llfllll'lr ll‘llfl'lvlflllllllflfifi Wm 3 1299 10802 will ” L This is to certify that the dissertation entitled A Comparison of Attitudes Toward Women's Participation in Sport Among Females and Males, Mexican—Americans and Anglo-Americans, and College Students and Members of the General Public presented by Judith Darlene Walton has been accepted towards fulfillment of the requirements for Ph.D. degree in School of Health Education, Counseling Psychology and Human Performance NSQLDK Major professor Date 59 own/144 J0 m» Ilf'li-n- AL‘ A ' F vi r r .1 .1 ‘ 042771 MSU LIBRARIES n RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. an :5 {A I" ,.-.~, I . 'Allfii? J.- r‘ at; «M 1 3 3i :1”: r“1 " l :3 0 Stag a. 1;? W31! 2 1 2003 J: 2 1 g o 3 Mixi- 652% A COMPARISON OF ATTITUDES TOWARD WOMEN'S PARTICIPATION IN SPORT AMONG FEMALES AND MALES, MEXICAN-AMERICANS AND ANGLO-AMERICANS, AND COLLEGE STUDENTS AND MEMBERS OF THE GENERAL PUBLIC BY Judith Darlene Walton A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY School of Health Education, Counseling Psychology, and Human Performance 1985 APPROVAL PAGE This dissertation has been approved by the following members of the Faculty of the Graduate School at Michigan State University. Dissertation AdvisoQ-mxflQszo RHW r u Committee Members mfl¢% 54 «2/6 ‘3 by ~ , / L/ ‘ ;_ 7 15*L// . .- ~- A; l.~~3/ JA 591407415“. 3 0, ”7575/ Dat'e of Acceptance by Committee ABSTRACT A COMPARISON OF ATTITUDES TOWARD WOMEN'S PARTICIPATION IN SPORT AMONG FEMALES AND MALES, MEXICAN-AMERICANS AND ANGLO-AMERICANS AND COLLEGE STUDENTS AND MEMBERS OF THE GENERAL PUBLIC by Judith Darlene Walton The purpose of this study was to assess and compare the attitudes of a: selected population toward the participation of women in sport, as well as attitudes toward the salience of sport to society and women's role in society. These attitudes were compared across ethnicity, gender, and educational status. Six: hundred, forty-one respondents stratified across the independent variables responded to a Likert—type scale developed for the study. The results of a 2 x 2 x 2 multivariate analysis of variance demonstrated that there were no significant interactions among the variables, but that there were significant main effects. A series of univariate analyses of variance indicated that Mexican—Americans and college students were significantly more positive in their «expressed attitudes toward the salience of sport to society JUDITH DARLENE WALTON than were Anglo-Americans and members of the general public; however, there was no difference between the attitudes of females and males. Women, Anglo-Americans, and college students expressed significantly more liberal attitudes toward women's role in society than did men, Mexican-Americans, and members of the general public. Supportive attitudes were expressed toward women's participation in sport, although females and college students exhibited significantly more supportiveness toward this role for women than did males and members of the general public. No difference was detected between the attitudes of Mexican-Americans and Anglo-Americans toward women in sport. Further ananlyses of the data revealed some serious limitations to the apparently supportive attitudes toward women in Sport and in society, however. Gender and educational status were salient determinants of attitudes toward women in sport and in society, but less important in explaining attitudes toward sport. Ethnicity appeared to be of little practical significance as a determinant of any of the three measured attitudes. ACKNOWLEDGMENTS Any major project encompassing the amount of effort involved in completing a doctoral degree could not be accomplished without the assistance, support and encouragement: of innumerable individuals. The following persons were instrumental in providing me with all that was needed to see this project through to the end and I most gratefully and sincerely acknowledge their much appreciated efforts. My parents, for providing me with unending support, encouragement, understanding and love and for instilling in me a thirst for knowledge and a capacity for hard work tempered with an appreciation for the enjoyment of the planet on which we all live. Dr. Annelies Knoppers, my major advisor and long-time friend, who provided me with constant support, encouragement and challenges. Her insightful editing and unwillingness to settle for less than what she considered to be the best stretched my self-imposed limits and helped me to grow as a writer and researcher more than I thought was possible. Dr. Marylee Davis, Dr. Marty Ewing and Dr. Donna Wanous Bullock, my doctoral committee members, who gave many helpful suggestions and provided constant inspiration ii to me by their examples of professional accomplishment and scholarship. Jesus Del Bosque, Acting Chair, and my colleagues and friends of the Texas Southmost College physical education department, who always understood and supported me, even when my split focus caused hardships for them. Dr. Juliet Garcia, Dean of Arts and Sciences at Texas Southmost College, who was instrumental in providing me with the time and support necessary to complete this degree. Her professional and personal acumen have been a source of inspiration and have given me goals to strive for---when I grow up I want to be just like her! Dr. R. Geneve Walton, my aunt and my first and foremost role model, who through her personal example taught me that there were no limits to what a woman could accomplish. She was doing things that a woman ”can't do" long before the contemporary women's movement taught the rest of us that there is nothing that a woman ”can't do" if she challenges the societal precepts which constrain her. Janet, Jacquelyn and Jessica, my sisters, whose well— timed letters and phone calls of encouragement helped lift my sagging spirits on many occasions. Dr. Wayne Lewis, who was always willing to make time for me in his busy schedule and whose professional support was invaluable in allowing me to complete the statistical analyses of my data. iii Dr. Nancy Van Noord, whose warm friendship, personal example of perseverance and triumph against great odds and high standard of graduate school performance was a constant source of strength and inspiration to me. Dr. Dianne Ulibarri, for providing me with shelter, support, encouragement and friendship when I needed it, and for never saying ”No" when I said ”let me ask you this?"! Barbara Reeves, Kay' Pratt, and Lisa Sheffield, my typists; Francisco, Olivette, Dawn and Elizabeth, my telephone assistants; and all persons who agreed to participate in my survey, for without their assistance my dissertation would have been impossible to complete. iv TABLE OF CONTENTS List of Tables vii List of Figures ix CHAPTER I: INTRODUCTION 1 Sport as a Transmitter of Values 2 Sport as a Socializing Institution 4 Statement of the Problem 9 Hypotheses 10 Significance of the Study 10 Definition and Explanation of Terms 12 Attitude 12 Educational Status 13 Ethnicity 13 Mexican-American 13 Sport 14 Limitations of the study 14 Summary 14 CHAPTER II: REVIEW OF RELATED LITERATURE 16 Introduction 16 Values of Anglo- and Mexican-American Cultures 17 Attitudes Toward the Value of Sport 23 Attitudes Toward Women in Anglo-American and Mexican-American Cultures 33 Attitudes Toward Women in Sport 42 Conclusion 52 CHAPTER III: PROCEDURES 55 Instrumentation 55 Development of the Instrument 57 The Pilot Study 63 Translation of the WASSS 70 Selection of the Sample 71 The Student Sample 72 The General Public Sample 74 Collection of the Data The Student Data The General Public Data Summary CHAPTER IV: RESULTS AND DISCUSSION Preliminary Analyses Comparison of Caller Groups Comparison of the Randomly Selected and Available Anglo Student Subsamples Reliability and Validity of the WASSS Analysis in Terms of the Hypotheses Summary of Results Further Analyses and Discussion Attitude Toward Sport (AS) Attitude Toward Women (AW) Attitude Toward WOmen's Participation in Sport (AWSP) Overall Discussion and Implications CHAPTER V: SUMMARY, CRITIQUE, AND RECOMMENDATIONS FOR FUTURE RESEARCH Summary Critique and Recommendations for Future Research Appendices References vi 75 75 77 80 82 82 84 85 92 100 102 103 115 130 141 150 150 158 164 178 LIST OF TABLES Composition of the Attitudinal Subscales and Their Sources Reliability Analyses of the Women in Sport and Society Survey Factor Analysis (Orthogonal Solution) of the Women and Sport in Society Survey (WASSS) Reliability Coefficients by Total Sample and Independent Variable Groups Factor Analysis (Orthogonal Solution) of the Women and Sport in Society Survey (WASSS) Factor Analysis (Oblique Solution) of the Women and Sport in Society Survey (WASSS) Summary of MANOVA for Subscales across Gender, Ethnicity, and Educational Status Oneway ANOVA Summary Table for Gender, Ethnicity and Educational Status for Dependent Variables Means and Standard Deviations According to Gender, Ethnicity, and Educational Status Discriminant Analyses of Subscales by Gender Discriminant Analyses of Subscales by Ethnicity Discriminant Analyses of Subscales by Educational Status Correlation of Selected Demographic Information with the Attitudinal Subscales Factor Analysis of the Attitude Toward Women Subscale vii 59 66 68 86 88 91 96 97 98 104 105 106 110 118 Discriminant Function Analysis on Attitude Toward Women and Attitude Toward Women in Sport Subscales Across Gender, Ethnicity, and Educational Status 123 Factor Analysis of the Attitude Toward Women in Sport Subscale (AWSP) 133 viii 4.1 LIST OF FIGURES Scheffe test of differences among caller groups for the Women in Sport and Society Survey (WASSS) 83 Pictorial presentation of the results of the ANOVA in relation to the hypotheses guiding 99 investigation ix CHAPTER I INTRODUCTION According to McPherson (1979), “Sport has been an integral facet of life in most societies throughout history” (p. 2). That this is true. of contemporary American society can hardly be disputed; one need only turn on a television set any weekend, glance at a newspaper, or peruse the local newsstand to realize the pervasiveness of sport in our society. Yet, it has only been in the past two~ decades that sociologists have "sensed in the institution of sport social parameters worthy of intellectual attention" (Ulrich, 1979, p. 11) and begun a concerted effort to discover and understand the relationships between sport and various aspects of American society (Coakley, 1982; Leonard, 1984; Snyder & Spreitzer, 1983). In order for sport to occupy the pre-eminent position that it does in American society, it must have some relevancy to the value structure of that society. Although it is extremely’ difficult to delineate typical .American values because of the enormous diversity of individuals and groups within this nation, there does appear to be a consensus concerning some of the dominant value systems operant in the country (Williams, 1970). Overall, Americans seem to value achievement, success, hard work, and upward mobility most highly. Sport as a Transmitter of Values Most Americans believe that sport transmits and expresses some of the dominant values of society (Snyder & Spreitzer, 1983). The Miller Lite Report on American Attitudes Toward Sports (1983) stated that 92% of their respondents believed that sports competition was good for children because it taught them to strive to do their best. Various authors (Berlage, 1982; Grove & Dodder, 1979; Jensen, Leonard, s Liverman, 1980; Nixon, 1979; Spreitzer & Snyder, 1975) have reported that adults believe that sport participation is good preparation for later life, makes better citizens, prepares one for life in the business world, teaches fair play, and develops self-discipline. For example, Berlage (1982) found that fathers of youth Sport participants believed that the emphasis in sports on teamwork, competitiveness, and tenacity was important for 1ater careers in business. Berlage agreed that these attributes are necessary for success in the business world and thus concluded that "children's competitive team sports are socializing agents for corporate America. From testimonials to corporate studies, athletes are shown to have advantages in business careers" (pp. 323-24). There is widespread support for the assertions that athletics are used to develop desirable attitudes toward the importance of achievement and success in our society and that the leaders of the board room learned necessary business skills on the playing fields of their youth (Bell, 1980; Berlage, 1982; Harragan, 1977; Snyder 8 Spreitzer, 1983). However, sport has historically been engaged in primarily by white males, so that they have been the main beneficiaries of the transmission of these values. Although women are certainly a viable part of society, they have been denied equal access to participate in sport throughout most of history (Gerber, Felshin, Berlin, & Wyrick, 1974). In addition, relatively little research exists concerning their limited involvement in sport. Hall (1977) surveyed 13 texts and anthologies focusing on the sociology of sport and found that only three of them contained separate chapters or sections concerning women's involvement in sport. This has changed somewhat since 1976 because of the impact of the women's movement, but the change has been minimal. We still know very little about women (Rosaldo, 1980) and even less about their involvement in sport, particularly Mexican-American women's involvement in sport. Not one reference to sport was found in this investigator's survey of four texts devoted to studying 'Las Mujeres," Mexican-American women, published between 1977 and 1980. Reference to minority women's involvement in sport in sociology of sport texts invariably pertains to that of black women. However, the Mexican-American population of the United States is the second largest minority population, and Mexican-American women's involvement in sport is a topic of legitimate concern for the researcher in sport sociology. Sport as a Socializing Institution As previously mentioned, sport participation in the United States has historically been primarily the prerogative of males. Many females have been shunted off to the sidelines and neither allowed the Opportunity nor encouraged to engage in competitive sports. Males, on the contrary, are not only encouraged, but also expected to be active sports participants. According to Hall (1977), "We know that sports participation has an important positive function in the deveIOpment of masculine identity but a negative function in the development of female identity" (p. 42). The differential consequences for each sex for participation in the same activity have generally been attributed to the differing role prescriptions for males and females in our society (Hall, 1977; Promoli, McCabe, & Shaw, 1977; Snyder, Kivlin, & Spreitzer, 1975). Traditionally, sex-role identity' has been interpreted in terms of masculinity and femininity, two concepts considered by some to be bipolar in nature. The characteristics ascribed to these terms have been thought to be biologically-based immutable differences determined by one's sex (Rosaldo, 1980). The concept of femininity encompasses behavioral attributes such as daintiness, expressiveness, and gentleness. Conversely, the concept of masculinity encompasses the behavioral attributes of strength, aggressiveness, confidence, and competitiveness (Chafetz, 1978). Therefore, the socially acceptable masculine image is defined in terms of leadership, dominance, and competitive tendencies. As indicated previously, these attributes are assumed to be developed partially through sports participation. Some members of society do not feel that these are desirable attributes for females. As a federal judge remarked, ”Athletic competition builds character in our boys. we do not need that kind of character in our girls" (Gilbert & Williamson, 1973, p. 95). Such persons evidently subscribe to the traditional view of a woman's proper role as being confined to the home as a wife and mother; therefore, she does not need to develop the so- called ”masculine" attributes necessary to compete and succeed in the business world. This view, however, denies reality. Over 50% of the adult labor force is comprised of women, many of whom are married. Many more are formerly married and now heads of households as well as sole supporters of households. Women work outside of the home for both psychological and economical reasons. They, therefore, need to develop the skills necessary to compete successfully for higher paying and more prestigious jobs in the world of work outside the home. Unfortunately, various factors have prevented women from developing the skills necessary for successful competition in the business world. The occupational system is highly differentiated by gender, with women concentrated in 10 positions found in the low-paying, low-prestige categories of clerical and service worker. Men, on the other hand, tend to be found in a wide variety of jobs in the high-paying, high-prestige managerial, craft, and laborer categories (Vanfossen, 1979). The concentration of women in such a narrow range of jobs effectively limits their opportunities for upward mobility while, at the same time, reducing competition with men so that the chances for mobility for men are not materially threatened. The mean salary for all male workers in 1980 was $25,973; for female workers the mean was about half that, or $14,914 (U.S. Bureau of the Census, 1983). The average salary for Mexican-American women is even lower. Comprising 47.5% of the Mexican-American labor force, the Mexican-American woman earns a mean salary of only $4,798 (U.S. Bureau of the Census, 1980). If participation. in competitive sport does, indeed, help to prepare the males of our society for life in the thsiness world, then it would appear that the exclusion of hubrnen from these competitive experiences would tend to Infiltzigate against their success in entering and effectiveness in staying in power positions in the same sphere. And, though evidence is less clear, the same may hold true for other facets of American society, such as politics. Furthermore, women's lack of participation in competitive sport may have other consequences as well. Oglesby (1978) identified psychological. damage which ‘may accrue when one is forced into stereotyped gender roles with which one is not comfortable. Ultimately, such enforced role ascription may result in total alienation from oneself and denial of the worth of one's existence. At a less severe level, this alienation from oneself may be manifested in lack of social acceptance, achievement problems, and fear of success. Thus far, most of what has been said about sport and females' involvement in sport has pertained to Anglo— American society. Most of the existing research concerning females in sport has been conducted with reference to white women. As stated earlier, almost no mention of women's participation in sport is found in literature pertaining to Mexican-American women. Therefore, it appears that little support for female involvement in sport in the Mexicanr American community exists and that the female Mexican- American athlete is invisible. This may be due, in part, t<> what some authors term the triple oppression of the Me at ican-Amer ican woman . Mirande and Enriquez (1979) stated that the oppression of the Chicana is threefold, ”as an ethnic minority, as a woman, and through internal Oppression within her own culture" (p. 130). Although the authors made no reference to sport in an entire volume devoted to the study of the Mexican-American female, some information pertinent to participation in sport can be extrapolated from their analysis of the Mexican-American female in education. They state that: All chicano (sic.) children are likely to view the Anglo teacher as an alien being with values and actions that deviate from cultural expectations, but the chicanita (sic.) also finds that the teacher does not conform to her expectations of feminine behavior . . . . Women are supposed to be warm and nurturing, but the teacher is aloof, cold, and assertive. Those very qualities which the school seeks to engender are typically in conflict with cultural expectations of feminine behavior. (p. 132) Mirande and ZEnriquez go on to say that the values the school seeks to develop, such as independence, assertiveness, and competitiveness, are in direct opposition to cultural norms for Mexican-American females. If a positive attitude towards achievement and success are not deemed desirable for females in Mexican-American culture, it seems reasonable to hypothesize that participation in sport would not be deemed desirable for the Mexican-American female. It would neither be aspired tC> by a female nor encouraged by male or female members of tile: Mexican-American society. However, as stated earlier, ;sL1c:h a hypothesis has little empirical verification. Statement of the Problem The purpose of this study was to assess and compare the attitudes of two selected populations of South Texas Anglo-American and Mexican-American adults toward the participation of women in competitive sports. Additionally, the attitudes of the same two populations toward women's role in society and the role of sport in society were assessed. A sample was drawn from the population of students enrolled in Texas Southmost College, Brownsville, Texas, during the spring and summer semesters of the 1983-84 academic year. A comparative sample was selected from the population of adults listed in the telephone directory of Brownsville, Texas, during the same time frame. The attitudes toward the role of sport in society, the role of women in society, and women's participation in sport were compared across gender, ethnicity, and educational status. It was assumed that the attitudes of the general pOpulation represented and were reflective of the status quo, whereas the students represented a new generation of leaders who will be instrumental in either maintaining the status quo or in changing existing conditions, values, attitudes, and opportunity structures. The attitudes of both of these groups were considered to be salient to an understanding of the current perception of women, competitive sport, and women's participation in competitive 10 sport in Anglo-American and Mexican-American society of South Texas. Hypotheses Specifically, the investigation was designed to test the following hypotheses. 