counsellor in gaining insight into the client's idiosyncratic pattern of perfectionism. For example, by noting the subscale scores that are highest, the client and counsellor can identify if issues related to perfectionism are primarily personal or interpersonal in nature. In addition, adaptive elements of perfectionism and client strengths could be highlighted and explored where scores are relatively lower. Overall, the H-MPS could be 102 useful in gaining insight into perfectionistic patterns, and scores could be used to target therapy in both European and Chinese university students. The current study confirmed the previously documented relationship between socially-prescribed perfectionism and depression for both European-Canadian and Chinese-Canadian university students, although a weaker link was demonstrated for the Chinese group. This suggests that it would be prudent for counsellors to screen for depression when working with perfectionistic clients from Chinese and European backgrounds, particularly for those clients who exhibit high scores on socially-prescribed perfectionism. Blatt (1995) found that perfectionists, as a subclass of depressives, responded poorly to brief therapy. If future studies extend Blatt's findings to a Chinese-Canadian sample, longer-term therapy may be more efficacious in treating perfectionism than the brief model of therapy that is typically adopted by University Counselling Centres. Future investigations of the clinical implications of perfectionism, particularly for Chinese-Canadian university students, are necessary in order to develop applicable and useful treatment plans. 103 References Adler, A. (1964). The individual psychology of Alfred Adler. In H.L. Ansbacher & RR. Ansbacher (Eds.). The Individual Psychology of Alfred Adler. New York: Harper & Row (Torch Books). Adler, A. (1926; reprinted 1972). The neurotic constitution. Freeport, NY: Books for Libraries Press. Arthur, N. & Hayward, L. (1997). The relationship between perfectionism, standards for academic achievement, and emotional distress in postsecondary students. Journal of College Student Development. 38(6). 622-630. Atkinson, D.R. & Gim, RM. (1989). Asian-American cultural identity and attitudes toward mental health services., Journal of Counseling Psychology 36(2). 209-212. Barrow, J.C. & Moore, C. A. (1983). Group interventions with perfectionistic thinking. Personnel and Guidance Journal. 61. 612-615. Beck, AT., Rush, A.J., Shaw, B.F. & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press. Beck, AT., Steer, R.A. & Garbin, MG. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review. 8. 77-100. Beck, AT., Ward, CM., Mendelson, M., Mock, J. & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry. 4. 561-571. Blatt, S.J. (1995). The destructiveness of perfectionism: Implications for the treatment of depression. American Psychologist. 50(12). 1003-1020. Blatt, S.J. & Zuroff, D.C. (1972). Interpersonal relatedness and self-definition: Two prototypes for depression. Clinical Psychology Review, 12, 527-562. Burns, D.D. (1980). The perfectionist's script for self-defeat. Psychology Today, Nov., 34-57. Chan, CM. & Tsoi, MM. (1984). The BDI and stimulus determinants of cognitive and related depression among Chinese college students. Cognitive Therapy and Research. 8. 501-508. Chan, D.W. (1991). The Beck Depression Inventory. What difference does the Chinese version make? Psychological Assessment: A Journal of Consulting and Clinical Psychology. 3(4). 616-622. Chang, E. (1998). Cultural differences, perfectionism, and suicidal risk in a college population: Does social problem-solving still matter? Cognitive Therapy and Research. 22(3). 237-254. Clavin, S.L., Clavin, R.H., Gayton, W.F. & Broida, J. (1996). Continued validation of the multidimensional perfectionism scale. Psychological Reports, 78.732-734. Cordova, F. (1983). Filipinos: Forgotten Asian-Americans. Dubuque. Kendall\Hunt. Cuellar, I., Harris, L. & Jasso, R. (1980). An acculturation scale for Mexican American normal and clinical populations. Hispanic Journal of Behavioral Science. 2,199-17. Delisle, J.R. (1990). The gifted adolescent at risk: Strategies and resources for suicide prevention among gifted youth. Journal of Education of the Gifted. 13,212-228. Demaria, TP., Kassinove, H. & Dill, C. A. (1989). Psychometric properties of the survey of personal beliefs: A rational-emotive measure of irrational thinking. Journal of 105 Personality Assessments. 53. 329-341. Driscoll, R. (1989). Self-condemnation: A comprehensive framework for assessment and treatment. Psychotherapy. 26. 104-111. Farrel, D.M. (1989). Suicide among gifted students. Roeper Review. 11.134-139. Flett, G.L., Hewitt, PL., Blankstein, K.R & Mosher, SW. (1991). Perfectionism, self-actualization and personal adjustment. Journal of Social Behaviour and Personality. 6, 147-160. Flett, G.L., Hewitt, PL., Blankstein, K.R. & Mosher, S.W. (1995). Perfectionism, life events and depressive symptoms: A test of a diathesis-stress model. Current Psychology. 14. 112-137. Flett, G.L, Hewitt, PL., Blankstein, K.R. & O'Brien, s. (1991). Perfectionism and learned resourcefulness in depression and self-esteem. Personality and Individual Differences. 12.61-68. Flett, G.L., Hewitt, PL. & DeRosa, T. (1996). Dimensions of perfectionism, psychosocial adjustment and social skills. Personality and Individual Differences. 20(2), 143-50. Flett, G.L., Hewitt, P.L. & Dyck, D.G. (1989). Self-oriented perfectionism, neuroticism, and anxiety. Personality and Individual Differences. 10. 731-735. Flett, G.L., Russo, F A & Hewitt, PL. (1994). Dimensions of perfectionism and constructive thinking as a coping response. Journal of Rational-Emotive and Cognitive Behaviour Therapy. 