1. There is no significant difference in the attitudes toward the role of sport in society across gender, ethnicity, and educational status. 2. The attitudes of males, Mexican-Americans, and the general public are significantly less positive than the attitudes of females, Anglo-Americans, and college students toward the role of women in society. 3. The attitudes of males, Mexican-Americans, and the general public are significantly less positive than the attitudes of females, Anglo-Americans, and college students toward women's participation in sport. Significance of the Study There appears to be a generally held perception in our society that participation in competitive sport is important training for developing characteristics deemed necessary for success, especially in the business world. Historically, access to this training has been almost exclusively a male prerogative in American society. When women were restricted to the traditional roles of wife and mother, this exclusion from competitive sport may not have had a great impact on their lives, for they rarely aspired to careers as business executives. However, since World War II, economic necessity has brought increasing numbers of women into the civilian labor force. In 1974, 28.1% of the civilian labor force consisted of women, but the 11 percentage has grown steadily until in 1982 women comprised 52.6% of the work force. At 47.5%, Hispanic women also comprise a high percentage of the Hispanic work force (U.S. Bureau of the Census, 1983). The contemporary women's liberation movement has done much to remove the stigma of women working outside of the home and also to convince women that they can and should aspire to traditionally "male" careers, such as business executive. This seeming change in attitude has not translated into effective occupational upward mobility and economic success, however. Women still are clustered at the lower rungs of the success ladder. Females comprise 80% of the clerical and kindred worker categories, 59% of the service workers, and 45% of the sales force. The percentages of women in managerial positions is much lower (U.S. Bureau of the Census, 1983). If participation in sport has helped prepare men for climbing the rungs of the ladder of success throughout history, it should do the same for women if they are provided with the opportunity. Historically, however, women have not been provided with the opportunity to participate in competitive sport. This has been due, at least in part, to attitudes limiting women's proper role in society to be that of a submissive, nurturing, supportive homemaker. There is some evidence to indicate that these attitudes are changing, at least for Anglo-American females (DeBacy, Spaeth, & Busch, 1970; Fisher, Genovese, & Morris, 1978; Harres, 1968; Hoferek & 12 Hanick, 1980; Kingsley, Brown, & Seibert, 1977; McGee, 1956; Nixon, Maresca, & Silverman, 1979; Promoli, McCabe, & Shaw, 1977; Snyder, Kivlin, & Spreitzer, 1975; Wittig, 1975; Woodford & Scott, 1982), but thus far there has been no such comparable work with reference to Mexican-American females. Yet, by the turn of the century, the Mexican- American population is predicted to be the largest minority group in the United States, and Mexican-American women are already represented in the labor force in (greater proportion than their Anglo-American counterparts. This investigation was an initial attempt to address some of the issues concerning the Mexican-American woman's participation in competitive Sport. Definition and Explanation of Terms For the purpose of clarification, the following definitions and explanations of terms are used in this investigation. Attitude "An attitude is an idea charged with emotion which predisposes a class of actions to a particular case of social situations" (Triandis, 1971, p. 2). It is composed of three components: cognitive, affective, and behavioral. This study was concerned with measuring the cognitive component. 13 Educational Status Educational status was a variable used to «classify respondents into one of two categories: (a) students who were currently attending a two-year college in South Texas, and (b) the general public, which consisted of any persons, 17 years and older, who were not currently attending an institution of higher education. Ethnicity Ethnicity was a variable used to classify respondents into one of two categories. Any person identifying her/himself as Caucasian or white was classified as Anglo- American, while any person identifying her/himself to be of Mexican ancestry was classified as Mexican-American. Mexican-American A Mexican-American was any person who considers her/himself to be of Mexican ancestry and who is residing in the United States of America. Several other terms such as "Latino," "Latin-American," ”Spanish-American," "Hispanic," and "Chicano" are also currently used to denote this population. However, the first four terms are considered to be too broad and inaccurate to describe the targeted population, as they also include members of diverse cultures such as Cuban and Salvadoran, who might confound the data in the current investigation. The fifth term, "Chicano," is considered to be pejorative by many Mexican-Americans of South Texas as it is associated with 14 the more radical elements of "La Raza." While the term is in vogue with many Midwestern and Farwestern Mexican- Americans, it is not considered to be the best term to refer to the targeted population of South Texas. Sport Sport is formally defined as the participation of highly skilled performers in athletic contests which are characterized by coaching, practice sessions, scheduled games or meets, the presence of Spectators, and the determinance of a winner of a contest. Participation in sport is considered to be distinct from participation in physical education or recreational activities. Limitations of the Study The investigation was subject to the following limitations: 1. selection of the study subjects from the student population of Texas Southmost College, Brownsville, Texas, during the Spring and summer semesters of the 1983-84 academic year; 2. selection of the general population sample from the telephone directory of Brownsville, Texas, during the Spring and summer of 1984; and 3. selection of the instrument designed to measure the cognitive attitudinal variables. Summary Although participation in sport has traditionally been a male prerogative, recent changes in our society have brought increasing Opportunities for females to participate 15 in the sport experience. 'This, however, has not been accompanied by a commensurate increase in studies concerned with women's participation in sport. Even more lacking are studies focusing on minority women's involvement in sport, specifically Mexican-American women. There seems to be a strict edict in the Mexican- American culture against women being independent, aggressive, and dominant. Thus, it would seem logical to postulate that attitudes toward their participation in sport would be negative since these are some of the qualities expressed in sport and developed by means of participation in Sport. If one of the functions of American Sport is to prepare individuals for success in society, the exclusion of women in general, and of Mexican- American women in particular, from that experience would appear to mitigate against their success in achieving parity in American society. However, little empirical evidence exists to support this hypothesis. This study will assess and compare the attitudes of a selected population of Anglo-American and Mexican-American adults and students toward the participation of women in Sport, the role of Sport in society, and the role of women in society. CHAPTER II REVIEW OF RELATED LITERATURE Introduction As discussed in Chapter I, sport is a significant institution in contemporary American society. To many people, individual achievement and success, as defined by the dominant members of American society, are a corollary of participation in competitive sport. In order for sport to occupy the prominent position that it does, it must have some relevancy to the value structure of American society. Values are those guidelines which determine what is desirable or acceptable behavior in a society or culture (Williams, 1970). American society is an enormous aggregate of people from many diverse races, nations, and cultures. AS in most societies, however, one group tends to dominate and, therefore, define the national value structure, while the other groups remain subordinate. In the United States, the dominant value structure has been defined by a white, Anglo—American, male society, while the Mexican-American culture has remained one of the subordinate groups. One of the conflicts which may arise between dominant and subordinate cultures is that of a difference in values. This may serve to effectively impede members of the 16 17 minority culture from achieving "success" as defined by the dominant culture, which usually means that minorities will remain the poorest and most destitute members of the society. Values of Anglo— and Mexican-American Cultures In a sociological analysis of American (Anglo) society, sociologist Robin Williams (1970) astutely observed that, ”Any attempt to delineate a national character or typical American values or a national basic personality type is extremely hazardous, not only because of serious gaps in the requisite data, but also because of the enormous value-diversity of the nation" (p. 450). He did, however, establish a set of criteria to which both dominant and subordinate values of a society must adhere and then delineated a number of values salient to contemporary Anglo-American culture. These highly prized values include belief in the importance of efficiency and hard work, individual achievement and material success, a pragmatic outlook on life, and a strong moral foundation with belief in democracy and freedom which extends to great national pride. Furthermore, Anglo-Americans tend to hold a universalistic outlook on the world with emphasis upon change and progress and a belief that the world can be mastered by individual effort. They have a strong faith in science and rationalism and generally emphasize equality in personal relationships. Despite the author's caution that 18 these are merely suggestive and certainly not immutable characteristics of dominant American culture, they have been interpreted as such (Guinn, 1977; Heller, 1971; Madsen, 1964). The Mexican-American culture within American society also has a value structure to guide the behavior of the members of its community. Some authors have ascribed value orientations to the Mexican-American culture which are antithetical to Anglo-American values (Guinn, 1977; Heller, 1971; Kluckholn s. Strodbeck, 1961; Madsen, 1964). Guinn (1977) summarized these attributed value differences as follows: Mexican-American Values Anglo-American Values Being rather than doing Doing rather than being Limited stress on material Material well-being ga1n Present time orientation Future orientation Simple patterns of work Individual action and organization and reaction group COOperation Central importance of the Impersonal relations family, personal relations Fatalism, accommodation Man's (sic.) mastery to problems over the universe Tradition, reluctance Change directed to change (p. 47) Success in American society is defined in terms of the dominant Anglo-American value structure. Accordingly, .Mexican-Americans' failure to exhibit the tangible signs of success such as upward mobility, status attainment, and economic well-being has been explained as emanating from 19 their supposedly differing value structure (Evans & Anderson, 1973; Heller, 1971; Kluckholn & Strodbeck, 1961; Madsen, 1964; Ramirez, Taylor, & Petersen, 1971). However, such a culturally deterministic explanation for the disparity between. Mexican-American and Anglo-American society has been widely criticized in social science literature (Gecas, 1980; Grebler, Moore, a. Guzman, 1970; Hernandez, 1977; Kuvlesky & Juarez, 1975; Montiel, 1973; Moore, 1976; Romano—V, 1973). Chandler (1979) succinctly summarized these criticisms in his observation that the studies which reported value differences between Mexican- Americans and Anglo-Americans were generally more than 10 years old, concentrated on small and/or isolated communities, were based on small sample sizes, focused on issues such as lack of achievement in schools rather than value orientations and did not make wide use of value- orientation measurement items. Chandler (1979) did, however, conduct a research study which supported the conclusions of the earlier studies which reported value differences. On the basis of data collected from a sample of 712 Anglos and 323 Mexican- Americans residing in Lubbock, Texas, Chandler reported that Mexican-Americans scored significantly lower than the Anglos on all dimensions of a scale designed to measure values deemed necessary to function effectively in modern societyu The scale included items designed to ‘measure fatalism, integration with kin, trust in people, and 20 occupational values. Items from each of these four areas were selected to comprise a total modernity scale, on which the Anglos scored significantly higher, i.e., more modern, than did the Mexican-Americans, 92% as compared to 44%. There are, however, several. methodological. problems ‘with this study. First of all, the Mexican-American sample was drawn from only one census tract in the central city, the only one with a heavy concentration of Mexican-Americans remaining after a tornado destroyed another barrio. The Anglo-American sample used all of the census tracts of the city, thus introducing a possible class bias. Secondly, the Mexican-American sample was interviewed by only male interviewers, whereas the Anglo-American sample was interviewed by both male and female interviewers, thus introducing a possible response sex bias. Thirdly, and perhaps most important, was the failure of the investigators to validate the instrument. Since no validity coefficients were reported, the interpretation of the scale as reported may be suspect. For example, agreement with the statement ”planning for the future only makes a person unhappy Since one's plans almost never come out right" (p. 157) is interpreted as being fatalistic and lacking a future time perspective, whereas it may actually be a quite rational adjustment to a minority's oppressed condition of existence (Moore, 1976). Finally, Chandler himself suggested that the generalizability of the findings 21 to other Mexican-American communities may be limited because of the "backward" status of the Mexican-American population of this particular locale. Other studies have refuted the traditional stereotype of the Mexican-American culture. Grebler, Moore, and Guzman (1970) reported that adult Mexican-Americans who were respondents to extensive interviews conducted in Los Angeles, California, and San Antonio, Texas, held predominantly modern views. That is, they believed they could actively master their environments, were future oriented, and, although familistic to some extent, believed in the worth of individual initiative. High educational achievement aspirations among Mexican-American teenagers and parents have been reported by .Anderson and .Johnson (1971), Espinosa, Fernandez, and ‘DornbuSch (1977), .Moerk (1979), and Schwartz (1971). Schwartz's data indicated that Mexican-American and Anglo-American high School pupils were similar in their goal aspirations, although the Mexican-Americans were more present oriented and more familistic than their Anglo counterparts. Montenegro (1976) reported that teenagers labeling themselves as Mexican-Americans held work and material success in high regard, were patriotic Americans planning to serve in the armed forces, had high educational achievement aspirations, and tended to subscribe to the traditional Anglo view of men and women having distinct roles based upon gender. 22 The preceding studies used samples of urban Mexican- Americans, so it is not surprising that the traditional stereotype of the fatalistic, particularistic, present- oriented Mexican-American was not substantiated. These traditionallv ascribed values of Mexican-Americans tend to be values associated with isolated folk societies (Hebding s. Glick, 1976), which Mexican-Americans once were. Now, however, Mexican-Americans tend to be concentrated in urban centers (Moore, 1976), so it is not surprising that their value structure appears to be in line with that of an urbanized society. At least one study (Gecas, 1980), however, reported high career occupational achievement aSpirations of rural Mexican—American youth similar to their urban counterparts and consistent with the achievement ethic of the dominant white male Anglo society. Gecas suggested that the Mexican-Americans' lack of high educational and occupational attainment may not be due to attitudes of value differing from Anglo society but rather to structural and situational obstacles stemming basically from the depressed economic circumstances of most rural Mexican-American families. As mentioned earlier, Moore (1976) concurred with this assessment and extended it to urban Mexican-American families as well. The picture that emerges from all of this information is one of a heterogeneous American society dominated by Anglo-American value orientations which are shared at least to some extent by the minority Mexican-American culture 23 (Bueter, 1972; Buzan & Phillips, 1980; Chandler, 1979; Eitzen & Sage, 1982; Garcia & Maldonado, 1982; Grebler et a1., 1970; Montenegro, 1976; Moore, 1976; Snyder & Spreitzer, 1983; Romano-V, 1973; Williams, 1970). Furthermore, the values of society as a whole as well as those of various subcultures do not appear to be immutable but, rather, constantly in a flux, changing as conditions within a society change (Bueter, 1972; Moerk, 1972; Williams, 1970). Although the full extent to which Anglo-Americans and Mexican-Americans share the same values has yet to be adequately demonstrated, limited evidence suggested that both cultures hold achievement in various spheres of endeavor in high regard. They both also seem to recognize the importance of hard work in striving to attain one's goals in life. These are two of the main values that participation in competitive sport is reputed to develop. Therefore, it would seem that competitive sport would be held in high esteem in both Anglo-American and Mexican- American societies. The following section will examine the attitudes of Anglos and Mexican-Americans toward the value of sport in American society. Attitudes Toward the Value of Sport There is a persistent belief among most Americans that participation in sport has positive benefits for those persons involved (Eitzen & Sage, 1982). A recent national 24 survey of American attitudes toward sport (Miller Lite Report on Sports, 1983) reported that only 3.7% of the United States' pOpulation was unaffected by sport as either a participant or spectator. ‘Participation in sport is believed to aid in the development of poise under pressure, help one gain control of aggressive tendencies, use leisure time more efficiently, build character, enhance one's self— esteem, develop self-discipline, and; promote physical fitness (Kniker, 1974). Despite criticism of its character-building capabilities (Coakley, 1982; Ogilvie & Tutko, 1971; Snyder & Spreitzer, 1983), competitive Sport continues to be held in high esteem by the American public. Sport is widely perceived as a tool to teach the skills and competencies deemed necessary to compete successfully and achieve status as defined by the dominant white male Anglo society. For example, one of the objectives of Little League baseball is to help youths develop the qualities of good citizenship. Similarly, Martens and Seefeldt (1979), using the results of a content analysis of over 400 articles concerning youth sports, identified 20 objectives of youth sports programs which included the following: --to teach children how to cooperate --to promote and convey the values of society --to develOp social competencies --to develop leadership skills 25 --to develop self-reliance and emotional stability by learning to make decisions and accept responsibilities --to teach children how to compete Obviously, the organizers of youth sport programs believe that these programs promote and convey the dominant values of American society and thus prepare its youth to assume a responsible place in that society. Parents of youth sports participants evidently agree with this belief. Griffith and Henschen (cited in Eitzen & Sage, 1982) and Phillips (1979) reported that parents of youth sports participants in New Mexico, Utah, and Virginia possessed favorable attitudes toward competition for children. The Miller Lite Report on American Attitudes Toward Sports (1983) reported that 92% of their respondents agreed that sports competition is good for children, mainly because it teaches them to strive to do their best. Additionally, Berlage (1982) found that 221 fathers of elite team participants on youth ice hockey and soccer teams in the Connecticut and New York metropolitan area held positive attitudes toward competitive youth sports. Ninety-three percent of the hockey players' fathers and 88% of the soccer players' fathers believed that involvement in competitive sports would be an asset to their sons in their adult business careers. Seventy-four percent of the fathers believed that a person has an advantage in applying for a job if he can attest to having played high school or college athletics. Furthermore, 57% stated that they would 26 be more