12.163-179. Frost, R.O., Heimberg, R G , Holt, C.S., Mattia, J.I. & Neubauer, AL. (1993). A comparison of two measures of perfectionism. Personality and Individual Differences. 14.119-126. Frost, R.O., Lahart, CM., Rosenblate, R. (1991). The development of perfectionism: A study of daughters and their mothers. Cognitive Therapy and Research. 15.469-89. Frost, R.O., Marten, P., Lahart, C. & rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research. 14.449-468. Garner D.M., Olmstead, M.P. & Polivy, J. (1983). Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders. 2. 15-34. Gotlib, I.M. (1984). Depression and general psychopathology in university students. Journal of Abnormal Psychology. 93. 19-30. Halgrin, R. & Leahy, P. (1989). Understanding and treating perfectionistic college students. Journal of Counseling and Development. 68. 222-225. Hamachek, D.E. (1978). Psychodymanics of normal and neurotic perfectionism. Psychology. 15. 27-3 3. Hays, R.L. & Sloat, R.S. (1989). Gifted students at risk for suicide. Roeper Review. 11. 134-139. Hewitt, PL. & Dyck, D.G. (1986). Perfectionism, stress, and vulnerability to depression. Cognitive Therapy and Research. 10. 137-142. Hewitt, P.L. & Flett, G.L. (1991). Perfectionism in the self and social contexts: Conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology. 60. 456-470. Hewitt, PL., Flett, G.L. & Blankstein, K.R. (1991). Perfectionism and neuroticism in psychiatric patients and college students. Personality and Individual Differences. 12, 273-279. Hewitt, PL., Flett, G.L., Turnbull-Donovan, W. & Mikail, S. (1991). The multidimensional perfectionism scale: Reliability, validity, and psychometric properties in psychiatric samples. Psychological Assessment: A Journal of Consulting and Clinical Psychology. 3. 464-468. Hewitt, PL., Flett, G.L. & Weber, C. (1994). Dimensions of perfectionism and suicide ideation. Cognitive Therapy and Research. 18.439-460. Higgins, E.T., Bond, R., Klein, R. & Strauman, T.J. (1986). Self-discrepancies and emotional vulnerability: How magnitude, accessibility and type of discrepancy influence affect. Journal of Personality and Social Psychology. 41. 1-15. Hill, R.W., Mclntire, K. & Bacharach, V.R. (1997). Perfectionism and the Big Five Factors. Journal of Social Behaviour and Personality. 12(1). 257-270. Hoge, D.R. & McCarthy, J.D. (1983). Issues of validity and reliability in the use of real ideal discrepancy scores to measure self-regard. Journal of Personality and Social Psychology. 44. 1048-1055. Hollender, MH. (1965). Perfectionism. Comprehensive Psychiatry. 6. 94-103. Hooley, J.M. & Teasdale, J.D. (1989). Predictors of relapse in unipolar depressives: Expressed emotion, marital distress and perceived criticism. Journal of Abnormal Psychology. 98. 229-235. Horney, K. (1945). Our Inner Conflicts. New York: WW. Norton. Jones, R.G. (1968). A factorial measure of Ellis' irrational belief system, with personality and maladjustment correlates. Dissertation Abstracts International. 29: 4379B -4380B. Kaplan, R.M. & Saccuzzo, DP. (1997). Psychological testing, principles, applications and issues (4th Ed.). Pacific Grove, CA: Brooks/Cole Publishing Company. Kassinove, H. (1986). Self-reported affect and core irrational thinking: A preliminary analysis. Journal of Rational Emotive Therapy. 4. 119-130. Kitano, H.H. (1969). Japanese-American mental illness. In S.C. Plog, and R.B. Edgerton, (Eds.), Changing Perspectives in Mental Illness. 256-284. New York: Holt, Rineholt & Winston. Kitano, M. (1986). Gifted and talented Asian children. In MKitano & P. Chinn (Eds.), Exceptional Asian Children and Youth. 54-60. Reston, VA: Council for Exceptional Children. Kuo, W. (1984). Prevalence of depression among Asian-Americans. Journal of Nervous and Mental Diseases. 172.449-457. Kwan, K. & Sodowsky, G.R. (1997). Internal and external ethnic identity and their correlates: A study of Chinese-American immigrants. Journal of Multicultural Counseling and Development. 25. 51-67. Lazzari, R., Fioravanti, M., & Gough, H.G. (1978). A new scale for the adjective checklist based on self versus ideal self discrepancies. Journal of Clinical Psychology. 34. 361-365. Missildine, W.H. (1963). Perfectionism - If you must strive to "do better". In W.H. Missildine (Ed.), Your inner child of the past. 75-90. New York: Pocket Books. Mordkowitz, E. & Ginsburg, H. (1986). The academic socialization of successful Asian American college students. Paper presented at the meeting of the American Educational Research Association, San Francisco. Morris, RP. (1961). Effect of the mother on goal-setting behaviour of the asthmatic child. Annals of Allergy. 19, 44-54. Ownbey, S.F. & Horridge, P.E. (1998). The suinn-lew asian self-identity acculturation scale: Concurrent and factorial validation. Educational and Psychological Measurement. 52. 1041-1046. Pacht, AR. (1984). Reflections on perfectionism. American Psychologist. 39. 386-390. Pang V.O (1991). The relationship of test anxiety and math achievement to parental values in Asian-American and European-American middle school students. Journal of Research and Development in Education. 24(4). 1-10. Pirot, M. (1986). The pathological thought and dynamics of the perfectionist. Individual Psychology: Journal of Adlerian Theory, Research and Practice. 42. 51-58. Preusser, K.J., Rice, KG. & Ashby, J.S. (1994). The role of self-esteem in mediating the perfectionism-depression connection. Journal of College Student Development. 35. 88-93. Rachman, S.J. & Hodgson, RJ. (1980). Obsessions and compulsions. Englewood Cliffs, NJ.: Prentice-Hill. Rice, KG., Slaney, R.B. & Ashby, J.S. (1998). Self-esteem as a mediator between perfectionism and depression: A structural equations analysis. Journal of Counseling Psychology. 45(3). 304-314. Rothblum, E.D., Solomon, LJ. & Murakami, J. (1986). Affective, cognitive and behavioral differences between high and low procrastinators. Journal of Counseling Psychology. 33. 387-394. Saddler, CD. & Sacks, L.A. (1993). Multidimensional perfectionism and academic procrastination: Relationship with depression in university students. Psychological Reports. 73. 863-871. Santos, L. (1983). The social and emotional development of Filipino-American children. In G. Powell (Ed.), The psychosocial development of minority group children. 131-158. New York: BrunnerVMazel. Sata, L. (1983). Mental health issues of Japanese-American children. In G. Powell (Ed.). The psychological development of minority children. 362-372. New York: BrunnerVMazel. Schwartz, A. (1971). The culturally advantaged: A study of the Japanese-American pupils. Sociology and Social Research. 55. 341-353. Shek, D.T.L. (1990). Reliability and factorial structure of the Chinese version of the Beck Depression Inventory. Journal of Clinical Psychology. 46(1). 35-42. Shek, D.T.L. (1991). What does the Chinese version of the Beck Depression Inventory measure in Chinese students - general psychopathology or depression? Journal of Clinical Psychology. 47(3). 381-390. Solomon, L. & Rothblum, E. (1984). Academic procrastination: Frequency and cognitive behavioral correlates. Journal of Counseling Psychology. 31. 503-509. Sorotzkin, B. (1985). The quest for perfection. Avoiding guilt or avoiding shame? Psychotherapy. 22. 34-57. Strauman, T.J. (1989). Self-discrepancies in clinical depression and social phobia: Cognitive structures that underlie emotional disorders? Journal of Abnormal Psychology. 98. 14-22. Suddarth, B.H. (1996). A multidimensional analysis of perfectionism: Exploring a continuum of psychological traits. Unpublished doctoral dissertation. Pennsylvania State University. Sue, S. & Chin, R. (1983). The mental health of Chinese-American children: Stresses and resources, In G. Powell (Ed.), The psychological development of minority group childrea 385-397, New York: Brunner\Mazel. Sue, S. & Sue, D.W. (1974). MMPI comparisons between Asian-American and non-Asian students using a student health psychiatric clinic. J. Consult. Psychol. 21.423-427. Sue, S. & Zane, N. (1985). Academic achievement and soci-emotional adjustment among Chinese university students. Journal of Counseling Psychology. 32. 570-579. Suinn, KM., Ahuna, C, Khoo, G. (1992). The suinn-lew asian self-identity acculturation scale: Concurrent and factorial validation. Educational and Psychological Measurement. 52. 1041-1046. Suinn, R.M., Rickard-Figueroa, K., Lew, S. & Vigil, P. (1987). The suinn-lew asian self identity acculturation scale: An initial report. Educational and Psychological Measurement. 47. 401-407. Tanaka-Matsumi, J. & and Kameoka, V.A (1986). Reliabilities and concurrent validities of popular self-report measures of depression, anxiety, and social desirability. Journal of Consulting & Clinical Psychology. 54(3). 328-333. Tashakkor, A. & Barefoot, J. (1989). What does the Beck Depression Inventory measure in college students? Evidence from a non-Western culture. Journal of Clinical Psychology. 45(4). 595-602. 112 University of British Columbia Office of Budget and Planning (1998). Who Are We Now? A Profile of U.B.C's undergraduates [ Brochure]. Klawe, M M : Author. Vaughn, CE. & Leff, J.P. (1983). Patterns of emotional response in relatives of schizophrenic patients. Schizophrenia Bulletin. 7. 43-44. Weisse, D.E. (1990). Gifted adolescents and suicide. School Counselor. 37. 351-358. Weissman, AN. & Beck, AT. (1979). Development and validation of the Dysfunctional Attitudes Scale. Paper presented at the annual meeting of the Association for the Advancement of Behavior Therapy, Chicago. Westerfield, J.S. & Furr, SR. (1987). Suicide and depression among college students. Professional Psychology. 18. 119-123. Woolfolk, A.E. (1995). Educational Psychology (6th Ed.). Needham Heights, MA: Simon & Schuster Company. Zheng, Y., Wei, L., Lianggue, Guochen, Z. & Chenggue, W. (1988). Applicability of the Chinese Beck Depression Inventory. Comprehensive Psychiatry. 29(5). 484-489. APPENDIX A Ethnic Origin Categories Used in the Current Study 114 Ethnic Origin Categories from the Census Dictionary (Statistics Canada. 1996) that wereUsed to Classify Participants in the Current Study European Sample Participants from the following list of ethnic origin categories were included in the European sample. • British Isles origins • Western European origins eg., French, Belgian • Eastern European origins eg., Ukrainian, Polish • Southern European origins eg., Italian, Spanish Chinese Sample Participants from the following list of ethnic origin categories were included in the Chinese sample. • Chinese origins • Mainland Chinese • Hong Kong Chinese • Taiwan Chinese • Indo-Chinese origins • Cambodian • Laotian APPENDIX B Characteristics of the Chinese and European Sampli 116 Variable Chinese European Percentage Male 33 46 Percentage Female 67 54 Mean Age 20.38 20.93 Percentage Years 1 -5: Year 1 50 38 Year 2 29 38 Year 3 15 11 Year 4 5 12 Year 5 0 1 Percentage by Faculty: Arts & Science 82 80 Arts 57 62 Science 25 18 Engineering 5 11 Commerce 5 1 Nursing 4 4 Other 4 4 Self-identify as a Perfectionist YES 49 43 NO 51 57 Percentage of sample using: Hard copy forms 75 68 Website forms 25 32 APPENDIX C Letter to Professors Information Sheet on Research Study Poster T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A 