inclined to hire an applicant who had played varsity Sports over one who had no such competitive sport experience. These fathers believed that the importance of sport for later business careers lay in sports' emphasis on teamwork, competitiveness, and tenacity. This positive attitude toward the benefits of sport extends from youth sports programs to high school and college athletics (Miller Lite Report on Spgrts, 1983). In a study of college students' attitudes toward intercollegiate athletics, Jensen, Leonard, and Liverman (1980) found that, overall, the respondents were positive about the benefits of this facet of higher education. The respondents were a pmrposive sample of 287 students enrolled at an Illinois university who responded to a 32 item Likert-type scale focusing on a variety of features of college athletics. The students perceived athletics as ”good" competition preparing one for competition in later life, stimulants to noble effort, promoting a concern for a healthy body, teaching a sense of fair play and good sportsmanship, contributing to the development of poise and self-assurance, providing an Opportunity for wholesome, organized activity, and preparation for life in the business world. (p. 9) There were no significant differences between ‘males and females in these overall perceptions as determined by a two-way analysis of variance and subsequent Tukey test. Another study' of (college students' attitudes toward sports (Nixon, 1979) found similarly that the majority of 27 525 undergraduate students, attending either a liberal arts college or a state university, believed participation in sport to be valuable. Specifically, 95% of the students believed that sport promoted self—discipline and over 80% indicated that participation in organized sport developed leadership qualities, built character, and made good citizens. However, only 23% of the respondents believed that sport participation was good preparation for later life, contrary to the findings of Jensen et a1. (1980). The attitudes of the general population toward sport have also been studied. Spreitzer and Snyder (1975) mailed a self-administered questionnaire to 945 respondents who were selected systematically from the city directory of a Midwestern city. Of the 945 selected, 510 returned the questionnaire, for a response rate of 54%. As has been the case with the studies cited thus far, at least 71% of the respondents believed that sport was important for the well- being of society, was valuable because it taught self- discipline as well as respect for authority and promoted good citizenship. Again, males and females did not differ significantly in these views. Grove and Dodder (1979) replicated this study with college students and faculty with essentially the same results. Both studies reported that persons with more education perceived sport to have greater psychosociological importance. However, Grove and Dodder (1979) found that older persons tended to perceive 28 greater social utility for sport than did younger persons, contrary to the original findings of Spreitzer and Snyder (1975). There is also evidence that young persons reflect the favorable attitudes toward sport held by adults. Participation in athletics is an important part of the adolescent status system. Coleman's epic study of adolescent society in 1961 concluded that the single most important criterion for high status among high school males was athletic prowess. This ranked considerably above intellectual achievement. Coleman's study was replicated by Eitzen (1975) and by Kane (1983), who found the results unchanged over a 20 year period. Evidently, Anglo-Americans of all ages, from teenagers to senior citizens hold primarily positive attitudes toward the role of sport in our society. They believe that sport participation has positive psychosociological functions for the individual as well as for society as a whole. One of the most prevalent beliefs, at least for adults, is that participation in sport teaches youngsters important Skills necessary for success in the competitive adult work world. All of the research cited thus far have used white samples and have focused primarily upon participation of white males in sport. There have been few studies in which the investigators have attempted to ascertain if the same positive attitudes towards sports' role in society are found among a minority, particularly a Mexican-American 29 sample. Is sport highly valued in Mexican-American culture? The current paucity of research in this area precludes any definitive answer to this question, though some indications of the value of sport in this culture are available. One such indicator, a survey of the portrayal of the athletic role in a dozen volumes concerning Mexican- Americans revealed not one textual reference to Sport. This indicates that sport may not be very highly valued in Mexican-American culture. However, at least one study contradicts this assumption. Hernandez (1977), on the basis of data collected from a sample of 100 Mexican-American and 112 .Anglo-American high school boys, concluded that Mexican-American adolescent males were more interested in Sport than were their Anglo counterparts, as measured by the Bohnke Athletic Attitude Inventory. Although sports participation and sports interest occupied a prominent place in life for both the Mexican-American and Anglo-American boys, this prominence was more pronounced for the Mexican—American adolescents. This was not true for the attitudes of the fathers of the boys toward sport, however, as perceived by the sons. The Mexican-American fathers tended to view their sons' participation in athletics less favorably than did the Anglo-American fathers. Hernandez found this to be a function of socioeconomic status, the poorer Mexican adults viewing sports less favorably than the more affluent 30 Anglos. However, the fact that Sport was more salient to the Mexican-American youths, who were generally from a poorer background than their Anglo counterparts, was consistent with Eitzen's (1975) findings that white adolescent males who came from poorer communities tended to view sport more favorably than those from more affluent communities. Another possible indicator of the value placed on sport in Mexican-American culture is whether or not youths aspire to a career in sport. Gecas (1980) studied career aspirations and expectations of rural Mexican-American youths in the state of Washington. He found that among the youths of 83 families, 25% of the 10-13 year old boys aspired to become entertainers or athletes when they grew up. This ranking was second only to the choice of a career in the professions among this age group. Only 9% of the 14-18 year old boys held such aspirations. However, when queried concerning occupational expectations, only 3% of each group actually expected to achieve their aSpirationS. Unfortunately, it is impossible to (determine the actual percentage of aspiring professional athletes because athlete and entertainer were considered together. It does seem plausible, considering the Hernandez (1977) study, that a greater number of young adolescent males aspired to the athlete role than to the entertainer role, reflecting a high value placed on Sport. The low expectation of actually achieving success probably reflects a: realistic 31 acceptance of the problems of poverty, discrimination, and lack of opportunity structures for rural Mexican-American youths (Gecas, 1980). Finally, leisure habits of sports ‘participation.:may also give a clue as to the value of sport in a society. McMillen (1981), in a study of leisure among Mexican- Americans in Houston, Texas, found that these persons had identifiable patterns of leisure involvement which did not differ significantly from a comparative sample of the general population. McMillen concluded that there was no significant difference between leisure patterns of Mexican- Americans in Houston and Anglo-Americans. By extrapolation, one :might infer that Sports participation was just as salient to one culture as to the other, provided that Similar Opportunity structures prevailed. Opportunities for participation were available in the community under study in the current investigation, although the opportunities for males were greater than for females at all levels. Sport programs for the youth of the community, conducted mainly by private organizations, included football, baseball, and soccer for boys, kickball for girls, and swimming for both. Although the baseball and soccer programs were ostensibly available for girls as well as boys, very few girls participated. The local Boys Club offered extensive programs for young boys and was Open to girls two days per week. 32 The Brownsville Independent School District sponsored an extensive sport program for high school athletes, with nine sports for boys but only seven for girls. Before the implementation of Title IX of the Education Amendments Act in the mid-19703, the only sport program offered for girls was track and field. The city parks and recreation department sponsored competitive sport programs for adults. 'Women were able to participate in one softball league and a mixed volleyball league. Men had the availability of three softball leagues, one basketball league, and a mixed volleyball league. The woman's softball league had 10 teams, while the men's leagues attracted close to 50 teams. There were eight public tennis courts, three swimming pools, and one municipal golf course available for this city of over 90,000 inhabitants. In the private sector, there were two country clubs, one health club, and two bowling alleys available for the leisure needs of the wealthier segments of the community. In summary, research evidence indicates that sport is highly valued, at least for males, in Anglo-American culture and may also be valued in Mexican-American culture, although the evidence in the latter case is less than substantial and is based upon indirect evidence. Similar Opportunities for participation do appear to be available in both cultural settings. But what about sport participation for females of both cultures? Is sport 33 valued for the benefits it may contribute to females in society as well as males? The final section will address these questions, after an examination of women's role in both Anglo-American and Mexican-American society. Attitudes Toward Women in Anglo-American and Mexican-American Cultures According to Dyer (1982), I'In modern western societies (and probably most other societies in the world) women have different roles, different expectations, carry out different tasks, have a different. range. of recreational interests, have different abilities, and are rewarded differently to men" (p. 4). In the United States during the first part of the twentieth century, those differences were manifested in the Victorian ideal of womanhood. This ideal emerged in England in the latter half of the nineteenth century during the reign of Queen Victoria, then transcended geographic boundaries to become integrally imbedded in American culture. The image of women which was ascendant during that era was that of a pale, frail, fragile female who was passive, obedient to her husband, and, above all, looked attractive. The ultimate goals of a fulfilling Victorian womanhood were twofold: attracting a man and bearing a child. Motherhood was considered to be the most womanly of all behaviors (Gerber et al., 1974). It should be noted that this image was pertinent only to the middle and upper class woman. The lower class woman 34 was too busy working to have the luxury of being frail or fragile. When American women gained the right to vote in 1920, the total restrictiveness of the Victorian ideal was temporarily cast aside in response to this new-found freedom. This was best exhibited by the ”flapper girls” of the Roaring Twenties. During the depression of the thirties, however, women were again restricted to their wife and mother roles as job Opportunities outside the home were drastically curtailed for both men and women. This situation changed once again with the advent of WOrld war II. Women were called upon to perform many of the duties vacated by the soldiers, including that of heavy labor. "Rosie the Riveter" became a symbol of women's reawakened sense of capability. This emergent feeling of fulfillment outside the home was not to be stifled once the war ended. Women were reluctant to return to their former homemaker duties to the exclusion of other jobs and this time were in the economic sector of the nation to stay. Unfortunately, the attitudes which were responsible for defining distinct male and female roles throughout history were also remarkably enduring. Traditionally, male and female roles, with their concomitant traits, were ascribed solely on the basis of gender. Males were expected to exhibit masculine traits such as strength, assertiveness, competitiveness, analytical thinking, self- confidence, independence, and logical thinking, while 35 females were to represent the polarity to 'these traits. They were to be passive, affectionate, cooperative, gentle, unintellectual, and have a strong need for security (Donelson & Gullahorn, 1977). The ascription of sex roles solely on the basis of gender has been increasingly challenged by feminists, and there are indications that a more egalitarian view of men and women's roles may be emerging in Anglo-American culture. In a study of sex-role attitudes of women from 1962 to 1977, Thornton and Freedman (1979) found a shift toward egalitarianism among the 1161 white Detroit-area women interviewed in 1962, 1963, 1966, and 1977. An egalitarian or liberal view toward women was defined basically as a belief that women and men should have equal rights and responsibilities in the occupational Sphere as well as in the division of labor at home. A. traditional or conservative orientation, on the other hand, was defined as the belief that being a wife and mother and taking care of the home is a woman's responsibility while the man should work outside the home to provide for it. In 1962 egalitarian responses to various questions concerning appropriate sex roles ranged from 32% to 56%. By 1977 the range of variation was from 60% to 77%. Thus, the lowest percentage of egalitarian response in 1977 was higher than the highest response percentage in 1962. Since age was not positively related to this shift in either sample, the authors attributed the change to the changing conditions of 36 the world rather than to maturational effects. Also, although there was a shift toward egalitarianism among all groups, those women with large families and those who were members of fundamentalist Protestant religions tended to retain more traditional attitudes toward appropriate sex roles than did the rest of the sample. Another longitudinal study of changes in sex role attitudes was undertaken by Parelius (1975). Two independent random samples of the students of a state- supported women's college in New Jersey were collected in 1969 and 1973. Three hundred, forty-seven women responded to a Likert-type questionnaire containing short descriptions of various women with either traditional or feminist sex role orientations. The respondents indicated the degree to which they were like the women described and whether or not men would want to marry women such as those described. The results indicated that women's attitudes Shifted substantially toward feminism between 1969 and 1973. The results also suggested that these women perceived men's attitudes toward gender roles to be basically traditional and less accepting of egalitarianism. The perception of men as less liberal than women in viewing gender roles has been substantiated by several studies utilizing the Attitudes Toward Women Scale (AWS), an instrument devised to measure attitudes toward woman's role in society. Hare-Mustin and Broderick (1979), Lunneborg (1974), Spence and Helmreich (1972), and Yoder, 37 Rice, Adams, Priest, and Prince (1982) focused on the attitudes of college students from various institutions around the country. The results consistently substantiated the view that females were Significantly more liberal in their attitudes toward woman's role in society than were the males. The results with respect to the relationship between age and attitudes were less consistent, however. Spence and Helmreich (1972) reported that their older respondents were more conservative than the younger ones, whereas Hare-Mustin and Broderick (1979) reported just the opposite. In two cross-cultural studies of attitudes toward women, Loo and Logan (1977) and Braun and Chao (1978) also found that the male respondents scored more conservatively than their female counterparts on the Attitudes Toward Women ScaLe. Loo and Logan found that the Canadian males were more liberal than the southern United States' males (Spence & Helmreich, 1972) but more conservative than northern United States' male students (Lunneborg, 1974). The Canadian females were more liberal than the southern female sample but were not significantly different from females in the northern U.S. Braun and Chao compared 74 white Americans matched for age and education with 85 Chinese-Americans who had been born in Asia. In general, the white females and Chinese males tended to share liberal views toward women, while the white males and Chinese females tended toward conservative attitudes towards 38 woman's role in society. Evidently there are widely divergent views of (appropriate behaviors and spheres of influence for men and women held by various segments of white American culture. This may also be true for Mexican- American culture. As is the case with Mexican-American attitudes toward sport, little empirical evidence exists to delineate attitudes toward apprOpriate role behavior of Mexican- American women. Some data-based studies do exist, however, as well as a growing body of phiIOSOphical and theoretical literature by Mexican-American scholars concerning the place of women in society. Flora (1971), in an attempt to assess the status of women in different cultures, studied the image of women as portrayed in the fiction found in women's magazines. On the basis of a content analysis of 202 short stories selected randomly from women's magazines published in the United States, Columbia, and Mexico, Flora reported that women were portrayed as Significantly more passive and dependent in the Latin American stories than they were in the North American stories. Very few examples of women actively controlling their lives were presented, with only seven percent of the fictional women improving their socioeconomic status through their own efforts. Evidently fiction in both cultures promotes the attitudes that passivity and dependence on males are desirable traits in females of both cultures, but the emphasis is greater in 39 the Latin culture. This picture of the passive female and her lack of status was reflected in Mexican-American women in a study of representation in municipal governments by Karnig and Welch (1979). These investigators reported that while women as a whole were underrepresented in participation in municipal government, Mexican-American women were considerably less evident than either Black or Anglo women. The authors computed an equity ratio by dividing the percentage of city council seats held by a particular group by the percentage of that group present in the city population. They reported that the equity ratios for Anglo women, Black women, and Mexican-American women were .33, .16, and .11, respectively. The investigators concluded that minority women, especially Mexican-Americans, suffered from what has been termed a ”double jeopardy"--they must overcome prejudicial attitudes against both their gender and their ethnicity. The traditional Anglo view of the female's role in society to be supportive of the male and to be limited to home and family is also prevalent in Mexican-American society. According to Mirande and Enriquez (1976), "It is clear that motherhood pervades the cultural heritage of the Chicana . . . . Chicanas who do deviate from the ideal wife-mother role are inevitably regarded as betrayers or 'vendidas,‘ cultural sell-outs" (p. 96). Most social science literature of the sixties and even the seventies 40 described the Mexican-American family as having a male- dominated, authoritarian. structure. Supposedly, husbands totally dominated their wives, made all the decisions, demanded total obedience and respect, and did pretty much as they pleased with impunity (Heller, 1976; Madsen, 1964; Penalosa, 1970). Increasingly, however, this concept of total male domination has been reexamined. Hawkes and Taylor (1975) studied decision-making in 76 Mexican-American farm labor families and concluded that decisions were made in an egalitarian mode. Eighty-three percent of the families interviewed 'Said that they shared decisions pertaining to money matters, family size, home location, disciplining children, and whether the wife should work outside the home. Grebler, Moore, and Guzman (1970) also reported that the Mexican-American family differed greatly from the cliche of the dominant male and submissive female, with decision-making comparatively as egalitarian as .Anglo families of similar SES. Mirande (1979) concluded in a review essay of male dominance in the Mexican-American family that sharp) sex role distinctions appear to be rare in Chicano families and that the traditional male dominant view should be seriously questioned. Although the results of these studies have suggested that the traditional male dominance patterns of Mexican- American families may be less universal than ‘previously assumed, various feminist writers have challenged this 41 assumption (Baca ‘Zinn, 1982a, 1982b; Gonzales, 1977, 1979; Longeaux y Vasquez, 1970; Mirande & Enriquez, 1979; Nieto, 1974; Sutherland, 1970). They have argued that while male dominance may not be the most prevalent mode of decision- making in Mexican-American families, such dominance certainly is present in families and other organizations within Mexican-American society. In fact, they have asserted that all women, Anglo and Mexican-American, are victims of sexism. The Mexican-American woman, like her Anglo counterpart, has traditionally been relegated to the role of supportive homemaker, with various forces operating to keep her out of the business and political spheres of influence. Yet it is outside the domestic sphere where money is made, laws are made, and decisions which impact greatly on the life of this planet are made-~spheres from which Mexican-American women are, for the most part, evidently excluded. Whereas attitudes are changing which allow for the Anglo—American woman to become increasingly active in areas outside the home, little evidence exists to determine if the same is true for the Mexican-American woman. This investigation was designed to provide data to help in determining the present status of the attitudes toward Mexican-American woman's role in society. If women are to make gains in performing influential duties outside the domestic sphere, participation in sport may well provide an important training ground for their 42 future success. Many Americans view Sports participation as important business preparation for males in our society, thus it ought to perform the same function for females. An important determinant of such participation by females is the current status of and attitudes toward women's participation in sport. The following section will examine this issue. Attitudes Toward Women in Sport Males and females have traditionally been ascribed differential roles in American society based solely upon gender. Those individuals who did not fulfill their role expectations have generally faced varying degrees of social sanctions. One of the greatest differences in role ascription has been in the realm of participation in sport (Sage & Loudermilk, 1979). For males in our society, sport participation has been used as an agent of socialization, to imbue boys with the masculine qualities deemed necessary to be a man. On the other hand, sport participation for females has been considered definitely unladylike, and girls have been discouraged from participation, especially once they entered adolescence. Sport participation is considered valuable for males of‘ our society, but not for females. Conditions are changing, however, and there is evidence that attitudes toward females' participation in sport may also be changing. 43 In a discussion of the value of sports participation for women in affecting social change, Westkott and Coakley (1981) ‘argue that there are essentially two divergent viewpoints concerning this matter. The critical viewpoint, which depends heavily on radical feminist scholarship, rejects patriarchal institutions altogether, particularly the values and attitudes which have traditionally been associated with masculine behavior. {Proponents of this viewpoint contend that women should not seek access to institutions such as Sport, which require masculine defined behavior for success. Instead, women Should seek to change the prevailing institution by forming an alternative one without such behavioral limitations. The opposite viewpoint is that of the assimilationist, who argues that “ . . . feminist change is defined in terms of women gaining access to culturally valued spheres of action rather than changing those cultural values and actions“ (p. 32). The proponents of this viewpoint maintain that if women are to be valued by society, they must gain access to the means to internalize cultural values from which they have so long been excluded, such as sport. Access to participation in sport is important for women because it has some important benefits to its participants. According to Westkott and Coakley, these benefits are the following. 44 1. It gives her the Opportunity to develop a sense of herself as she acts for herself as Opposed to meeting her needs through serving others (individuation). 2. It helps her to separate herself from the limitations and demands of family traditions and ascriptive roles. 3. It helps her reconnect with the power of her body. The assimilationists contend that by developing the skills and attributes that accrue from sports participation, women are more capable of entering and being successful in spheres of influence outside the home. Increasingly, girls and women are gaining access to sports participation, due mainly to federal legislation. There is evidence, however, that this increased access has not always been viewed favorably by all segments of society, especially those who possess decision-making powers. McGee (1956), in one of the earliest studies of attitudes toward intensive competition for girls, studied the attitudes of administrators, teachers, and parents of high school girls in Iowa and Illinois. Subjects were selected by stratified random sampling procedures from high schools in Iowa and Illinois which did not Sponsor girls' sports. Subjects (N = 1347) were asked to respond to a self-administered Likert-type scale designed to measure attitudes toward intensive competition for high school girls. The data were then subjected to an analysis of Variance with subsequent t-tests to test significances between any two means. 45 As expected, all respondents from schools which sponsored competitive sports for girls exhibited more favorable attitudes toward competition for girls than did those respondents from schools not sponsoring such programs. However, parents from all three groups showed significantly' more favorable attitudes than did administrators and teachers other than coaches. This favorable response of parents toward competition for girls appears to have remained fairly constant over the years. In 1983, the Miller Lite Report on American Attitudes Toward Sports reported that 86% of the respondents believed it important for girls to participate in sports, and 95% felt it important for boys to participate in sports. A great majority also stated they would be happy if their daughters (88%) or sons (92%) became professional athletes when they grew up. A number of studies have attempted to assess the attitudes of college students toward sport competition for women. Harres (1968) administered a Likert-type attitude inventory and demographic questionnaire to a stratified random sample of 300 undergraduate students at the University of California at Santa Barbara. Two hundred eighty-four (94%) correctly completed forms were subjected to analysis of variance with subsequent Newman-Keuls tests of differences between ordered means. The results indicated that the respondents were favorable, although not highly favorable, toward athletic competition for women, 46 with no significant differences between men and women's attitudes. A pilot study to determine the attitudes of college men toward women's competition was undertaken by De Bacy, Spaeth, and Busch (1970). A modified Likert instrument was used to collect data from 90 male physical education majors and 90 male non-physical education majors at the State University of New York at Buffalo. The results indicated that both groups were only slightly positive in attitude toward women's competition in sports, with no significant difference between the physical education majors and non- majors' attitudes. The authors concluded that while competitive opportunities for girls and women are increasing rapidly, the male support necessary for continued success is not increasing commensurately. Kingsley, Brown, and Seibert (1977) presented evidence to indicate that attitudes towards women's participation in sport were becoming more positive among women. The investigators asked 240 female athletes and non-athletes at a Southwestern university to respond to vignettes depicting a high or low aspiration female dancer or softball player by rating them on four points concerning social acceptability. Contrary to their expectations, the authors found no significant differences in the social acceptability ratings of the dancer and softball player by the non-athletes. The sample of athletes rated the softball player as significantly more acceptable than the 47 dancer, giving some indication that some sports once considered socially unacceptable for female participation are now enjoying increased acceptability at least among female college students. A projective technique was used by Wittig (1975) to study the attitudes of 225 volunteer students at a Midwestern university. Male and female students responded to stories depicting a male or female in a sports situation by describing the individual's appearance and behavior. Results from the first data collection Showed that both men and women responded positively to the successful male-in- sport cue, but the men responded significantly less favorably than the women to the successful female-in-sport cue. Wittig, however, was not certain whether this difference was due to men's overall disapproval of women's participation in sport or to the cue which depicted the female as invading a previously all-male team; so in a subsequent data collection six months later, he changed the cue to depict a female as achieving success on a female sports team. This time there was a significantly positive shift in the men's attitudes toward the approval of women succeeding in sport. Wittig again was unsure how to explain this change. The change in cue may have been re5ponsible for the positive shift in attitude, or it may have been due to extraneous influence occurring within the intervening six months, such as publicity concerning Title IX of the 1972 Education Amendments Act. 48 Another study of attitudes toward participation in sports was undertaken by Promoli, McCabe, and Shaw (1977). They compared the attitudes of 160 randomly selected male and female physical education majors and students enrolled in introductory psychology classes at a Canadian university. Subjects responded to vignettes depicting ”Pat” variously as a female figure skater, male figure skater, female ice hockey player, or male ice hockey player by describing Pat's appearance and behavior. The only significant differences found among all comparisons of males, females, physical education majors, and psychology students was the attitude toward Pat when she was depicted as a female ice hockey player. In this instance, female students viewed Pat significantly more positively than did physical education majors. Overall, respondents' attitudes toward females competing in sports were generally positive except for male physical education :majors ‘who’ responded negatively to female involvement in a male dominated sport. Fisher, Genovese, Morris, and Morris (1978) examined the perceptions of male/female and athlete/non-athlete students in a liberal arts college located in upstate New York. The subjects viewed 15 slides of female athletes participating in various Sporting events and then rated each slide on a scale of 1—10 indicating how close the female depicted came to being an "ideal female." A rating of l denoted the subjects' perception of an "ideal female," whiLe 10 was the farthest removed from that ideal. Males 49 in the study rated the slides as significantly less ideal than did females, and non-athletes rated the slides as significantly less ideal than did athletes. These findings were consistent with those already reported in which males viewed females' participation in sports less favorably than did females. Finally, Nixon, Maresca, and Silverman (1979) surveyed student attitudes toward females in sport at two colleges in Vermont. A randomly selected sample of 760 students was mailed a self-administered questionnaire to which 65% (N = 494) of the students responded. On all 11 questionnaire items, women were significantly more positive toward women participating in sport than were men. For example, 53% of the women felt that females should be given more encouragement to participate in sports of all kinds, whereas only 29.7% of the men agreed with this statement. Although males (85.5%) and females (95.9%) both rejected the idea that sport should be restricted mostly or entirely to males, men had significantly more reservations about which Opportunities were permissible, i.e., only 46.3% thought girls should be allowed to compete in Little League baseball, whereas 73.2% of the women felt they should be permitted to do so. Overall, the results of the studies concerning attitudes of college students toward the involvement of females in sports indicated that women view sport involvement by females significantly more positively than 50 do males. Although a few studies reported no significant differences in the attitudes of males and females (Harres, 1968; Wittig, 1975), most did find differences, with the attitudes of males being less positive than those of females (Fisher et al., 1978; Nixon et al., 1979; Promoli et al., 1977; Wittig, 1975). 'This would seem to have important ramifications for future female involvement in sports, as men still control most decision-making positions in organized sport. Several authors have investigated the attitudes of the general public towards females' participation in sport. Snyder, Kivlin, and Spreitzer (1975) asked subjects of a probability sample of the general pOpulation of MetrOpolitan Toledo, Ohio, if certain Sports would enhance or detract from a girl/woman's feminine qualities. Over half the respondents» answered that swimming, tennis, (and gymnastics would enhance feminine qualities, while leSS than 15% believed softball, basketball, and track would do so. Conversely, 20 to 30% believed track, basketball, and softball detracted from feminine qualities, whereas only two to six percent believed tennis, gymnastics, and swimming to do so. Unfortunately, ”feminine qualities" were not defined, nor were the respondents asked how important feminine qualities were believed to be. Although not stated in such terms, these results might be interpreted to mean that the general pOpulation would discourage girls/women from participating in team sports, 51 which again has possible negative consequences for the success of women outside the homemaking sphere. Hoferek and Hanick (1980) replicated the Snyder, Kivlin, and Spreitzer study by utilizing a random sample of adults living in permanent structures in an Iowa town. The results, based on a 51% response rate, indicated that respondents viewed participation in sports as neither detracting from nor enhancing femininity. The authors speculated that traditional, rigid sex role stereotypes had been transcended by the subjects in their sample. Using a sample of 353 respondents randomly selected from the directory of Oklahoma City and its suburbs, Woodford and Scott (1982) assessed the attitudes of the general public concerning women's participation in sports. Subjects responded on a four-point scale to statements regarding women's participation in Sports. Overall, the results indicated that the respondents viewed the participation of women in competitive Sports favorably. Fewer than 12% agreed that participation in organized sports takes time away from other important things a woman ought to be doing or that a woman cannot be both feminine and a good athlete. There were no significant differences between males' and females' responses. However, when compared by education levels, those with less education were significantly more negative toward women's participation in sport than those with some college education. In addition, the respondents over 50 years of 52 age were also more negative in their attitudes than their younger counterparts. Contrary to some other studies, however, there were no significant differences reported in attitudes toward team sports' participants as compared to those in individual sports. The results of these studies of attitudes of the general population toward the participation of females in sport are somewhat contradictory. ' Whereas one study (Snyder et al., 1975) reported that respondents viewed participation in team sports significantly less favorably for women than participation in individual sports, others (Hoferek & Hanick, 1980; Woodford & Scott, 1982) reported no such difference. One explanation for the conflicting results may be that Snyder et al. reported their results from data which had been collected in 1973. Since that time, there has been a significant increase in opportunities for female participation in sport as well as increased media attention. Conclusion Sport appears to occupy a prominent position in American society, mainly because it is believed that Sport helps to develop and transmit the dominant values of that society to its future citizens. Historically, the benefits derived from Sport participation accrued mainly to the male citizens of American society, especially white males. Traditional attitudes defining women's role in society to 53 be confined mainly to the home have existed in both Anglo— American and Mexican-American cultures and have helped to effectively discourage active female involvement in sport. There is growing evidence to suggest that such restrictive attitudes toward women's role in society and women's participation in sport are changing within the Anglo-American culture, but no such evidence» exists for Mexican-American women. Additionally, the majority of studies indicating changes in attitudes have focused on college student populations. There have been differences reported between the attitudes of female and male college students, however, with males being generally less favorable toward women's expanded role in society and toward women's participation in Sport. The few studies that have focused on attitudes of the general public have also generally reported an increase in favorable attitudes toward expanding women's role in society and participation of women in sport. Unfortunately, only one study focused on the importance of Sport participation in Mexican-American society, and that study focused on adolescent males. Also, none of the studies focusing on the participation of females in Sport provided comparative data concerning the attitudes of the sample population toward the role of females in society, nor even the salience of sport to that particular pOpulationJ .A comparative analysis among different studies is difficult because a variety of 54 instruments for measuring attitudes toward sport have been used,with no report of reliability or validity coefficients for the measuring devices. This investigation sought to remedy some of these problems by providing a three-pronged investigation of the attitudes of male and female Mexican-Americans and Anglo- Americans. Students as well as members of the general public responded to an instrument designed to measure attitudes toward sports, toward women's role in society, and toward women's participation in Sport. CHAPTER III PROCEDURES The investigator assessed and compared the attitudes of a selected pOpulation of South Texans toward the role of Sport in society, the role of women in society, and women's 'participation in sporta The attitudes, which were the dependent variables of the investigation, were measured by means of a summated rating scale developed for the study. Attitudes were compared across gender, ethnicity, and educational status, which were the independent variables of the investigation. The scale was administered to a sample of the target pOpulation of students during the spring and summer semesters of the 1983-84 academic year. The responses from a sample of the general public were collected during the same time period. Selected demographic information was also collected via the survey instrument. Instrumentation In general, two methods of data collection are most widely used in survey research such as the current investigation. These are the self-administered questionnaire and the interview. Both of these techniques have inherent advantages and disadvantages, but the self- 55 56 administered questionnaire appears to be more amenable to the surveying of large numbers of respondents, most cost efficient and time effective, and more appropriate for dealing with what may be sensitive issues for some respondents because of its greater assurance of anonymity than an interview (Babbie, 1983). Since all of these issues were considerations in the present investigation, the self-administered questionnaire was used to collect the data, with slight modifications of the technique. Both the student and the general public samples responded to the same survey instrument. The students, however, responded to a written questionnaire while the general public responded to the same instrument over the telephone. The reasons for this difference in the collection of data will be delineated subsequently in the section of this chapter concerning sampling. As early as 1967, Bonjean, Hill, and McLemore expressed concern about the lack of continuity in sociological research. They pointed out that in a survey of 2080 written instruments used for gathering data during a 12 year span, only 2.8% of the measures were used more than once, while only an abysmal 2.26% were used more than five times. According to Bonjean et al., this lack of continuity severely restricted the comparative interpretations of the research findings. This same concern about the lack of continuity in measuring instruments is also applicable to the assessment 57 of attitudes toward the participation of females in competitive sport, which was the primary focus of this investigation. Although a search of the related literature indicated the existence of a small variety of measuring devices, none of them had been utilized with consistency over a number of different studies. The form of measurement device which was most widely reported in the literature, however, was the summated rating or Likert-type scale (DeBacy et al., 1970; Harres, 1968; McGee, 1956; Nixon et al., 1979; Selby & Lewko, 1976; Sherriff, 1971; Snyder et al., 1975; Woodford & ‘cott, 1982). Development of the Instrument This investigation employed survey research procedures utilizing the Likert-type Scaling technique to measure the dependent variables. Such a scaling technique is an excellent means