1 1 9 Department of Counselling Psychology Faculty of Education 2125 Main Mall Vancouver, B.C. Canada V6T 1Z4 Tel: (604) 822-5259 Fax: (604) 822-2328 Information Sheet on Research Study Entitled: Perfectionism in university students from Chinese and Western-European cultural backgrounds PLEASE READ THE FOLLOWING INFORMATION SHEET PRIOR TO COMPLETING THE QUESTIONNAIRES $$ CHANCE TO WIN $50.00 CASH PRIZE FOR YOUR PARTICIPATION $$ Background: Perfectionism has been linked to both positive achievement strivings and adjustment difficulties in university students. Perfectionism has been studied almost exclusively in individuals from Western-European cultural backgrounds, while little, if any, attention has been given to studying perfectionism in other cultural groups. Given the cultural diversity of British Columbia, it is important to include individuals from different backgrounds so that results can be generalized to more individuals. Purpose: To test the current definition of perfectionism for meaningfulness and applicability to other cultural groups and to identify similarities and differences in perfectionism between the two cultural groups included in the study. Procedures: Individuals will be invited to participate in the study via entering classes to inform students of the study, with permission from the instructor, and by hanging posters with the appropriate contact number. Note: Participation in this study is completely voluntary and participants have the right to refuse to participate, or to withdraw at any time without penalty or without jeopardizing class standing. APPENDIX D List of U.B.C. Classes Entered to Recruit Participants 123 List of U.B.C. Classes Entered to Recruit Participants 1. One Anthropology 100-level class (daytime). 2. Four Sociology 100-level classes (2 daytime, 2 evening) 3. Three Asian Studies classes (daytime) • One 100-level class • One 200-level class • One 300-level class 4. Four Computer Science Classes (daytime) • Two 100-level classes • Two 200-level classes 5. Two 100-level Economics classes (daytime) 6. Two Engineering classes (daytime) • One 200-level Electrical Engineering class • One 300-level Chemical Engineering class 7. One 100-level Physics professor imported a project poster, advertising the study, into his website rather than having the researcher enter class. Two sections of 100-level physics students (daytime) accessed the website on a daily basis. APPENDIX E Pictures of the Website Used by Participants to Complete Questionnaires 125 The perfectionism website hosted at URL http://www.4di.ca/perfectionism. The website was comprised of the set of questionnaire forms that participants completed. Validity of the responses was checked automatically and if any responses were inappropriate or omitted, participants were prompted to correct them. If all responses were found to be legitimate, then the data were immediately entered into a database. JS HkHI-tLMUNISM S1UUY l-ALL I99B Nclscjpe Eb Edt » « » Co Eowaimaloc H * 3 3 ^ * ^ w ^ a * \ S f **w,ity>&#^  | Y a b > & "3 (PWheft netted UHNERStTY OF BRITISH COLUMBIA DEPARTMENT O f COUNSELLING PSYCHOLOGY PERFECTIONISM STUDY i Researchero O v t s t t n s V s r o Y ( M A CarwfidBto) A Dc B B t h r i a v w ^ a m c ( S u p c m r o r ) Information Sheet on Research Study Entitled: P**feetu*/usot in aosircrsss> jtudattxfivm Chines* and W«zt*m-Kump«an culiuml backgrounds PLEASE READ THE FOIXOWTNG INFORMATION SHEET PRIOR TO COMPLETING THE QUESTIONNAIRES. § § C H A N C E TO WIN S50.00 C A S H PRIZE F O R YOUR PARTICIPATION] I § § Background: Perfectionism has been linked to both positive achievement stnvingt and adjustment difficulties in university identify similarities and differences in perfectionism between the two cultural groups included in the study. Procedures: Individuals will be invited to participate in the study via entering classes to inform students of the study, with permtsrion from the instructor, and by hanging posters with me appropriate contact numbi iff t^iksfc^c " » iDot»»r tDS? students. Perfertionism has been studied almost exclusively in mdividuals from Western-European cultural backgrounds, while httte. if any. attention has been given to studying perfectionism m other cultural groups. Given the cultural diversity of British Columbia, it is important to include individuals from cSfierent backgrounds so that results can be generalized to more individuals. P ° T ° " ; T o definition of perfectionism for meattkgfulnest and applicability to other cultural groups and to 2i |S!*tFFELMIUNlSMSiUDrsFALL199S INclJCOPC nnt3 ,lh,l±'/^:^iXa<xamia (Jet I W ^ j ^ a 4 ^ ^ «• if v 2 2 1 n e w r f d B t e I c n a i f mee t m y pa ren t s ' expecta t ions . 2 X r I do do tt t t f iEr p u p ^ 2*: Other pacpta t o e m to a c c a t f tower s tandard* ftnen tt*yn*i>Mtt frwn I dn. 2ft lay pa ren t ! have a tvrayi h a d higher expec t t$t/ f \ ^ % f t ^ a n p w i QsJMJtin29 • 28 : l u s u a a y h a w Uuutjts about the aanple Z K N s a t n e ^ s M C f y k S f a n r t a r a l o ^ 3t t I expect M Q J K T per formance In mt tsasytaslts than mos t people. 3 t I s m 80Cat tau toa^B«rao! l- l ^^:«-^^ . i^< . * , . . .%^^^ 32 : I t a n d t o not bahsnd l i m y w o r k b e c a u s s 1 r e p M trdngs owar a n d n S M t a k t t n s 0 toaj ffeno to do tomot l snf l y a - - ;\'...-'/.;rj:.:w^ M : TTM f m w c m t s l n k a s ! crofca, 1S« m e r e pnnpkj w C Ed<B m * . Sot I oovar fefl Mta I couaa moot m y p a i a n U ' a t o n r l o f t s i . ^ ^ o t h r r a s r l u m p ( S u p e r j s o c j s ' ' ci C i C ] » i r - s f t r J r 3 c i r s C l f l C K l C i te«ti£igf :7:r ,3',c « ' f *8 Cl C 2 C] Pa r 5 C | C j C n r ? c i e i r a n c i CitjCl;,!? 