of measuring attitudes and orientations prevalent with a large population. The summated rating scale appears to be the most frequent type of scale employed in behavioral research (Babbie, 1983; Kerlinger, 1973) because it permits a straight forward method of index construction and allows for intensity of attitude expression. Such scales are considered to have high reliability because all subjects are presented with identical stimuli and the same scoring procedure. Unfortunately, these scales are also considered to be somewhat weaker in validity, due mainly to the 58 artificiality of the format. However, some researchers have contended that the concept of validity itself is rather ambiguous, for to say that some measure is valid or invalid assumes that there is a 'real" definition (Babbie, 1983; Kerlinger, 1973). Of the six available instruments utilized in reported studies to measure the attitudinal variables of concern in this investigation (Harres, 1968; McGee, 1956; Nixon et all... 1979; Snyder et al., 1975; Snyder & Kivlin, 1977; Woodford & Scott, 1982), only Harres and McGee cited reliability and validity information for their instruments. Beth of those instruments were considered to be much too lengthy for use in this investigation, however. McGee, for e3v-tample, used a 70 item scale and obtained a reSponse rate of only 34% with a mailed questionnaire, somewhat low from wh ich to make valid statistical observations or generalizations. Since the current investigator assumed th at the response rate of the sample contacted by telephone would be similar to one obtained by a mailed questionnaire, bitf‘evity without sacrificing quality was deemed to be ixTl‘lportant in the selection of the instrument used in this i1'1\restigation. The current investigator intended to locate an in strument which demonstrated an acceptable level of Va lidity and reliability and which had been used previously in a related study in an effort to contribute to the CO!) tinuity of research efforts in the area of attitudes 59 toward women in Sport. However, none of the existing instruments stood out. as the unequivocal best, nor were they comprehensive enough for the Scope of the current investigation. Therefore, the investigator ldevelOped an eclectic instrument for this investigation by selecting statements judged to be most pertinent to the current study frxmn those used by six investigative teams (Jensen et al., 1980; Nixon et al., 1979; Snyder & Kivlin, 1977; Spence & lie: lmreich, 1972; Spreitzer & Snyder, 1975; WOodford & Scott, 1982). Prior to conducting a pilot study of the instrument, the investigator experimented with different lengths of $1.1 bscales and formats for the questionnaire. As a result of this experimentation, the instrument used in the pilot Study consisted of the 30 items displayed in Table 3.1 and of questions pertaining to demographics. Ta ble 3.1 Composition of the Attitudinal Subscales and Their Sources \ \ S’x“-1.L=>scale/Items Source” At: i: itude toward sport 1— .. Competitive sports are not particularly important for the well-being of society. A 3 -— If more peOple were involved in competitive sports, we would not have so much trouble with drugs in our society. A 5 - Competitive sports are valuable because they help youngsters become good citizens. A 7‘- The emphasis that sports place on competition causes more harm than good. A 60 Table 3.1, continued Subscale/Items 10. Competitive sports are valuable because they teach youngsters respect for authority. 12. Competitive sports are valuable because they teach youngsters self-discipline. 15. Competitive sports are valuable because they contribute to the development of patriotism. 18. Competitive sports are valuable because they provide an opportunity for individuals to get ahead in the world. 22. Competitive Sports are valuable because they teach young people a sense of fair play and good sportsmanship. 30. Competitive Sports teach young persons to compete and thus prepare them for later participation in the business world. Attitudes toward women's role in society 6. 8. 13. 16. 17. 23. 24. 26. One of the most important things a mother can do for her daughters is prepare them for the duties of being a wife. It goes against nature to place women in positions of authority over men. A woman's personal ambitions should be sub- ordinated to her family (husband and children). Women Should compete on an equal basis with men for all jobs. The major responsibility of a wife is to keep her husband and children happy. Both husband and wife Should be allowed the same grounds for divorce. he modern woman is entitled to the same freedom from regulation and control that is given the modern man. Telling dirty jokes should be mostly a pre- rogative of men. Source* 61 Table 3.1, continued Subscale/Items Source* 28. 29. It is worse for a woman to be drunk than for a man to be drunk. D On the average, women should be considered as less capable of contributing to economic production than men. D Attitude toward women's participation in sport 2. 11. 14. 19. 20. 21. 25. 27. Opportunities for participation in organized sports should be restricted mostly or entirely to males. E Participation in organized sports only takes time away from other important things a woman ought to be doing. F Men seem naturally to know more about com- petitive sports than women. F Participation in competitive sports is more important for a male's physical and social development than for a female's development. E To excel in competitive sports, a woman has to be more like a man than a woman. F A woman can be both a good athlete and a truly feminine person. F WOmen ought to stick to cheerleading and leave participation in organized sports to men. E,F Women should be given more encouragement to participate in organized sports of all kinds. E Women must try extra hard to Show their femi- ninity if they participate in competitive sports. E Males and females should not be allowed to compete with each other in organized sports competition. E SOURCE KEY: Spreitzer & Snyder, 1975 Jensen et al., 1980 Snyder & Kivlin, 1977 Spence & Helmreich, 1973 Nixon et al., 1979 Woodford & Scott, 1983 WINDOW? 62 These statements comprised three subscales which were combined into one instrument with which to measure the three attitudinal variables under investigation. This instrument was entitled "The Women and Sport in Society Survey" (WASSS). Each of the three subscales which were embedded in the WASSS consisted of 10 statements which were randomly interspersed within the instrument (see Appendix A). Respondents were asked to (1) disagree strongly, (2) disagree mildly, (3) agree mildly, or (4) agree strongly with each of the 30 statements of the WASSS. For example, after being presented with the statement "sports are not particularly important for the well—being of our society," respondents chose one of the aforementioned responses. A ”neutral” or "no opinion" category was not included in the response choices because the investigator believed that forcing respondents to choose either a positive or negative response would result in the emergence of a more accurate reflection of true attitudes (Spence & Helmreich, 1972; Woodford & Scott, 1982). The forced choice of either agreement or disagreement with the statements of the WASSS, with only the intensity of the responses differing, seemed reasonable since a recent national survey of American attitudes toward sport (Miller Lite Report on Spprts, 1983) revealed that less than four percent of the pOpulation of the United States remained unaffected by sport in some manner. Similarly, it is difficult to conceive of an 63 individual who does not possess an attitude toward women's role in society. Thus, forcing respondents to choose from four response categories without a "no opinion" Option appeared to be reasonable. Some questions were scored in reverse order to fit the pattern of scoring. For example, agreement with the statement ”sports are not particularly important for the well-being of society” would indicate a negative attitude toward sport, whereas agreement with the statement ”Sports promote the development of fair play" would indicate a positive attitude toward sport. Therefore, the first statement was scored in reverse, with disagreement with the statement scoring higher than agreement. The interspersing of subscale statements and reverse ordering of some statements was done in order to minimize the possibility of response-set bias occurring to confound the results of the data (Kerlinger, 1973). Respondents were also asked to supply information pertaining to certain demographic variables. Gender, ethnicity, and educational status variables were utilized as the independent variables in this investigation. The Pilot Study Although the statements incorporated in the WASSS were drawn from previously published studies and should, therefore, have possessed an acceptable level of validity and reliability, this presumption was not substantiated by 64 any available information. Therefore, a pilot study was undertaken to determine the extent of the validity and reliability of the WASSS with the target population. An available sample of 150 students from an introductory psychology class and two evening physical education classes were surveyed for the pilot study. Seventy—one of these students were available two weeks later for the second administration of the questionnaire. Although there appears to be no general consensus concerning what constitutes an acceptable level of reliability for an instrument, some guidelines have been suggested. Carmines and Zeller (1979) suggested that most scales should have an alpha coefficient above .80, while Mehrens and Lehman (1973) suggested that for decisions involving groups rather than individuals, a reliability coefficient as low as .65 would be acceptable. Additionally, the length of an instrument has an effect on the extent of the reliability of the instrument, with short scales generally demonstrating lower reliability scores than longer scales, and must be taken into account when considering the extent of reliability demonstrated. Because the WASSS was designed to determine the attitudes of various groups of respondents and because each subscale consisted of only 10 items, a coefficient of .70 was selected as the minimum level of reliability acceptable for each subscale. 65 The extent of the reliability of the WASSS was examined from two perspectives. The extent of external consistency, or time-associated reliability, of the instrument was assessed by the test-retest method, while the extent of internal consistency was estimated by use of Cronbach's coefficient alpha. Although the test-retest procedure for estimating the extent of time related reliability is subject to the possible confounding effect of recall from one administration to the next, the investigator believed that this effect would be minimal with a two-week interval between administrations of the instrument, considering the nature of the variables under investigation. The test-retest procedure produced an overall Pearson coefficient of correlation of .79 (N=7l). The test-retest coefficients of correlation for the Attitude Toward Sport (AS), Attitude Toward Women's Role in Society (AW), and Attitude Toward Women's Participation in Sport (AWSP) subscales, were .76, .79, and .65, respectively. These results are depicted in Table 3.2. Cronbach's coefficient alpha, which was used to assess the extent of the internal consistency of the WASSS, has been recommended for use with multiple-item scales or for scales which are not scored dichotomously (Carmines & Zeller, 1979; Mehrens & ZLehman, 1973), such as in the present investigation. Coefficient alpha is a very general reliability coefficient encompassing both the Kuder- 66 Table 3.2 Reliability Analyses of the Women in Sport and Society Survey Internala Externalb Scale (N=150) (N=7l) Women in Sport and society survey (WASSS) .82 .79 Attitude toward sport (AS) .75 .76 Attitude toward women's role in society (AW) .70 .79 Attitude toward women's participation in Sport (AWSP) .79 .65 aestimated by Cronbach's coefficient alpha bestimated by Pearson's correlation coefficient Richardson 20 and Spearman-Brown Prophecy formulas for estimating internal consistency. Alpha is considered to be a conservative estimate of internal consistency where the actual scale reliability never falls below the alpha coefficient. The results of the internal reliability analysis Shown in Table 3.2 indicated that the alpha coefficients achieved for the 30 item WASSS and the AS, AW, and AWSP subscales were .82, .75, .70, and .79, respectively (N=150). These coefficients were all above the minimum level of alpha established for the WASSS, as were the test-retest correlation coefficients (N=7l) in all except one instance. Although one time related reliability coefficient did fall below .70, this was not considered crucial since the research design involved only one administration of the instrument and was not used to measure change. Therefore, 67 the instrument was judged to possess an acceptable level of reliability for further use in this investigation. Although the WASSS appeared to possess content validity, an exploratory factor analysis was undertaken in order to identify the basic structure underlying the items, since the statements comprising the WASSS had been drawn from a variety of sources. A principal components factor analysis with orthogonal rotation was conducted, following the guidelines of Kim and Mueller (1978). These authors suggested that the orthogonal solution be used in an exploratory factor analysis even if the factors are somewhat correlated, since the primary purpose of such an analysis is to simplify the data into theoretically meaningful constructs. The varimax criterion for rotation was chosen because it simplifies the columns of the factor matrix, thus presenting a clearer indication of the subscale groupings in the present investigation. A minimum loading of .40 was adopted as the criterion for a statement to be considered as contributing to a particular factor (Korth, 1975). The factor solution produced for the WASSS by the above procedure yielded seven factors which could be <:lear1y interpreted as facets of the three subscales. These factors accounted for 82.8% of the scale variance and are depicted in Table 3.3. Three other factors which contained an even mixture of items from the subscales and we re, therefore, uninterpretable accounted for the 6E3 .uoouo HcoHLoES: :H cars Lo:och mucwosoww wamomnpm SH peacomopo one mu0uomm .>o_ucfio mo oxzm czo Leo coucomouo up: our oc. Boaos mucwocca noboce L H. q. w.HN ~.m c.c <.m o.< m. oocmwnc> co czoonom mo. H~.m mm. Am. on. Ms. cc.m ozae>coawm ucoEooHo>oo HmHOOm me. etc accomxzo .woaoE Lou ucmupanw ouoE out monocm .AH 00. mewocm booed wuoE 30:: SHHmHSumc co: .m on. AuaSoCHEou 305m umsE uuocm CH :oEoB .mm 00. mupoom hmao :oE mommawmozu coEo3 .cm om. o:«c«Eow cam owuoaruc on cue ccEc3 .oH mm. Amooe: xfioscw coax ob Oufi3 no Suwawnwmcoomou Meme: .NH cm. >H«Eco Op mcofiuwose Ouecwpuocsm :oEc3 .mfi cc. ocoxou43 new mumuzmaco ounce»; muo:ucz .o A3o>u3m auoHoom Cw uuoom czc :cEcB ecu co ACOHDS—om Hmccuosuuov mHmSHmc< HODOmm m.m oHan 69 remainder of the variance. As shown in Table 3.3, factors 1, 7, and 10 together contained eight of the ten items comprising the AS subscale. Considered as a unit, these three factors accounted for 42.8% of the total variance of the WASSS. Factors 4 and 6 together contained half of the items for the AW subscale and accounted for 11.8% of the variance. Factors 2 and 5 combined to account for 28.2% of the total variance and together contained five of the ten items comprising the AWSP subscale. The fact that each subscale consisted of more than one factor indicated that each of the measured attitudes was multidimensional, a construct consistent with attitude theory (Dawes, 1972; Triandis, 1971). However, the combinations of factors did produce a clear factor structure which conformed to the composition of the three attitudinal subscales, with very little overlap between factors. Some factors were retained even though they failed to demonstrate an eigenvalue of 1.0 (Kim & Mueller, 1978) because they corresponded to one of the respective subscale constructs and also because the increase in size of the actual study sample could result in an increase in the amount of variance explained by these factors. Therefore, the WASSS was determined to possess an acceptable level of construct validity for further use in this investigation. .A correlational analysis of the three subscales indicated relatively little relationships between the 70 subscales. The AS subscale demonstrated almost no correlation with the AW (r=.00) subscale or with the AWSP (r=.23) subscale. The AW subscale, however, demonstrated a moderate correlation with the AWSP subscale (r=.64). This indicated that the measured attitudes were related, but judgment was reserved and the two subscales were not combined in the study since the factor structure was clearly delineated. Translation of the WASSS Many of the residents of the geographical area from which the data were collected did not speak English at all or had a very limited proficiency in the language, Spanish being their only or primary language. Therefore, in order to enhance the response rate of the sample, the WASSS was translated into Spanish, and the respondents were given the option of responding to the instrument in the language with which they were most comfortable. A member of the modern language department at Texas Southmost College translated the WASSS into the Spanish language. This individual was a native of Brownsville who had been an instructor at the college since 1973. She had had previous translating experience, having translated numerous business documents for local business persons dealing with with Mexican national businesses. The Spanish translation of the instrument was given to three local native speakers, an elementary school teacher, 71 a secretary, and a college student not included in any of the samples, all of whom translated the Spanish version back into English. The back translations were compared, and any differences which the back translators noted were considered. The instrument was then modified where necessary, although such modifications were minimal. Thus, the final Spanish version of the instrument was assumed to be parallel to the final English version (see Appendix B). Selection of the Sample Research in the area of attitudes toward women in sport has been criticized because it has generally focused on college student populations rather than on the general public (Nixon et al., 1979; Selby & Lewko, 1976; Woodford 8 Scott, 1982). Therefore, in an effort to address this concern, the investigator surveyed both a student sample and a sample of the general public for comparative purposes. This independent variable was named ”educational status." The sample of students and members of the general public was randomly selected from students attending Texas Southmost College and from the population of Brownsville, Texas. Brownsville is a city of approximately 90,000 residents situated in South Texas on the border with Mexico. The nature of this investigation dictated the availability of a large population of Mexican Americans, which is the case in Brownsville where they are the 72 majority pOpulation. The study also necessitated that both the Mexican-American and the comparative Anglo-American pOpulations have access to their cultural roots, again an occurrence in Brownsville. The farther away from the Mexican border the Mexican-American population lives, the greater would be the possibility of acculturation and assimilation into the dominant. .Anglo society, thus minimizing attitudinal differences. ApprOpriate sample size in survey research is generally determined by the Size of the research budget and the method of data collection (Babbie, 1983). Since each of the three independent variables in this investigation (gender, ethnicity, educational status) consisted of two categories, a total of eight (2 x 2 x 2) types of respondents or subsamples were needed. A subsample size of 100 was considered to be appropriate for this investigation, making the necessary total sample size approximately 800. The Student Sample A sample was drawn from the student population of Texas Southmost College, Brownsville, Texas, during the 1983-84 academic year. Texas Southmost College is a two- year community college located at the southernmost tip of Texas on the Mexican border. The college contains both academic and vocational-technical divisions and had a combined enrollment of 5591 students during the 1983-84 73 academic year. The ethnic composition of the college at the time of the collection of data for this investigation was 77% Mexican-American and 18% Anglo-American. Over 53% of the students were enrolled in academic transfer programs, while 42% were enrolled in two-year terminal degree programs (Registrar, Texas Southmost College, 1984). Since Texas Southmost College did not publish a student directory, students could not be sampled individually as was subsequently done with the general public. Instead, randomized cluster sampling was employed to draw a sample of physical education activity classes from a numbered list. As two semesters of physical education activity were required of all students enrolled in both transfer and terminal degree programs, students enrolled in those classes were assumed to be representative of the entire student population. Therefore, a random sample of physical education activity classes should have generated a representative sample of individual students comparable to an individual sampling of the general public. Consequently, a table of random numbers was used to obtain a starting sample of 18 physical education activity classes. Since the investigator needed a total of 400 students divided equally across the four subsamples and the average size of activity classes was 25 students, the investigator estimated that 18 classes were needed to generate