4;f^4;cisl f i c 2 c 3 r4 r 5 <=• i <• 3 r c4 r 5 b^w•>^^3BllTroNtoro^^fioE^^,^^ APPENDIX F Questionnaire Package 127 1. 2. 3. 4. 5. 6. BACKGROUND INFORMATION SHEET The following questions request some background information. Given the cultural diversity of British Columbia, I hope to include individuals from different cultural backgrounds so that the results are relevant to more individuals. Thank you for taking the time to complete this questionnaire. Age Today Date Today DATE OF BIRTH (dd/mm/yy) Sex: Male Female Please check one related to your status as a student: Part-time Full-time Please check one related to your status as a student: Undergraduate Graduate Which university do you currently attend? Please check one. U.B.C. S.F.U. Please list your academic faculty/school: Other Other Other 8. What year in your program are you currently in? Please check the one that best fits. YEAR 1 YEAR 2 YEAR 3 YEAR 4 What is your cultural or ethnic background? 10. If Chinese, please indicate where you would place yourself, on a scale of 1-5, in terms of your cultural identity. Please circle the number that best applies to you. Highly Chinese Where were you born? 5 Highly Western If you immigrated to Canada, at what age did you come? 11. Where was your father born? 128 If not in Canada, how many years has your father lived in Canada? 12. Where was your mother born? If not in Canada, how many years has your mother lived in Canada? 13. PATERNAL GRANDPARENTS (Your father's parents): Where was your grandfather born? Where was your grandmother bom? 14. MATERNAL GRANDPARENTS (Your mother's parents): Where was your grandmother bora? Where was your grandmother born? 15. The first language you learned was: (Please select the one that best applies) English French Mandarin Cantonese Other If other, please list 16. Currently, what language io you use most? (Please select the one that best applies) English French Mandarin Cantonese Other If other, please list 17. What age did you start school here in Canada? 18. How many years have you attended school in Canada? 129 19. Do you consider yourself to be a perfectionist? Please check only one. YES NO 20. Please list 5 adjectives that you would use to describe someone who is a perfectionist. 1. 2. 3. 4. 5. 21. Have you ever participated in individual or group therapy for perfectionism? YES NO If yes, when? (month/year) MPS 130 Listed below are a number of statements oonoeming personal characteristics and traits. Read each item and decide whether you agree or disagree and to what extent. If you strongly agree circle 7; if you strongly disagree, circle 1; if you feel somewhere in between, circle any one of the numbers between 1 and 7. If you feel neutral and undecided the midpoint is 4. Disagree Agree t. When I am working on something, I cannot relax until it is perfect 1 2 3 4 5 6 7 2. I am not likely to criticize someone for giving up too easily 1 2 3 4 5 6 7 3. It is not important that the people I am close to are successful 1 2 3 4 5 6 7 4. I seldom criticize my friends for accepting second best 1 2 3 4 5 .6 7 5. I find it difficult to meet others' expectations of me 1 2 3 4 5 6 7 6. One of my goals is to be perfect in everything I do 1 2 3 4 5 6 7 7. Everything that others do must be of top-notch quality 1 2 3 4 5 6 7 8. I never aim for perfection in my work 1 2 3 4 5 6 7 9. Those around me readily accept that I can make mistakes too 1 2 3 4 5 6 7 10. It doesn't matter when someone close to me does not do their absolute best 1 2 3 4 5 6 7 11. The better I do, me better lam expected to do.... 1 2 3 4 5 6 7 12. I seldom feel the need to be perfect 1 2 3 4 5 6 7 13 Anything I do that is less than excellent will be seen as poor work by those around me 1 2 3 4 5 6 7 14. I strive to be as perfect as I can be 1 2 3 4 5 6 7 15. It is very important that I am perfect in everything I attempt 1 2 3 4 5 6 7 16. I have high expectations for the people who are important to me 1 2 3 4 5 6 7 \7„ I strive to be the best at everything I do 1 2 3 4 5 6 7 18. The people around me expect me to succeed at everything I do 1 2 3 4 5 6 7 19. I do not have very high expectations for those around me 1 2 3 4 5 6 7 20. I demand nothing less than perfection from myself 1 2 3 4 5 6 7 21. Others will like me even if I don't excel at everything 1 2 3 4 5 6 7 Copyright (c) Paul L Hewitt Ph.D., & Cordon L Flett, Ph.D., I98S 131 I Disagree Agree 22. I can't be bothered with people who won't strive to better themselves 1 2 3 4 5 6 7 23. It makes me uneasy to see an error in my work 1 2 3 4 5 6 7 24. I do not expect a lot from my friends 1 2 3 4 5 6 7 25 Success means that I work even harder to please others 1 2 3 4 5 6 7 26. If I ask someone to do something, I expect it to be done flawlessly 1 2 3 4 5 6 7 27. I cannot stand to see people close to me make mistakes 1 2 3 4 5 6 7 28. I am perfectionistic in setting my goals 1 2 3 4 5 6 7 29. The people who matter to me should never let me down 1 2 3 4 5 6 7 30. Others think I am okay, even when I do not succeed 1 2 3 4 5 6 7 31. I feel that people are too demanding of me 1 2 3 4 5 6 7 3Z I must work to my full potential at all times 1 2 3 4 5 6 7 33- Although they may not show it, other people get very upset with me when I slip up 1 2 3 4 5 6 7 34. I do not have to be me best at whatever I am doing 1 2 3 4 5 6 7 35. My family expects me to be perfect 1 2 3 4 5 6 7 36. I do not have very high goals for myself 1 2 3 4 5 6 7 37. My parents rarely expected me to excel in all aspects of my life 1 2 3 4 5 6 7 38. I respect people who are average 1 2 3 4 5 6 7 39. People expect nothing less than perfection from me 1 2 3 4 5 6 7 40. I set very high standards for myself 1 2 3 4 5 6 7 41: People expect more from me than I am capable of giving 1 2 3 4 5 6 7 42. I must always be successful at school or work 1 2 3 4 5 6 7 43. It does not matter to me when a close friend does not try their hardest 1 2 3 4 5 6 7 44. People around me trunk I am still competent even if I make a mistake 1 2 3 4 5 6 7 45. I seldom expect others to excel at whatever they do 1 2 3 4 5 6 7 Copyright (c) r«ul L Hewitt Ph.D., & Cordon L nclt, n«.D.. 1988 132 Multidimensional Perfectionism Scale (MPS - Frost et al. 1990) Please circle the number that best corresponds to your agreement with each statement below. Use this rating system Strongly disagree 1 2 3 5 4 Strongly agree Strongly Disagree Strongly Agree 1. My parents set very high standards for me. 1 2 2. Organization is very important to me. 1 2 3. As a child, I was punished for doing thing less 1 2 than perfectly. 4.If I do not set the highest standards for myself, 1 2 I am likely to end up a second rate person. 5. My parents never tried to understand my mis takes. 1 6. It is important to me that I be thoroughly 1 competent in everything I do. 7.1 am a neat person. 1 8.1 try to be an organized person. 1 9. If I fail at work/school, I am a failure 1 as a person. 10.1 should be upset if I make a mistake. 1 11. My parents wanted me to be the best at everything. 1 12.1 set higher goals for myself than most people. 1 13. If someone does a task at work/school better than 1 me, then I feel like I failed the whole task. 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 14. If I fail partly, it is as bad as being a complete 1 2 failure. 2 15. Only outstanding performance is good enough in my family. » 16. I am very good at focusing my efforts on attaining a goal. 17. Even when I do something very carefully, I often feel that it is not quite done right. 18. I hate being less than the best at things. 19. I have extremely high goals. 20. My parents have expected excellence from me. 21. People will probably thing less of me if I make a mistake. 22. I never felt like I could meet my parents' expectations. 23. If I do not do as well as other people, it means I am an inferior human being. 24. Other people seem to accept lower standards from themselves than I do. 25. If I do not do well all the time, people will not respect me. 26. My parents have always had higher expectations for my future than I have. 27.1 try to be a neat person. 28. I usually have doubts about the simple everyday tilings I do. 29. Neatness is very important to me. 30.1 expect higher performance in my daily tasks than most people. 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 2 3 4 5 31. I am an organized person. 32. I tend to get behind in my work because I repeat things over and over. 33. It takes me a long time to do something "right". 34. The fewer mistakes I make, the more people will like me. 35. I never felt like I could meet my parents' standards. Date: 135 Name: Occupation: Marital Status: Education: Age: Sex: Instructions: This questionnaire consists of 21 groups of statements. Please read each group of statements carefully, and then pick out the one statement in each group that best describes the way you have been feeling during the past two weeks, including today. Circle the number beside the statement you have picked. If several statements in the group seem to apply equally well , circle the highest number for that group. Be sure that you do not choose more than one statement for any group, including Item 16 (Changes in Sleeping Pattern) or Item 18 (Changes in Appetite). 1. Sadness 0 I do not feel sad. 1 I feel sad much of the time. 2 I am sad all the time. 3 I am so sad or unhappy that I can't stand it. 2. Pessimism 0 I am not discouraged about my future. 1 I feel more discouraged about my future than I used to be. 2 I do not expect things to work out for me. 3 I feel my future is hopeless and w i l l only get worse. 3. Past Failure 0 I do not feel like a failure. 1 I have failed more than I should have. 2 A s I look back, I see a lot of failures. 3 I feel I am a total failure as a person. 4. Loss of Pleasure 0 I get as much pleasure as I ever did from the things I enjoy. 1 I don't enjoy things as much as I used to. 2 I get very little pleasure from the things I used to enjoy. 3 I can't get any pleasure from the things I used to enjoy. 5. Guilty Feelings 0 I don't feel particularly guilty. 1 I feel guilty over many things I have done or should have done. 2 I feel quite guilty most of the time. 3 I feel guilty all of the time. 6. Punishment Feelings 0 I don't feel I am being punished. 1 I feel I may be punished. 2 I expect to be punished. 3 I feel I am being punished. 7. Self-Dislike 0 I feel the same about myself as ever. 1 I have lost confidence in myself. 2 I am disappointed in myself. 3 I dislike myself. 8. Self-Criticalness 0 I don't criticize or blame myself more than usual. 1 I am more critical of myself than I used to be. 2 I criticize myself for all of my faults. 3 I blame myself for everything bad that happens. 9. Suicidal Thoughts or Wishes 0 I don't have any thoughts of ki l l ing myself. 1 I have thoughts of ki l l ing myself, but I would not carry them out. 2 I would like to k i l l myself. 3 I would k i l l myself i f I had the chance. 10. Crying 0 I don't cry anymore than I used to. 1 I cry more than I used to. 2 I cry over every little thing. 3 I feel like crying, but I can't. Subtotal Page 1 THE PSYCHOLOGICAL CORPORATION* Harcourt Brace & Company Orlando • Boston San Diego • Philadelphia • Austin SAN ANTONIO New York • Chicago • San Francisco • Atlanta • Dallas Fort Worth • Toronto • London • Sydney Copyright © 1996 by Aaron T. Beck All rights reserved. Printed in the United States of America. Continued on Back 0154018392 11. Agitation 0 I am no more restless or wound up than usual. 