the necessary number of respondents. When these 18 classes failed to yield sufficient numbers of students 74 to fill all subsamples, sampling continued in a similar manner until each of the groups was sufficient in size or until the entire subsample population was sampled. The General Public Sample The sample of the general public was drawn from the telephone directory of the city of Brownsville, Texas, during the spring of 1984. The ethnic composition of the city at the time of the collection of data for this investigation was approximately 84% Hispanic (including persons of Hispanic heritage other than MexicaneAmericans) and 15% Anglo-American (Brownsville City Planning Office, 1984.) In order to obtain a general public sample comparable to the size of the student sample, 400 respondents stratified by gender and ethnicity were needed. Therefore, a random sample of 804 names stratified by ethnicity according to surname was generated by drawing 12 names from each page of the 67-page telephone directory. The names were selected by randomly drawing six Spanish surnames from the page and then alternately choosing the first Anglo surname above and below the Spanish surname. In some cases, where nol.Anglo names were available on a page, succeeding pages were over-sampled until the necessary number of names was achieved. This procedure resulted in some error of classification, for surname is not always an 75 indication of ethnicity, but this error was calculable by the self-report of respondents' ethnicity. Collection of the Data All data were collected during the four-month period of April to July, 1984, by the investigator and four trained assistants. The Student Data The 18 sections of physical education activity classes selected with the use of a table of random numbers contained a wide variety of classes and at least one class of every instructor in the department. The investigator obtained permission from the instructor of each class selected to use part of one class period to administer the WASSS. Permission was granted for all classes except one. The investigator conducted the intitial administration in all classes, whereas at subsequent class meetings the instructors of the classes administered the survey instrument to class members who were absent during the initial administration. Students who had already completed the survey were instructed not to complete another questionnaire. This procedure ensured that no duplication of respondents would occur to confound the results of the data. The instructions read to every class prior to the administration of the WASSS explained the purpose of the study, informed the students that they could refuse to participate in the study, and assured complete anonymity 76 and confidentially to those who did choose to participate (see Appendix C). Respondents from 12 of the first 13 classes listed yielded the necessary numbers of Mexican-American female (N=118) and male (N=108) respondents, but not of Anglo- American female (N=12) and male (N=7) respondents. The instructor of the class selected ninth would not allow the survey to be given as She was behind schedule in that class and could not afford the time needed for the administration of the survey. The response rate for these 12 classes was 90%, with 246 respondents out of a possible 271. However, because only 19 of these respondents were Anglo-American, whereas 200 were needed, the investigator discontinued sampling MexicanaAmericans but continued sampling the remaining physical education classes until the required number of Anglo respondents was reached. Unfortunately, a sampling of all 38 sections of physical education activity classes yielded only 23 Anglo females and 22 Anglo males, far short of the number of Anglo respondents needed. Since the end of the spring semester had arrived, the sampling process was continued during the first summer session in college classes in departments other than physical education. Classes in the departments of business administration, biology, chemistry, English, mathematics, modern languages, and social sciences were sampled by having the instructors of classes ask only Anglo students to complete the questionnaire. This 77 procedure produced a sufficient quantity of female respondents (N=102) but not of male respondents. In a final attempt to procure a sufficient number of Anglo male respondents, an assistant who was doing telephone calling of the general public was assigned to go to the college student center and ask all Anglo males she could find who had not already done so, to complete the survey instrument. .A two-week period of this sampling still failed to produce the target number of 100, but all available respondents were probably contacted (N=72). Five of the Anglo-American students failed to meet the age criterion of 17 and older because they were high school students who were enrolled in summer college classes through a concurrent enrollment plan between the college and the local school district, so their data were not included in the final sample. 'Thus, the final student sample (N=395) was comprised of 118 Mexican-American females, 108 Mexican-American males, 98 Anglo-American females, and 71 Anglo-American males. The General Public Data Responses from the general public were collected via the telephone by four bilingual callers. Originally, the callers consisted of two females and one male who had been trained to elicit responses over the phone. However, during the data gathering process, one female caller quit after amassing a substantial pool of data. This data pool 78 was retained, and another female who had been a part of the original training process was hired to collect more data, which brought the total number of callers and subsequent caller groups to four. Although a mailed questionnaire may have been more amenable to gathering data from the large number of respondents needed, the possibility existed that many of the recipients of a written questionnaire would be unable to read it in either English or Spanish. The illiteracy rate of South Texas is reported to be very high; for example, in 1970 the average adult had attained less than seven years of formal education (Moore, 1976). A 1984 survey by the curriculum commission of the Brownsville Independent School District found that the dropout rate for high school students was 55% (Brownsville Herald, 1984). Therefore, contacting respondents by telephone seemed to be more efficient in this situation then a mailed questionnaire. Since the students and the general public responded to the same questionnaire and both samples had the opportunity to ask for clarifications of the items, it was assumed that the responses of the two samples were comparable, even though the techniques for collecting the data differed. The original telephone assistants, three bilingual college students, were trained in a three-step process by the investigator. First, they were given c0pies of the instrument in both English and Spanish to review and 79 clarify any questions about the instrument itself. General techniques for interviewing and telephone courtesy were discussed, and the callers were instructed to administer the questionnaire to each other over the phone. Next, the telephone assistants called the investigator's secretary and administered the survey to her with the investigator listening on an extension. Problems were noted and suggestions for improving the process were given. Finally, the assistants called five friends or relatives not included in the sample and administered the questionnaire to them with either the investigator or her secretary listening on an extension to make helpful suggestions. The participation of the investigator's secretary in the training process proved to be very fortuitous. Since one of the female callers quit during the course of the data collection, the secretary, also bilingual, was able to replace her. The procedures followed for training the telephone assistants not only standardized the data collection, but also enabled the assistants to achieve familiarity with the instrument and gain confidence in the survey administration process. All callers used the same introductory statement, in either English or Spanish, depending upon which language the respondents used to answer the phone (see Appendix D). As individuals were called, the assistants kept track of the gender of the respondent and attempted to equalize the number of females and males who responded to the survey. A 80 running total of respondents stratified by gender and ethnicity was kept in order to ensure that all categories were filled with the necessary numbers. As each respondent was called, the assistants kept a record of whether or not each respondent agreed to participate, needed to be called back, or was never reached. Telephone numbers which were discontinued or where no one answered after three attempts to reach them were replaced by resampling; those numbered 317. The calling was done at various times of the day from 8:00 am to 10:00 pm so as to maximize the variability of the respondents by including persons working outside the home as well as inside it, the retired, and the unemployed. The final sample size achieved for the general public was 411 respondents. The response rate of this sample was 43%, as 237 persons refused to participate. Of the 411 respondents, 397 met the age criterion, including 104 Mexican-American females, 102 Mexican-American males, 93 Anglo-American females, and 98 Anglo-American males. The sample size was considered to be adequate since it. was comparable in Size to the student sample and assumed to be representative of the general public in Brownsville. Summary A search of the literature failed to produce a suitable instrument with which to measure the attitudinal variables under investigation in this study, so the 81 investigator developed an eclectic one by drawing statements from instruments reported in related literature. The statements drawn from those instruments were combined into one instrument comprised of three subscales with which to measure the dependent variables of attitudes toward sport, toward women's role in society, and toward women's participation in sport. The instrument was named the Women and Sport in Society Survey (WASSS). A pilot study conducted with an available sample of college students (N=150) demonstrated that the WASSS possessed an acceptable level of reliability and validity for use in this investigation. The survey instrument was used to collect data from 400 college students and 411 members of the general public in a South Texas city during the spring and summer of 1984. The students responded to a written questionnaire, while the general public reSponded to a reading of the same questionnaire over the telephone by trained callers. The results of this data collection are presented in Chapter IV. CHAPTER IV RESULTS AND DISCUSSION The data which were collected by means of the procedures described in Chapter III were subjected to various statistical analyses. This chapter will encompass a presentation of the results of these analyses as well as a discussion of those results. For a summary table of the demographic information collected from respondents, see Appendix E. Preliminary Analyses Comparison of Caller Groups The initial analysis leading to the eventual interpretation of the data focused on an assessment of the responses to the four telephone callers employed to collect the data from the sample of the general public. Although the interviewing technique had been standardized and the subjects were randomly assigned to the telephone callers so as to minimize the caller effect, it was necessary to test for the presence of the effect because significant differences among the caller groups to the WASSS would make any subsequent differences detected across independent variables uninterpretable, since the source of such differences would be unknown. 82 83 In order to test the assumption of no difference in the responses of the four caller groups, labeled A, B, C, and D, a one-way analysis of variance (ANOVA) across callers was conducted for the WASSS. The results of the ANOVA across the four caller groups indicated a significant difference among the mean responses of the groups to the WASSS (F(3,393)=21.50, p .001). Thus, the null hypothesis of no difference among the group means was rejected. Since the results indicated a significant difference among the groups, but not which of the groups differed significantly from one another, a post hoc multiple comparison procedure employing Scheffe's test was conducted. The results of this analysis are presented in Figure 4.1. 9223 Leg a 2:922 .12 a .8. 9 85.41 71 D 86.63 175 A 90.26 97 B * 99.33 54 C * * * *p(.05 Figure 4.1. Scheffe test of differences among caller groups for the Women in Sport and Society Survey (WASSS) The investigator experimented with various combinations of caller groups to eliminate the caller effect. The results of this experimentation indicated that 84 the best way to eliminate the caller effect while minimizing data loss was to discard the data from groups B and C. The retention of the data from groups A and D resulted in the largest and most balanced sample across gender and ethnicity. A subsequent one-way ANOVA across caller groups A and D supported the assumption of no difference between the mean responses of the two groups to the WASSS (F(l,244)=.52, p) .47). Therefore, the final size of the sample of the general public which was utilized in this investigation was 246 respondents. Of these respondents, 66 were female Mexican-American, 61 were male Mexican- American, 52 were female Anglo-American and 67 were male Anglo-American. Comparison of the Randomly Selected and Available Anglo Student Subsamples All of the members of the subgroups in this investigation were obtained by random sampling procedures with the exception of the Anglo-American students. Although some of these subjects had been selected randomly (N=45), the majority of the sample (N=124) were part of an available sample for reasons delineated in Chapter III. This caused the investigator to be concerned about possible differences in response to the WASSS due to sampling error. However, the available sample of male Anglo-American students represented the entire pOpulation enrolled in the institution, so sampling error was not an issue. For the 85 female Anglo-American students, though, this issue was germane. Therefore, a t-test of the difference between the means of the Anglo-American female subsamples was conducted for the WASSS. The results of this analysis indicated that the responses of the two subsamples were homogeneous (t(96) = -.82, 2 > .41). Reliability and Validity of the WASSS Although the results of the pilOt study described in chapter III demonstrated that the WASSS possessed an acceptable level. of reliability' and validity for use in measuring the dependent variables under investigation, the data collected from the target population were subjected to the same analyses. This was done in order to ensure that the two methods used to collect the data were indeed comparable, since the pilot study had not included responses obtained by the telephoning process. In addition, the pilot study sample consisted only of students, not of members of the general public. Thus, the extent of the reliability and validity of the instrument with the actual sample population of the study had to be examined. The extent of the reliability of the WASSS was again examined by the use of Cronbach's alpha as the estimate of internal consistency. The test-retest measure of time- associated reliability was not conducted this time because 86 of the difficulty, and in most cases the impossibility, of surveying the respondents a second time. The reliability for the 30 statement instrument, presented in Table 4.1, produced alpha coefficients of .84 for the total instrument, and .73, .73, and .78 for the AS, AW and AWSP subscales, respectively. Since these coefficients met the ndnimum criterion of .70, the instrument was judged to possess an acceptable level of reliability for use in this investigation. In addition, since the alpha coefficients obtained with this sample were quite similar to those demonstrated in the pilot study, the WASSS was considered to be a stable instrument. Table 4.1 Reliability Coefficientsa by Total Sample and Independent Variable Groups Group WASSS AS AW’ AWSP Pilot Sample (N=150) .82 .76 .70 .79 Study Sample (N=641) .84 .73 .73 .78 Females (N=334) .84 .76 .71 .76 Males (N=307) .81 .70 .71 .75 Anglo-Americans (N=288) .83 .75 .72 .77 Mexican-American (N=353) .84 .71 .73 .79 Students (N=395) .81 .77 .72 .74 General Public (N=246) .82 .65 .69 .78 aCronbach's coefficient alpha Key: WASSS Women and Sport in Society Survey AS Attitude Toward Sport AW Attitude Toward Women's Role in Society AWSP Attitude Toward Women's Participation in Sport 87 A further assessment of the extent of the reliability of the instrument for each level of the independent variables was performed, with the results summarized in Table 4.1. Since only two of the 24 alphas fell below .70, and in those instances the divergences were small, the data collected via the WASSS were considered to be reliable for all subgroups of the investigation. The extent. of the validity' of. the WASSS was again examined by means of a principal components factor analysis. Although the results of the pilot study had indicated that the WASSS possessed an unambiguous factor structure, another exploratory rather than confirmatory factor analysis was conducted since a sample of the general public had not been included in the pilot study. Since the goal of this analysis was the same as that of the pilot study, the orthogonal solution with varimax rotation to a final solution was again employed. The results of this analysis are presented in Table 4.2. The orthogonal solution yielded five interpretable factors accounting for 96.4% of the scale variance. Factor 2 contained seven statements which loaded (above the .40 cut-off point, all of which were part of the AS subscale. This factor alone accounted for 28% of the scale variance. Factors 3, 4 and 5 contained eight of the ten statements from the AW subscale and combined accounted for 18.8% of the total variance. Factor 1 contained Six of the statements which comprised the AWSP subscale and accounted 88 .uopuo aooauoe=c c“ can» unsung .uocuosouo cannons. ouuuuucOIoc 0» adobe cs :— pOueoumuo and uncuoomn hug-nae no oxen day ago coueoaoum age One 0.. radon accupoo~ uOuoomo h.av v.n ~.o «.h o.o~ coca—uo> no vacuuom h«.o on. we. or. pa.n oauaaeommm no. “It. >-o0.30cooo amen ousa_uueou nose: .am no. nuuoou undo eon .oaOHuuoso c0303 .ou cc. col ox«~ on anal color .aOOxo OB .v~ on. unosmo~o>oo Aauooa pea aquaaazo .uo~oa you assumed!“ coon yucca .- Hc. uuuomu ozone anal soc: adamu=uoc co: .0 un. .3¢. :05 noon solo: uozu anusuoc uoz .o on. e030) uOu ocean» penance-a Iona aura clay noxau guano .v on. node! 0» oououuuQOL uuomu cu xuucauuomoo .N .amxac Duca- c. cosuua.o.uuaa ..cono: eucaou oeausuu< cm. can can» xenon on Ou eosoz uOu ouuoz .on mv. noxOn Sumac uuou pusozn cue aaeo .ow an. snag; »~.aou some a» no.2 s».~.n_ucoauo. .oflax .- mm. >~_aou cu acomuabaa cane—puonsu :oeoz .n~ Ho. oOO:Ou«3 uOu nuouzmzoo oLaQOLQ uuozuox .o mm. such men Heuucoo loom lacuna» .vn on. ocuo>ao non apesouo Anson .mu «v. .m<. m.zacanuuuomu can scan u_au cocoa .- cc. Amaze. uuuoau as. :. ouna.osu.om 0» camouaouco cueoz .- vv. .mmttv ocuCuIOH can oquumzua on coo c0203 .o~ ov. anon you :0! cu“) aanasgo euomeoo c0303 .o~ .34. >u0«00u cu o~0u n.:osoz canto» opsuuuut ouuox caucuses uou commend pea ouoozoo Ou coach .on . av. Queues-nanomu pea aoan nuau suave .- av. capo: as once. you as»: .od 5v. Io_uo.uuoo cocoa .mu «m. o=_~a_oa.euu~ou cocoa .«a co. suauozusa uOu acumen» cocoa .o~ an. nucucou_uuo coca some? .m .m¢. anode euazou unauLuu< u m c n bu uuSOEououm\aosoz uOuoom cue—once uOuoou Anna‘s. roausm aneuooo can uuomm e. ease: ecu no “rouge—om unnecesuuo. nau>~oc£ neuoom ~.v odnoe 89 for 53.3% of the scale variance. Although it was theoretically desirable for each of the subscale constructs to account for 33.3% of the scale variance if the factors were totally independent, the fact that the AWSP subscale construct accounted for a greater amount of the variance than the other two subscales was not considered to be critical since the major focus of this investigation was on the participation of women in sport., The statements which were not accounted for in their respective factor constructs either loaded on another subscale construct (See Tables 4.2 and 4.3) or did not load on any of the factors identified. Although the WASSS factored into five interpretable factors which could not only be combined into the three subscale constructs but also accounted for 94.6% of the scale variance, the fact that some subscale statements loaded on inapprOpriate subscale factors led to further analysis of the structure of the WASSS. An analysis of the correlation between the three subscales indicated that there was negligible correlation between the AS subscale and either the AW (£=.06) or the AWSP (£=.26) subscales. However, the AW and AWSP subscales demonstrated a fairly substantial (r=.70) correlation with each other. Therefore, the investigator conducted a further analysis of the factor structure of the WASSS by employing an oblique solution with direct oblimin rotation to a final solution. Several authors (Kendall, 1975; Kim & Mueller, 1978) have recommended that the oblique solution 90 be used when factors are known to be correlated. The direct oblimin rotation was chosen since it does not require the introduction of reference axes and allows the factors to be correlated if such a correlation exists in the data. The oblique solution, presented in Table 4.3, also yielded five interpretable factors accounting for 96.4% of the variance. The amount of variance accounted for by each of the factors and the identity of each factor as being comprised of the statements from a certain subscale were identical to the results yielded by the orthogonal solution. However, the structure of each factor became clearer as items inapprOpriate to the respective subscales were eliminated. Although the AW subscale factors did not account for a large amount of the variance of the WASSS, the subscale was retained since the constructs were clear and similar to those demonstrated by the instrument from which they were drawn (Spence & Helmrech, 1972). Thus, the investigator concluded that the WASSS possessed an acceptable level of construct validity for use in this investigation. Thus far in this chapter the presentation of results has consisted of analyses assessing the extent of the reliability and validity of the measuring instrument and the comparability of the data collected with different sampling procedures. Attention is now turned to an 91 .uopuo anouuoesc cu can» .uonuau .uoeuQSOLU oaaounsu canny-cocoa Ou nopuo ea c. ovucouoLc eon uncuooho auuuouo uo oxen or» San paueouuum yo: and ov. 30~on noeupuo~ Lagoon. h.ov v.m ~.o «.5 o.o~ cocauua> no u:00uom h« n on. we. on. no.