1 I feel more restless or wound up than usual. I am so restless or agitated that it's hard to stay still. I am so restless or agitated that I have to keep moving or doing something. 12. Loss of Interest 0 I have not lost interest in other people or activities. 1 I am less interested in other people or things than before. 2 I have lost most of my interest in other people or things. 3 It's hard to get interested in anything. 13. Indecisiveness 0 I make decisions about as well as ever. 1 I find it more difficult to make decisions than usual. 2 I have much greater difficulty in making decisions than I used to. 3 I have trouble making any decisions. 14. Worthlessness 0 I do not feel I am worthless. J I don't consider myself as worthwhile and useful as I used to. 2 I feel more worthless as compared to other people. 3 I feel utterly worthless. 15. Loss of Energy 0 I have as much energy as ever. 1 I have less energy than I used to have. 2 I don't have enough energy to do very much. 3 I don't have enough energy to do anything. 16. Changes in Sleeping Pattern 0 I have not experienced any change in my sleeping pattern. la I sleep somewhat more than usual. lb I sleep somewhat less than usual. 2a I sleep a lot more than usual. 2b I sleep a lot less than usual. 3a I sleep most of the day. 3b I wake up 1-2 hours early and can't get back to sleep. 17. Irritability 0 I am no more irritable than usual. 1 I am more irritable than usual. 2 I am much more irritable than usual. 3 I am irritable all the time. 18. Changes in Appetite 0 I have not experienced any change in my appetite. l a M y appetite is somewhat less than usual, lb M y appetite is somewhat greater than usual. 2a M y appetite is much less than before. 2b M y appetite is much greater than usual. 3a I have no appetite at al l . 3b I crave food all the time. 19. Concentration Difficulty 0 I can concentrate as well as ever. 1 I can't concentrate as well as usual. 2 It's hard to keep my mind on anything for very long. 3 I find I can't concentrate on anything. 20. Tiredness or Fatigue 0 I am no more tired or fatigued than usual. 1 I get more tired or fatigued more easily than usual. 2 I am too tired or fatigued to do a lot of the things I used to do. 3 I am too tired or fatigued to do most of the things I used to do. 21. Loss of Interest in Sex 0 I have not noticed any recent change in my interest in sex. 1 I am less interested in sex than I used to be. 2 I am much less interested in sex now. 3 I have lost interest in sex completely. NOTICE: This form is printed with both blue and black ink. If your copy does not appear this way, it has been photocopied in violation of copyright laws. Subtotal Page 2 Subtotal Page 1 Total Score 136 TO BE COMPLETED BY THOSE FROM CHINESE CULTURAL BACKGROUNDS ONT.Y. CHINESE CANADIAN SELF-IDENTITY ACCULTURATION SCALE (C-CSIA) INSTRUCTIONS: The questions which follow are for the purpose of collecting information about your historical background as well as more recent behaviors which may be related to your cultural identity. Circle the one answer which best describes you. 1. What language can you speak? 1. Chinese only 2. Mostly Chinese, some English 3. Chinese and English about equally well (bilingual) 4. Mosdy English, some Chinese 5. English only 6. Other (please specify): 2. What language do you prefer to speak? 1. Chinese only 2. Mostly Chinese, some English 3. Chinese and English about equally well (bilingual) 4. Mostly English, some Chinese 5. English only 6. Other (please specify): 3. How do you identify yourself? 1. Chinese 2. Regionally defined ethnicity (e.g. Taiwanese, Indonesian, Vietnamese) 3. Chinese-Canadian 4. CartaoUan-Cliinese 5. Canadian 4. Which identification docs (did) your mother use? 1. Chinese 2. Regionally defined ethnicity (e.g. Taiwanese, Indonesian, Vietnamese) 3. Chinese-Canadian 4. Canadian-Chinese 5. Canadian 5. Which identification does (did) your father use? 1. Chinese 2. Regionally defined ethnicity (e.g. Taiwanese, Indonesian, Vietnamese) 3. Chinese-Canadian 4. Canadian-Chinese 5. Canadian 6. What was the ethnic origin of the friends and peers you had, as a child up to age 6? 1. Almost exclusively Chinese, other Asians, Asian-Canadians 2. Mostly Chinese, other Asians, Asian-Canadians 3. About equally Chinese/Asian groups and Anglo groups 4. Mostly Anglo or other non-Asian ethnic groups 5. Almost exclusively Anglo or other non-Asian ethnic groups 7. What was the ethnic origin of the friends and peers you had, from age 6 to 18? 1. Almost exclusively Chinese, other Asians, Asian-Canadians 2. Mostly Chinese, other Asians, Asian-Canadians 3. About equally Chinese/Asian groups and Anglo groups 4. Mostly Anglo or other non-Asian ethnic groups 5. Almost exclusively Anglo or other non-Asian ethnic groups 8. Whom do you now associate with in the community? 1. Almost exclusively Chinese, other Asians, Asian-Canadians 2. Mostly Chinese, other Asians, Asian-Canadians 3. About equally Chinese/Asian groups and Anglo groups 4. Mostly Anglo or other non-Asian ethnic groups 5. Almost exclusively Anglo or other non-Asian ethnic groups 9. If you could pick, whom would you prefer to associate with in the community? 1. Almost exclusively Chinese, other Asians, Asian-Canadians 2. Mostly Chinese, other Asians, Asian-Canadians 3. About equally Chinese/Asian groups and Anglo groups 4. Mostly Anglo or other non-Asian ethnic groups 5. Almost exclusively Anglo or other non-Asian ethnic groups 10. What is your music preference? 