~ usua>comuu . . canon. "Ha cu mucououuuem cu comausooeo c0803 .- nv. ucoemo~o>oc deuce. one nooduazm .uOHol Lon unnamed!“ 0.05 anomm .- ov. .3<. :03 noon coco: use» moneys: go: .o nv. :oeoa LOu coeds» aeouuoos_ eouu acre also ooxuu uuoam .v on no~QI o» payouuuoOL yucca ed quSSuuomoo .N .mmzto anon. cu Susana—oduuoo u.coaoz punsou oosuauu4 nu. col can» scape on o» :0503 ecu undo: .o~ an. noxom Sande duo» nascen con amco .oa om. ammo: adaaeu moo: Ou cud: >uuuunumcomu0u none: .nd om. madaau o» acouuubuo OuocLOLObsu coeoz .m~ no. coosouua new uncusosoo «uneven uuozuo: .o 5v. I soon non dauucoo IOuu £0006um .v~ nv. u ocuo>uc Log apesouo Anson .mn cv. I when new :03 cu.) Sausage ouomeoo :oaox .o~ Ast. hue—00¢ cu cuOu u.coeoz punJOu oosuauut vm. capo: unocdusn uou unoccum pea ouooeoo Ou cocoa .om av. mucosaSuuuoan can undo Luau noose .- «v. auauoauuoo some? .m~ mm. oeuum.on.ouu~ou suave .m— om. auuuonusa Lou UOOQuOL cones .o~ on. odsucou.u_o pogo cocoa .m .me. Leona etu30u oesuauu< a n v n bu coco-vueum\uoaaz Hobos» amcuoaoq uOuoah 0"",II'I'!‘ I I'll ill. I! It lane‘s. >o>usm Suwanee ecu uuoam :. coco: ago no .co.u:Hom osoa~no. nau>~uc¢ bosons m.v mange 92 analysis of the data in terms of the hypotheses presented in Chapter 1. Analysis in Terms of the Hypotheses Data were collected from 811 respondents, of whom 641 met all criteria established for this investigation (and whose data were therefore included in the statistical analyses. The investigator quantified the three attitudinal variables under investigation by summing the responses of the subjects to each of the attitudinal subscales and calculating the mean responses for each of the subscales. Scores could range from 10 to 40, and a score above the midpoint (25) of each subscale was assumed to indicate a favorable attitude toward the variable being measured, while a score below the scale midpoint was assumed to indicate an unfavorable attitude toward the variable being measured. The design of this study included three dependent variables (attitude toward sport, attitude toward women's role in society and attitude toward women's participation in sport) and three independent variables (gender, ethnicity and educational status) each with two» levels. Since the three dependent variable measures were obtained from the same subjects and therefore were presumed to be correlated, it was necessary to test for the combination of effects on these dependent variables (Hair et al., 1979). 93 This type of approach, which allows simultaneous testing of all variables and considers the various interrelationships among them, was necessary in order to control the inflation of Type I error which would be inherent in performing a series of univariate analyses in this instance (Barker & Barker, 1984; Cooley s. Lohnes, 1971; Hair et al., 1979; Huck et al., 1974). Therefore, a multivariate analysis of variance (MANOVA) was selected as the appropriate statistical technique. Since various test statistics are available to determine the significance of a MANOVA, and since the choice of the apprOpriate statistic depends upon the dispersion of the data, the investigator first determined whether the assumption of homogeneity of dispersion matrices associated with this procedure was met by the data (Hair et al., 1979). This assumption, tested by Box's M test, yielded a highly significant value (F(42,430242)=2.l3,p < .001), thereby indicating heterogeneity of dispersion matrices. The seriousness of violating this assumption is questionable, however, Since many contemporary statisticians have indicated that the MANOVA procedure is extremely robust, that is, unaffected by violations of this assumption (Cooley & Lohnes, 1971; Dowdy s. Weardon, 1983; Hair et al., 1979; Harris, 1975; Kerlinger, 1973; Winer, 1971). In view of these recommendations, the investigator concluded that it was apprOpriate to proceed with the multivariate analysis of 94 variance of the data collected in this investigation. However, the heterogeneity of dispersion matrices was taken into account in the choice of the appropriate test statistic for assessing the probability of attaining equal mean vectors. Due to the presence of heterogeneity' of (dispersion matrices, the Pillai-Bartlett trace criterion, V, was chosen to evaluate the equality of mean vectors rather than the more conventionally used Wilks' lambda statistic. Pillai's V is considered to be more sensitive than its counterparts in the presence of extreme heterogeneity of dispersion matrices and has been suggested for data such as those found in this investigation (Barcikowski, 1983; Olson 1976, 1979; Stevens, 1979). Since Pillai's V represents the amount of explained variance, a higher calculated value of V yields a lower p value, indicating increased significance. In the event that the results of the current investigation had been affected by the heterogeneity of dispersion matrices, the Pillai trace, V, could have been slightly inflated. However, since the V was overwhelmingly significant (p (.001), there appeared to be little chance of making a Type I error. Therefore, the results of the MANOVA were examined with the understanding that the significance level of the Pillai trace criterion might be somewhat higher than the value yielded, although not high 95 enough so as to fail to reject the null hypothesis of no difference. The results of the MANOVA, summarized in Table 4.4, shows that all interaction effects were non-significant (p<.05) as evaluated by Pillai's V. Each main effect, however, achieved a high level of significance (p (.001). The rejection of the null hypothesis of equality of mean vectors for the three main effects based upon the Pillai trace criterion indicated that each independent variable differed with respect to the dependent variables. It did not, however, indicate which of the three dependent variables were contributing to the significant differences detected. Therefore, a follow-up test was necessary to determine whether the AS, AW or AWSP subscales were contributing to the significant Pillai's V. In the absence of significant interaction. among the 'variables, the univariate analysis of variance (ANOVA) was deemed the appropriate procedure to further study the results (Barker & Barker, 1984; Hummel & Sligo, 1971; Morrison, 1976). The results of the ANOVA are presented in Table 4.5, followed by Table 4.6 which shows the means and standard deviations of the independent variable groups on the attitude subscales. Figure 4.2 illustrates the results of the .ANOVA in conjunction with the hypotheses stated in Chapter I. 96 Hmo.mnuou mwn. voH.H woo. magnum Honouumooom x aufiooczum x noocoo wov. vac. voo. . noumum Hmcooumosom x >ufioficsum omm. Non. Hoo. msumum Hmcooumosom x noocmo . com. ovo. ooo. . xuaoficcum x Loocoo ooo. mvo.o~. HNH. woumum Hmcooumooom Hoo. vom.m cNo. apocaccum ooo. www.mm va. noocoo a no cocooouucmdm mm oumefixouom< > coupe m.omadfim oocofium> mo ocuoom i msucuw accomucosom one aufiooccum .uoocoo anchor moamvmnsm new <>Oz¢:.uo SLMEEom v.v canoe 97 Table 4.5 Oneway ANOVA Summary Table for Gender, Ethnicity and Educational Status for Dependent Variables Significance Source 22 of F Attitude toward sport (AS) Gender .092 .762 Ethnicity 6.043 .014 Educational Status 17.394 .001 Attitude toward women's role in society (AW) Gender 48.003 .001 Ethnicity 9.618 .002 Educational Status 60.953 .001 Attitude toward women's participation in sport (AWSP) Gender 87.541 .001 Ethnicity 2.253 .134 Educational Status 68.810 .001 adf=1, 9E3 .lz..mmu** 8.3:. on.m oo oucoccum occ memo: o.q ofiomh 99 O._oaa _et~coo m:.mum _ ::_~zc_.¢t>:_ _ . t _ 7.3.. _=Lc:ec A 1.2.2:. _.o.t.:.::1- _ 2.5:.u 22.:— .r..:._.n._ aces—tc5< crowns: A Seats; woeo....:=< o_oc< cauucaonn- accuuou:om ocut9r25 .cc..xa.o:::_ 3. yucca .._ z... 3.2.—:8 Ere—.153. .2: .L EEO—.752 5.11:: Accra—Leaé c_xc< «EczAmo—eef _.s.._:c._:c- >._.._c.._um 2.3.59.3 - tomato .1 . y. . . .ioli-.-.'i-- Illlulnlh Soe_oom :. =Ssc3 co o_oL ecu o.¢3:. w.::=:.m e::__es tee w:e».;is< =.::< .mo_eaoo ho ve=:o_..e oz. :e;_ e>_..wc: was. >..:¢u_o_:c.w cue u__::= _~Ltccc 0;. tea «eeo.ue6< ecu_x&. .me_eE oo «ot=~.uua as» “N: 0. 33. .5256 A 356:3 oeuuoanam uoc- escrow ~¢couusoaom flex Ave-5...... outfit—3 . Leocec mzsouhcé< A o—&c< :aoaxoz emotoaasm so: - xuooocguu .‘ z T - I l..l|:.119 4 ('I. plll...ll ::_L=_ez c. <>oz< 0:9 He ¢o_:aoz ozu we :cuueocomeuo _: mcsodbos< «Lc.u_o ~.¢ cu:&_u me_=zrma~m39m tougcoé.’ I L.€:a co_ucoao«uuaa a c0503 pooJOu cutep=.c owe—ace can acooqhoe< O~wc< .mO—«Eou ac cep:u_eos 6;“ :e;. o>uu~noo cued >_ucoo«umcuon a»: u__r=o _euz:ex ecu van accouuoa< c-uuxox .eougE no noo:u_uuo eFr “n: :=.o2:m.¢c>:_ 1;. wooed» ~o:o_oou:om 2:: xo.omcz.e .Lot:o: «mecca x~o_ucw :_ «Leon .0 o_ot o;s stereo meosu_uue oz. :1 coerce.._c Leeuoo.:o_m c: m. ates» 100 Summary of Results Responses to a survey instrument designed to measure the respondents' attitudes toward the role of Sport in society, toward the role of women in society and toward women's participation in Sport were collected from a total sample of 811 respondents consisting of college students and members of the general public of Brownsville, Texas. Of these 811 responses, 641 met all criteria established for the study and were therefore utilized in the ensuing statistical analyses. Various analyses were conducted in order to determine the comparability of the data which were collected by different methods, as well as to determine the extent of the reliability and validity of the survey instrument with the sample population. The results of these analyses indicated that the data collected via the WASSS were comparable and that the instrument demonstrated an acceptable level of reliability and validity for measuring the dependent variables under investigation. A MANOVA, in which the dependent variables consisted of the attitudes toward sport, toward women's role in society and toward women's participation in sport and the independent variables were the categorical variables of gender, ethnicity and educational status, was used to test the three hypotheses guiding this investigation. The results of the MANOVA demonstrated significant main effect differences but no significant interaction effects. Since 101 there were no significant interactions, the investigator performed several univariate analyses of variance as a follow-up procedure to determine which of the dependent variables were contributing to the significant main effects. On the AS subscale, Mexican-Americans and college students exhibited significantly more positive attitudes toward the role of sport in society than did the Anglo Americans and members of the general public. This was contrary to the hypothesis of no difference in attitudes. There were no significant differences in attitudes toward sport demonstrated between females and males, as hypothesized. Significant differences on the AW subscale were observed across all three independent variables. Males, Mexican-Americans and members of the general public had a Significantly less liberal orientation toward the role of women in society than did females, Anglo-Americans and college students. This finding supported the second hypothesis of this investigation. On the AWSP subscale, males and members of the general public demonstrated a significantly less supportive attitude toward the participation of women in sport than did females and college students, as hypothesized. The attitudes of Anglo-Americans and Mexican-Americans did not differ significantly, however, contrary to the hypothesized difference. 102 Further Analyses and Discussion The purpose of this investigation was to explore the effects of gender, ethnicity and educational status on attitudes toward sport, toward women's role in society and toward women's participation in sport. Although these attitudes were found to differ across the three independent variables, the means of the three scaled attitudes were on the positive side (i.e., greater than 25) of their respective subscales. This indicated that the respondents held overall favorable attitudes toward sport, as well as toward women in society and in sport. For example, although women were significantly more suportive of women's participation in sport than were men, men also expressed positive attitudes toward the participation of women in sport. These results were not unexpected Since they were congruent with those found by other investigators who also reported positive attitudes of adults toward sport (Berlage, 1982; Griffith & Henschen, 1982; Grove & Dodder, 1979; Jensen et al., 1980; Miller Lite Report on Sports, 1983; Phillips, 1979; Spreitzer & Snyder, 1975); toward women's role in society (Braun & Chao, 1978; Cherlin & Walters, 1981; Loo & Logan, 1977; Lunneborg, 1974; Martin et al., 1980; Spence & Helmreich, 1972; Yoder, et al., 1982); and toward women's participation in sport (DeBacy et al., 1970; Kingsley et al., 1977; Harres, 1968; McCabe & Shaw, 1977; Miller Lite Report on Sports, 1983; Nixon et 103 al., 1979; Wittig, 1975; Woodford & Scott, 1982). Since an integrated discussion of these three attitudinal dimensions could be rather confusing, the discussion of the results of the current investigation will be divided into three sections, each one corresponding to one of the attitudinal subscales. Each section of the discussion will include the results of further analyses which were performed on the data to allow the investigator to explore dimensions of the Significant differences detected in the primary analyses of the data. These analyses included discriminant function analyses, discriminant classification analyses, factor analyses of the individual subscales and various analyses of additional demographic data. Attitude Toward Sport (AS) The results of the current investigation showed that these Americans believed that sport is an important component of contemporary society. This conclusion was based on the fact that ALL means of both the total subscale and of the individual subscale statements were on the positive side for all subgroups1 (see Tables 4.6, 4.7, 4.8, 4.9). Furthermore, only 25% of the respondents agreed with the statement that ”competitive sports are NOT particularly important for the well-being of society" (#1). The 1A statement mean score above 2.5 was considered to reflect a positive, or supportive attitude toward sport in society, since 2.5 was the midpoint for the individual subscale statements. 104 Table 4.7 Discriminant Analyaea of Subscale! by Gender Subscale/Statement M SD StandardlZvd — ”— Discriminant Wilks' Function Female Male Female Male Lambda Significance Coefficient Attitude toward sport (AS) a). Not important to uncloty 3.06 3.00 1.00 3.06 .999 .762 .049 3. Lessons societal dlug problems 2.79 2.50 1.02 1.16 .99] .010 - .63] a5. Teachcu quad cltlzunflhip 3.10 3.16 .95 .91 .999 .412 .162 7. CompetillOn does more harm than UOUJ 3.14 3.17 .8] .91 .999 .707 .055 10. Teaches respect for authority 3.12 3.l9 .91 .90 .999 .336 .301 12. Teaches self-discipline 3.47 3.47 .73 .83 1.000 .971 - .036 15. Teaches patriotism 2.70 2.06 .92 .90 .998 .297 .203 18. Helps get ahead in world 2.90 3.05 .93 .89 .993 .040 .546 22. Teaches fair play and sports- manuhip 3.46 3.32 75 .02 .993 .032 - .760 30. Teaches to compelc and plepalo for lnnalneun worl.) 3.06 3.ll .92 .(H. .999 .499 .141 Attitude tuwunl won-mu (AN) 66. Mothers prepare daughters for wifehood 2.70 2.3l 1.06 1.00 .965 .000 .215 a 8. Not natural that women boss men 3.35 2.00 .95 1.1] .951 .000 .375 13. Women subordinate amhItJOns to family 2.92 2.44 1.06 1.03 .950 .000 .347 16. Women compete equally wlth men a for jobs 3.14 3.00 1.00 .95 .999 .440 - .087 17. Major responsibility of wife to ‘ keep family happy 2.52 2.19 1.09 1.07 .977 .000 - .032 23. Equal gr0nnds for divorce 3.40 3.30 .91 .93 .999 .049 - .294 624. Freedom from control for both 3.37 3.08 .89 .97 .976 .000 .316 26. Only men should tell dirty a jokes 2.93 2.80 1.09 1.06 .997 135 - .160 28. Worse for women to be drunk than a men 2.7] 2.38 1.23 1.27 .98) .000 .083 29. Women contribute less economi- cally 3.44 2.96 .92 1.04 .943 .000 .492 Attitude toward women in Sport (AWSP) a2. Opportunity in sport restiicted a to males 3.67 3.40 .72 .07 .974 .000 .012 4. Sport takes time away [10m 6 important things for Home” 3.51 3.20 .80 .91 .969 .000 .114 9. Men naturally know more about a Sports 2.99 2.32 1.04 1.03 .906 .000 .585 11. Sports more important for males' physical and social development 3.37 2.09 .93 1.02 .943 .000 .207 14. To excel, wowan must be like man 3.37 2.90 .91 1.00 .959 .000 .126 19. Women can be athletic and feminine 3.68 3.40 .69 .07 .968 .000 .070 a 20. Women Cliccrlcad, men play Sports 3.67 3.28 .69 .09 .944 .000 .23l 21. Women encouraged to engage in a Sports 3.43 3.07 .02 .93 .957 .000 .142 25. Women in sport must Show a femininlly' 2.92 2.54 1.01 1.03 .967 .000 .132 27. Halcru .uud lunnileu lKJT tnimlwato against each other 2.0) 2.54 1.06 1.11 .905 .002 .004 a Statements which were revcruo scored Table 4.3 Discriminant 105 Analyse» of Subscales by lillmii'lly M 51) 8t anvlar :11 and -. _. Disc: iminanl 140): i can Anglo Merican Aug 1 0 1.111113 Fur-1‘1. inn Subscale/Statement American American American American Lambda Significance roelficienl Attitude toward Sport (AS) ‘1. Not important to Hlvvloly 1.11 1.112 1.60 .96 .909 .287 .112 .1 Lessons socielal 111m] l-lul-lu-mn 2.1111 2.56 1.14 1.01 .988 .006 .314 as. Teach-:8 (10011 (:il in-nship 1.10 1.16 1.112 .111 .999 . 192 - . 1"" 7 Con-petition tincn Iunlo‘ lnum linili 900d 3.24 3.115 .119 .115 .9119 .007 .101 10. Teaches respecl {m aulluuity 3 19 3.10 .94 .05 .997 .1114 .123 12. Teaches Self-discipline 3.47 3.40 .83 .72 .999 .814 - .265 15. Teacher! palriotisn 2.88 2.74 .911 .90 .995 .065 .196 18. "er9 get ahead in wui id 3.13 2.70 .91 .06 .963 .000 .751 22. Teaches fair play and npuiluumn— ship 3.43 3.15 79 .77 .997 .195 .109 10. Teaches tn C(1m|'t'14' ail-1 plvpnies for lnisineas Hnl141 3.03 3.16 '1'; .82 994 059 - .526 Attitude toward Hum-in (01") at". Mothers I‘lt‘p-Ut.’ 41.|11I]11|4‘lti IHI wifehomi 2.90 2.51 1.11 .9" 99') .671 - .211 1‘8. Not liatulal lhal wi-nu-n bus-o 1mm 1.07 1.19 1.1) .‘H. 997 .174 .101 d13. Hun-uh aul-udlnale muhitinnu to family 2.61. 2.73 1.11 1.01 999 422 - .126 16. Women con-pull: l'illld' 1y will: men a {m join-1 3 14 1.07 .on .9" 999 .392 - .247 17. Major iespmmil-illly ul U144! lo kin-Ii family happy 2.211 2,45 1.14 1.412 .994 .052 .07. 2 I. l-pml qlulnuln inn illvnleu 3. 311 1.411 .06 ,0] .999 .781 - .036 324. Fi‘cmhim luau I‘Hllllul inl luilh 1.10 1.211 .911 .1114 .991] .251 .054 .26. Only 1|4l'1l h11-II11I1 It‘ll Iiilfy 111L144 2.75 3.(Hl 1.15 .137 .9114! .003 .233 28. Horse iul Hlmn-n 101-u clam-1i than a men 2 11 2 11‘» 1.27 1.111 .957 .000 .828 29. Women (‘untl Hullu 1t'931 tuft-Inimi- caily 3 11 1. 12 1.0'» .94 997 020 .164 Attitude towni-l mum-n in fillnll (ANY-1') a2. Opportunity in spoil rot-1 I 1(‘11'41 a to malvu 3.47 3.63 .90 .67 .991 .015 .567 4. Sport takes film: away from a important lliimytl fur win-urn 3.32 3.41 .90 .82 .997 .193 .155 9. Men naturally know more about a sports 2.67 2.66 1.14 1.02 .999 .925 - 218 11. Sports more important fur males' a physical and uncinl development 3.13 3.15 1.04 .95 .999 .873 -' .212 14. To excell, wluuan must he like man 3.13 3.26 1.05 .86 .995 .079 .319 19. Women can be athlul ic and a feminine 3.58 3.5 .79 .79 .998 .259 - .541 20. Women cheerlead, nwn play sports 3.46 3.51 .81 .80 .999 .373 .101 21. Women encouraged to engage in sports 3 27 3.25 .96 .80 .999 .780 - .203 a25. Women in 9 out mun! show femininity‘ 2.65 2.05 1.07 .98 .991 .014 .507 ‘27. Males and fumnlva HUT compute against each other 2 64 2 71 1.14 1.01 .998 .258 .176 a . Statement: which weie reverse acurcd 106 Table 5.9 Dtecr-ninant Anaiyees of Subscales by Educational Statue .. £2 Starcard — Disc:::i General General Uilks' Functizn Subscale/Statement Student ?ublic S:;ce:: Public La:h¢a Significance Cce:fi: Attitude toward sport (AS) a. h) (u m . . e) \D (I) on N .000 .665 he Not imp rtan: :: society 3. Lessens societal drug e—‘O u 2.6 problems 2 9 2.68 1.07 l 12 .999 . 4L - 33. ‘5. Teaches good citi enship 3. 6 3.07 .91 97 996 .‘23 3 7 Cocoetition coes more harm than good 3.24 3.02 35 SS 985 .302 1:2 10. Teaches respect for authority 3.1L 3.16 .86 98 999 .347 - 247 12. Teaches self-disczpline 3.5# 3.52 .51 .73 998 2:5 - 315 15. Teaches pa:::o:;sa 2.79 2.57 .92 1.30 993 293 - 192 13. Help: 3e: heac 2: world 2.96 2.00 .90 .93 999 S99 - .240 Z . Teaches fair play and sports- - nanship 3.57 3 11 ‘4 .77 919 000 7L5 30. Teaches :3 :o::e:e and pre- pares 5c: buszness world 3.23 2 36 3‘ .93 .961 00C “‘ Attitude toward women (AV) ‘6. Mother: orepare daughters for , ‘- ‘ vifehood 59 2 22 '.93 1 32 953 DOC 256 3. So: natural that vcaen been _ . . ‘ zen 3.23 2 96 9 . ‘5 .9S~ 50‘ 3-5 “ 30cc: subordinate aaoitions to f :ilv -.56 2 ‘1 ' 1 1‘ .955 000 135 16. Vomen cocoete ecuall? ' th . ‘ :en for 30:: 3 " 3.37 . 31 .39 999 .