1. Chinese music only 2. Mostly Chinese music 3. Equally Chinese and English music 4. Mostly English music 5. English music only 6. Other (please specify): 11. What is your movie preference? 1. Chinese movies only 2. Chinese movies mostly 3. Equally Chinese /English-language movies 4. Mostly English-language movies 5. English-language movies only 6. Other (please specify): 138 12. Where were you bom? • Canada • China/Hong Kong/Taiwan • Other Where was your father bom • Canada • China/Hong Kong/Taiwan • Other Where was your mother bom • Canada • China/Hong Kong/Taiwan • Other Where was your father's father born? • Canada • China/Hong Kong/Taiwan • Other Where was your father's mother bom? • Canada • China/Hong Kong/Taiwan • Other Where was your mother's father born? • Canada • China/Hong Kong/Taiwan • Other Where was your mother's mother bom? • Canada • China/Hong Kong/Taiwan • Other Based on the above answers, circle the generation that best applies to you: 1. 1* generation (I was bom in China/Hong Kong/Taiwan or other) 2. 2nd generation (I was bom in Canada; either parent was bom in China/Hong Kong/Taiwan or other) 3. 3rd generation (I was bom in Canada; both parents bom in Canada; both grandparents bom in China/Hong Kong/Taiwan or other) 4. 4* generation (I was bom in Canada; both parents bom in Canada; at least one grandparent bom in Canada and at least one grandparent bom in China/Hong Kong/Taiwan or other) 5. 5* generation (I was bom in Canada; both parents bom in Canada; all grandparents also bom in Canada) 6. I don't know since I lack some information. 13. Where were you raised? 1. Asia (China/Hong Kong/Taiwan) only 2. Mostly in Asia, some in Canada 3. Equally in Asia and Canada 4. Mostly in Canada, some in Asia 5. Canada only 14. What contact have you had with Asia (China, Hong Kong, Taiwan)? 1. Raised one year or more in Asia 2. Lived for less than one year in Asia 3. Occasional visits to Asia 4. Occasional communications (letters/phone calls, etc.) with people in Asia 5. No exposure or communication with people in Asia 139 15. What is your food preference at home? 1. Exclusively Chinese food 2. Mosdy Chinese food, some Canadian (Western) 3. About equally Chinese and Canadian (Western) 4. Mostly Canadian (Western) 5. Exclusively Canadian (Western) 16. What is your food preference in restaurants? 1. Exclusively Chinese food 2. Mostly Chinese food, some Canadian (Western) 3. About equally Chinese and Canadian (Western) 4. Mostly Canadian (Western) 5. Exclusively Canadian (Western) 17. Do you-1. Read only Chinese 2. Read Chinese better than English 3. Read both Chinese and English equally well 4. Read English better than Chinese 5. Read only English 18. Do you-1. Write only in Chinese 2. Write Chinese better than English 3. Write Chinese and English equally well 4. Write English better than Chinese 5. Write only in English 19. If you consider yourself as a Chinese-Canadian, how much pride do you have in this group? 1. Extremely proud 2. Moderately proud 3. Little pride 4. No pride, but do not feel negative toward group 5. No pride, but do feel negative toward group 20. How would you rate yourself? 1. Very Chinese Z Mostly Chinese 3. Bicultural (equally Chinese and Canadian) 4. Mostly Canadian (Westernised) 5. Very Canadian (Westernised) 140 21. Do you participate in Chinese occasions, holidays, traditions, etc.? 1. Nearly all of them 2. Most of them 3. Some of them 4. A few of them 5. None at all 22. Rate yourself on how much you believe in Chinese values (e.g. about family, marriage, work, etc.) 1 (do not believe) 2 3 4 5 (strongly believe) 23. Rate yourself on how much you believe in Canadian (Western values). 1 (do not believe) 2 3 4 5 (strongly believe) 24. Rate yourself on how well you fit when with other Chinese people. 1 (do not fit in) 2 3 4 5 (fit very well) 25. Rate yourself on how well you fit when with other Canadian who are non-Chinese (Westerners). 1 (do not fit in) 2 3 4 5 (fit very well) 26. There are many different ways in which people think of themselves. Which ONE of the following most closely describes how you view yourself? 1. I consider myself basically a Chinese person. Even though I live and work in Canada, I still view myself as a Chinese person. 2. I consider myself basically as a Canadian. Even though I have a Chinese background, I still view myself as a Canadian. 3. I consider myself as a Chinese-Canadian, although deep down, I always know that I am Chinese. 4. I consider myself as a Chinese-Canadian, although deep down, I view myself as Canadian first. 5. I consider myself as a Chinese-Canadian. I have both Chinese and Canadian characteristics, and I view myself as a blend of both. APPENDIX G Perfectionism Means Reported in the Current Study and in Previous Studies 142 Table 28 Perfectionism Subscales Sample _Self Other Social Samples from Previous Studies 65.27-73.42 53.38 -59.57 45.92-56.88 Samples from Current Study: European 69.74 58.6 53.2 Chinese 70.54 55.22 54.28 European males 68.24 57.56 48.41 European females 71.01 59.49 57.27 Chinese males 73.69 56.61 52.54 Chinese females 68.96 54.52 55.15 Chinese - 1st generation 71.25 54.45 51.6 Chinese - 2nd generation 69.15 55.95 57.3 Chinese females - 1st generation NA NA 53.25 Chinese males - 1st generation NA NA 48 Chinese females - 2nd generation NA NA 57.9 Chinese males - 2nd generation NA NA 55.5 Chinese - low acculturation 70.5 55.2 51.6 Chinese - high acculturation 68.55 56.1 55.95 Note: Ranges of means listed from previous studies were drawn from five samples of Canadian university students reported by Flett et. al., (1995) and Hewitt and Flett (1991). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0053940/manifest

Comment

Related Items