- Z - 2L5 17 5:30: :esponeitilit7 :f wife to keep famil? ha:;v 2 5- 2.37 1.28 1.34 .955 .SC -159 33. Sc" 1 staunc: for :Lvote 3 «9 3.24 55 .97 .95~ 331 1'3 IL. Freedoa its: control for ‘ both 3.39 2.99 34 1.3» 937 JCC 29$ 25. Only :e: shouli tell #1::7 ‘ jokes 2.93 2.76 L :7 1 :8 994 352 - 150 23 gorse for wane: to be rur' ‘ than zen 30 2.19 1 22 1 24 9&6 SCC .45, 2?. ioaen cents: ute less econo- :zcally 3.32 3 3L 9 2 :6 :81 .30C 232 A::‘:uce tow :: were: i: so r: (A35?) ‘2 Oaportunztv 1: saczz :e- ‘ stricted :o :aLes 3.6L 3.33 76 .35 .976 .30C .365 4. 560:: :ases :Lze away from . i:oor:an: t;;:gs for women 3.50 3 ‘b 79 92 9E9 ,300 .22‘ 9 Ken natura..y gnaw :ore about a so r: 2.32 2 -2 1 )2 1 16 .965 33C “” ll. Soorts :ore i:oor:ant for sales' anvezca; as: social . a. develoogent 3.2 2 99 96 . 05 .966 903 - '9" .3. To exce;. u _ 1“ man oven ‘ust be exe3.32 2 97 91 ‘ 03 .970 300 3-: 19 women can be athletic and ‘ feminine 3 70 3 29 .68, 39 .936 000 .25“ 23. women cheerlead. :en play he 0‘ O u u C) ‘5 U) -O 0 o m C) O O D (J N (4 sports 21. women encouraged to engage ‘ in soorts 25. Bowen 1n sport aunt show ‘ fenxninity 2.82 2.61 27. Sales end females NOT compete agaxnst eecn other ‘ ’ .‘8 .92 .880 .000 .671 .99l .015 - .076 .09 1.06 .976 .300 .162 U M 0 r4 r~ - o e F r- O \J 'x H D) .46 a- statement: which were reverse scored 107 importance accorded Sport in this investigation was approximately the same as that reported in a study of adults conducted a decade ago in the midwest (Spreitzer & Snyder, 1975). The similarity of results in these two investigations indicated that favorable attitudes toward Sport have remained fairly constant over the past decade and are not bounded by geography. Although the respondents in this investigation appeared to believe that sport is important to society, significant differences were detected in the degree of support across the two independent variables of ethnicity and educational status. Therefore, the investigator conducted further analyses to explore these differences. The results of the initial factor analysis of the WASSS had indicated that the attitude measured by the AS subscale was unidimensional, since only one factor construct had emerged in this analysis. Therefore, the AS subscale was assumed to consist of only one dimension and, consequently, was not subjected to a separate factor analysis. Since no significant difference was detected between the attitudes of females and males toward sport, and since this result was consistent with the reports of other investigators (Grove & Dodder, 1979; Jensen et al., 1980; Spreitzer & Snyder, 1975), the variable of gender with respect to the AS subscale was not further explored. However, since the results of the initial 2 x 2 x 2 (gender, ethnicity, educational status) ANOVA had 108 demonstrated significant differences in attitudes toward Sport across ethnicity and educational status, the investigator used the results of a discriminant function analysis of the 10 statements which comprised this subscale to determine where those differences existed. The results of that initial ANOVA had revealed that educational status accounted for 2.5% of the AS subscale variability, which, although this was relatively little of the subscale variability, was more of the explained variability than that accounted for by either gender or ethnicity. Educational status appeared to be of greater salience in explaining the attitudes of the students than of the members of the general public, because the results of a discriminant classification analysis showed that 64% of the students but only 56% of the general public were correctly classified by their scores on the AS subscale. The students held Significantly more positive views toward Sport than did the members of the general public on all four of the subscale statements which were found to discriminate between the two status groups (#1, #7, #22, #30). However, as stated previously, all mean item Scores for both groups were on the positive Side of the cut-off point (see Table 4.9). The positive attitudes toward Sport expressed across educational Status in this investigation were expected, since they were consonant with the findings of other investigations among samples of both college students and members of the general public (Berlage, 1982; 109 Grove & Dodder, 1979; Jensen et al., 1980; Nixon et al., 1979; Spreitzer & Snyder, 1975). Since the independent variables of this investigation had explained so little of the variability in the AS subscale, the investigator examined other demographic information which had been collected from the respondents. A correlational analysis was conducted between the dempographic information and the three attitudinal subscales (see Table 4.10) in order to determine if any other variable correlated more highly with the respective subscale than did gender, ethnicity or educational status. An examination of the results of this analysis indicated that no other variable correlated with the AS subscale more highly than did educational status, and only current participation approximated the same level of correlation as educational status. It thus appeared that educational status was the most salient variable for explaining attitudes toward Sport, although current participation also appeared to be involved. Since current participation appeared to be related to attitudes toward Sport, the investigator performed a one- way ANOVA on this variable across educational status. The results indicated that students currenly participated in Sport Significantly more than did members of the general public (F(1,639)=32.82,p< .001). This difference in degree of current participation in sport may have been partially responsible for the more positive attitudes toward Sport 110 9.. e_. e_.. o~.r cc . n—. e_. c~.. a... _a c~. . , 5.. 2..- .2; .= .3: 32...; z. c..:....., 15...»..2. 3.5.: 3:... 232;; 3 Qatari—.3... _.......__ ..._:.- ......._.¢... f... z— _:...._...... .ux ... 3....I... $.33. ,5 :1 me;: f V. .:._.:;_;_.n:; .:._.:;..__.:L . ‘1:;u ‘35; cf. .2: at - c. .1... :. .:........ .339. I...’ .I..-o.’ a: .0... . v: — b o.... .n —:.- u... «a 4:... .c. ...—. Q .9. .of" o.:;.4.._ozu =aouo_...3u =3..‘saa :z..!x..otr:_ 3.5. a: 00—5....) .605Claufi:_ ; 10.3: u. .;u.- n.:a_a..—ozu ::.u1.o..au :vI—aozao .~ . I..-'—md. .3. :— cl ...... .lma<. ..o¢. :_ :3.uso_u.u.aa n .c?-.;) ’6 . On. Quclbu 09:..uu( .:<. .....o. a. 0—0— 0.60i2) ~_ o~. v-oao. ou:....< Aw.H Nm. ms.H mm. m~.H m~.~ mc.m cs.H mo.m cm a wwmmmm mb. Ho.N Nm. Hm.H mm. m~.H mm. hN.H om. vb.H mm. NM.H Nb. N~.~ mm. mm.v an m mmw::sm mDB.H comumamofiuomm occussu Hm. mm.H ov. c¢.~ mm. Ob.~ Hm. vb.H muuomm cm cmummmufiuumm uw>mz he. mm.~ cm. cv.~ He. mm.~ om. mm.H muuoam amsom> [wocH Amcomumwsowm c“ cowuum_owuomm om. o«.~ mv. cw.~ cc. we.~ Hm. Oe.fi muuoom Esme Hocowumwuowm cm coflummwowucmm am. ~m.H mm. em.H om. as.~ om. ma.~ momssoo cs cosuomwowuomm uuoam as. mm.H cm. 44.3 mm. o..~ om. ~¢.H Hoocom now: :1 co_unawuwu.om uuoom c~.~ we." we. =V.H so. qv.~ .c._ ~c.. magnum Hau_.mz -.H cc.e mH.H so.» a~.~ 51.4 c~.~ k~.v columoscm Lo mama» cm a mm x mm r em a zD_v_:zbm .Locccc mmoLo¢ mucoccoamw: oz» LC mu.ummumum v>_uLwLUmo: REFERENCES REFERENCE S Ajzen, I., & Fishbein, M. (1979). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, _8_4, 888- 918. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall. Anderson, J. G., & Johnson, W. H. (1979). Stability and change among three generations Of Mexican-Americans: Factors affecting achievement. American Educational Research Journal, 8, 285-309. Babbie, E. R. (1983). The practice of social research (3rd ed.). Belmont, CA: Wadsworth. Baca Zinn, M. (1982a). Chicano men and masculinity. The Journal of Ethnic Studies, 19, 29-44. Baca Zinn, M. (1982b). Mexican-American women in the social sciences. Signs: Journal of Women in Culture and Society, 8, 259-272. Barcikowski, R. S. (1983). Computer packages and research designs. New York: University Press of America. Barker, H. R., & Barker, B. M. (1984). Multivariate analysis of variance (MANOVA) . Alabama: The University Of Alabama Press. Bell, M. M. (1980). Role conflict of women as athletes in the United States. ARENA Review, 4, 22-31. Berlage, G. I. (1982). Are children's competitive team Sports socializing agents for corporate America? In A. O. Dunleavy, A. W. Miracle, & C. R. Rees, Studies in the sociology of sport. Fort Worth: Texas Christian University Press. Birrell, S. (1983). The psychological dimensions of female athletic participation. In M. A. Boutilier & L. SanGiovanni, The sporting woman. Champaign, IL: Human Kinetics. 178 179 Bonjean, C. M., Hill, R. J., & McLemore, S. D. (1967). Sociological measurement. San Francisco: Chandler. BOSlOOper, T., & Hayes, M. (1973). The femininity game. New York: Stein & Day. Boutilier, M. A., & SanGiovanni, L. (1983). The sporting woman. Champaign, IL: Human Kinetics. Braun, J. S., & Chao, H. M. (1978). Attitudes toward women: A comparison of Asian-born Chinese and American Caucasians. Psychology of Women Qkuarterly, 8, 195-201. Brownsville City Planning Office. (1983). Personal communication. Brownsville Herald. (1984, July 17). Education not a top priority. Bueter, R. J. (1972). Sports, values and society. The Christian Century, 88, 389-392. Buzan, B. D., & Phillips, D. B. (1980). Institutional completeness and Chicano militancy. Aztlan, _]._l_, 33- 64. . Carmines, E. G., & Zeller, R. A. (1979). Reliability and validity assessment. Beverly Hills: Sage. Catholic Diocese of Brownsville. (1985). Personal call. Chafetz, J. S. (1978). Masculinequeminine or human? An overview Of the sociology of gender roles. Itasca, IL: F. E. Peacock. Chandler, C. R. (1979). Traditionalism in a modern setting: A comparison of Anglo and Mexican-American value orientations. Human Organization, 88, 153-159. Cherlin, A., & Walters, P. B. (1981). Trends in United States men's and women's sex-role attitudes: 1972- 1978. American Sociological Review, 88, 453-460. Clark, B. R“ (1962). Educating the expert society. San Francisco: Chandler. Coakley, J. J. (1982). Sport in societj (2nd ed.). St. Louis: C.V. Mosby. Coleman, J. S. (1961). The adolescent society. New York: Free Press. 180 Cooley, W., & Lohnes, P. (1971). Multivariate data analysis. New York: John Wiley & Sons. Curry, T. J., 5. Jiobu, R. M. (1984). Sports, a social perspective. Englewood Cliffs, NJ: Prentice-Hall. Dawes, R. M. (1972). Fundamentals of attitude measurement. New York: John Wiley & Sons. DeBacy, B. R., Spaeth, R., & Busch, R. (1970). What do men really think about athletic competition for women? Journal of Healthy Physical Education and Recreation, 88, 28-29. Donelson, E., & Gullahorn, J. E. (1977). Women: A psychological perspective. New York: John Wiley & Sons. Dowdy, S., & Weardon, S. (1983). Statistics for research. New York: John Wiley & Sons. Dowling, C. (1981). The Cinderella complex. New York: Pocket Books. Dyer, K. F. (1982). Challenging the men. St. Lucia, Queensland: University of Queensland Press. Eitzen, D.S. (1975). Athletics in the status system of male adolescents. Adolescence, 88, 268-276. Eitzen, D. S. (1982). In conflict and order: Understanding society. Boston: Allyn-Bacon. Eitzen, D. S., & Sage, G. H. (1982). Sociology of American sport (2nd ed.) . Dubuque, IA: William C. Brown. Espinosa, R. W., Fernandez, C., & Dornbusch, S. M. (1977). Chicano perceptions of high school and Chicano performance. Aztlan, 8, 133-155. Evans, F. B., & Anderson, J. G. (1973). The psychocultural origins of achievement and achievement motivation: The Mexican-American family. Sociology Of Education, 88, 396-416. Federico, R. C., & Swartz, J. S. (1983). Sociology. Reading, MA: Addison-wesley. Fine-Davis, M. (1983). A society in transition: Structure and determinants of attitudes toward the role and status of women in Ireland. Psychology of Women Quarterly, 8, 113-132. 181 Fisher, A. C., Genovese, P. P., Morris, K. J., 8 Morris, H. H. (1978). Perceptions of women in Sport. In D.M. Landers 8 R. W. Cristina (Eds.), Psychology of motor behavior and spgrt. Champaign: Human Kinetics. Flora, C. B. (1971). The passive female: Her comparative image by class and culture in women's magazine fiction. Journal of Marriage and the Family, 88, 435- 444. Garcia, P., 8 Maldonado, L. A. (1982). America's Mexicans: A plea for specificity. The Social Science Journal, L2, 9-24. Gecas, VK (1980). Family and social structure influence on the career orientation of rural Mexican-American youth. Rural Sociology, 88, 272-289. Gerber, E. W., Felshin, J., Berlin, P., 8 Wyrick, W. (1974). The American woman in sport. Reading, MA: Addison-Wesley. Gilbert, B., 8 Williamson, N. (1973, May 28). Sport is unfair to women. Sports Illustrated. 88, 88-92+. Gonzales, S. (1979). The Chicano perspective: A design for self-awareness. In A. D. Trejo (Ed.), The Chicanos. Tucson, AZ: The University of Arizona Press. Gonzales, S. (1977). The white feminist movement: The Chicana perspective. The Social Science Journal, 88, 67—76. Grebler, L., Moore, J. W., 8 Guzman, R. C., (1970). The Mexican-American people: The nation's second largest minority. New York: Free Press. Griffith, L., 8 Henschen, K. (1982). Attitudes and values associated with Little League football competition. Cited in D. S. Eitzen 8 G. Sage, Sociology of American sport (2nd ed.). Dubuque, IA: William C. Brown. Grove, S. J., 8 Dodder, R. A. (1979). A study of functions of Sport: A subsequent test of Spreitzer and Snyder's research. Journal of Sport Behavior, 8, 83-91. Guinn, R. value clarification in the bicultural classroom. Journal of Teacher Education, 88, 46-47. Hair, J. F., Anderson, R. E., Tatham, R. L., 8 Grablowsky, M. (1979). Multivariate data analysis with readings. Tulsa, OK: PPC Books. 182 Hall, M. A. (1977). The sociological perspective Of females in sport. NAGWS Research Reports, III. Washington, DC: AAHPER. Hare-Mustin, R. T., 8 Broderick, P. C. (1979). The myth of motherhood: A study of att i tudes toward motherhood . Psychology of Women Quarterly, 8, 114-128. Harragan, B. (1977). Games mother never taught you: Corporate gamesmanship for women. New York: Rawson. Harres, B. (1968). Attitudes of students toward women's athletic competition. Research Quarterly, 88, 278- 284. Harris, D. V. (1980). Femininity and athleticism: Conflict or consonance? In D. F. Sabo 8 R. Runfola (Eds.), Jock: Smarts and male identity. Englewood Cliffs, NJ: Prentice-Hall. Harris, R. J. (1975). A primer of multivariate statistics. New York: Academic PreSS. Hawkes, G. R., 8 Taylor, M. (1975). Power structure in Mexican 8 Mexican-American farm labor families. Journal of Marriage and the Family, 81, 807-811. Hebding, D., 8 Glick, L. (1976). Introduction to sociology. Reading, MA: Addison-Wesley. Heller, C. S. (1976). Mexican-American youth: Forgotten youth at the crossroads. New York: Random House. Heller, C. S. (1971). New converts to the American dream? Mobility aspirations of young Mexican-Americans. New Haven, CT: College 8 University Press. Hernandez, M. A. (1977). A comparative study Of attitudes toward athletics as expressed by Mexican-American and non-Mexican-American high school boys. Eugene: Microform. Hesselbart, S. (1976). A comparison of attitudes toward women and attitudes toward Blacks in a Southern city. Sociological Symposium, 81, 45-68. Hoferek, M. J., 8 Hanick, P. L. (1980). Woman and athlete: Toward role consistency. In J. Thomas (Ed.). Abstracts: 1980 AAHPERD COnvention. Washington D.C.: AAHPERD. 183 Huck, S. W., Cormier, W. H., 8 Bounds, Jr., W. G. (1974). Reading statistics and research. New York: Harper 8 Row. Hummel, T. J., 8 Sligo, R. J. (1971). Empirical comparisons of univariate and multivariate analysis Of variance procedures. Psychological Bulletin, _7_6_, 49- 57. Jensen, T. M., Leonard, W; M., II, 8 iLiverman, R. I). (1980). College students attitudes toward intercollegiate athletics. Paper presented at the first annual conference Of the NOrth American Society Of Sport Sociology. Denver, CO: October 16—19. Kahn, A. (1984). The power war: Male response to power loss under equity. Psychology of WOmen anrterly, 8, 234-247. Kane, M. J. (1983). Title IX notwithstanding, girls expected to 'glow'. The Physician and Sportsmedicine. 88, 40-41. Karnig, A. K., 8 Welch, S. (1979). Sex and ethnic differences in municipal representation. Social Science Qparterly, 88, 465-481. Kendall, M. (1980). Multivariate analyses. New York: MacMillan. Kerlinger, F. (1973). Foundations of behavioral research (2nd ed.). New York: Holt, Rinehart 8 Winston. Kiesler, C.A. (1969). Attitude change: A critical analysis of theoretical approaches. New York: John Wiley 8 Sons. Kingsley, J. L., Brown, F. L., 8 Seibert, M. E. (1977). Social acceptance of female athletes by college women. Research Quarterly, 88, 727-733. Kim, J-O., 8 Mueller, C. W. (1978). Factor analysis. Beverly Hills: Sage. Kim, J-O., 8 Mueller, C. W. (1978). Introduction to factor analysis. Beverly Hills: Sage. Kluckhohn, F. L., 8 Strodbeck, F. L. (1961). Variations in value orientations. New York: Row, Peterson. Kniker, C. R. (1974). The values of athletics in schools: A continuing debate. Phi Delta Kappan, 88, 116-120. 184 Korth, B. (1975). Rotations in exploratory factor analysis. In D. J. Amick 8 H. J. Walberg, (Eds.). Introductory multivariate analysis. Berkeley: McCutchan. Kuvlesky, W. P., 8 Juarez, R. Z. (1975). M-A youth and the American dream. In J. S. Picou 8 R. E. Campbell (Eds.). Career behavior of special groups. Columbus: C. F. Merrill. Leonard, W. M. (1984). A sociological perspective of Sport (2nd ed.). Minneapolis: Burgess. Lever, J. (1978). Sex differences in the complexity of children's play and games. American Sociological Review, 88, 471-481. Lewko, J. H., 8 Greendorfer, S. L. (1978). Family influence and sex differences in chi ldren' S socialization into sport: A review. In D. M. Landers, 8 R. W. Cristina, (Eds.). Psychology of motor behavior and sports. Champaign: Human Kinetics. Longeaux y Vasquez, E. (1970). The Mexican-American woman. In E. Sutherland (Ed.), Sisterhood is powerful. New York: Random House. LOO, R., 8 Logan, P. (1977). Investigation of the attitudes toward women scale in western Canada. Canadian Journal of Behavioral Science, 8, 201-204. Lunneborg, P. (1974). Validity of attitudes toward women scale. Psyghological Reports, 88, 1281-1282. Madsen, W. (1964). The Mexican Americans of South Texas. New York: Holt, Rinehart 8 Winston. Martens, R., 8 Seefeldt, V. D. (1979). Guidelines for children's sports. Washington, DC: AAHPERD. Martin, P. Y., Osmond, M. W., Hesselbart, S., 8 Wood, M. (1980). The significance of gender as a social and demographic correlate of sex role attitudes. Sociological Focus, 88, 383-396. Mason, K. O., 8 Bumpass, L. L. (1975). U. 8. women's sex- role ideology, 1970. .American Journal of Sociology, 88, 1212-1219. McConville, B. (1985). Women under the influence. New York: Schocken Books. 185 McGee, R. (1956). Comparisons of attitudes toward intensive competition for high school girls. Research McMillen, J. B. (1981), Social organization of leisure among Mexican-Americans in Houston, Texas. Abstracts: 1981 AAHPERD Convention. Washington, D.C.: AAHPERD. McPherson, B. D. (1979). Past, present and future perspectives for research in Sport sociology. In A. Yiannakis, et al., Sport sociology: Contemporary themes (2nd ed.). Dubuque: Kendall/Hunt. Mehrens, M. A., 8 Lehman, I. J. (1973). Measurement and evaluation in education and psychology. New York: Holt, Rinehart 8 Winston. Miller Brewing Company. (1983). Miller lite report on American attitudes toward sports 1983. New York: Research 8 Forecasts. Mirande,.A. (1979). A.rednterpretation of male dominance in the Chicano family. The Family Coordinator, 88, 473-479. Mirande, A., 8 Enriquez, E. (1976). Chicanas: Their triple oppression as colonized women. Sociological Symposium, 81, 91-102. Mirande, A., 8 Enriquez, E. (1979). La Chicana. Chicago: The University Of Chicago Press. Moerk, E. (1972). The acculturation of the Mexican- American minority to the Anglo-American society in the U.S. Journal of Human Relations, 88, 317-325. Montiel, M. (1973). The social science myth of the Mexican-American family. In 0. I. Romano-V(Ed.), Voices. Berkeley: Quinto Sol. Montenegro, M. (1976). Chicanos and Mexican-Americans: Ethnic self-identification and attitudinal differences. San Francisco: R 8 E Research Associates. Moore, J. (1976). Mexican Americans (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall. Morrison, D. F. (1976). Multivariate statistical methods. New York: McGraw-Hill. 186 Nicholson, L. J. (1983). The ethics of gender discrimination. In B. C. Postow (Ed.), Women, philOSOJhy, and Sport: A collection Of new essays. Metuchen, NJ: Scarecrow Press. Nielsen, L. (1983). Putting away the pom-pom: An educational psychologist's view of females and sports. In B. C. Postow (Ed.), Women, philosophy, and sport: A collection Of new essays. Metuchen, NJ: Scarecrow Press. Nieto, C. (1974). The Chicana and the women's rights movement--a perspective. Civil Rights Digest, 8, 485- 490. Nixon, H. L., II. (1979). Acceptance Of the 'dominant American sports creed' among college students. Review of Sport and Leisure, 8, 141-59. Nixon, H. L., II. (1984). Sport and the American dream. New York: Leisure Press. Nixon, H. L., 11, Maresca, P. J., 8 Silverman, M. A. (1979). Sex differences in college students: Acceptance Of females in Sport. Adolescence, 88, 755- 764. Ogilvie, B., 8 Tutko, T. (1971). Sport: If you want to build character, try something else. Psycholggy Today, 8, 60-63. Oglesby, C. A. (1978). Women and Sport: From myth to reality. Philadelphia: Lea 8 Febiger. Olson, C. L. (1976). On choosing a test statistic in multivariate analysis of variance. Psychological Olson, C. L. (1979). Practical considerations in choosing a MANOVA test statistic: a rejoinder to Stevens. Psychological Bulletin, 88, 1350-1352. arelius, A. P. (1975). Emerging sex-role attitudes, expectations and strains among college women. Journal of Marriage and the Family, 81, 146-153. Penalosa, F. (1970). Toward an Operational definition Of the Mexican-American. Aztlan, 8, l-12. Phillips, J. H. (1980). A comparison of parental attitudes toward competition in ,youth sports in relation to the sex of the participant and the selected sport. Eugene: Microform. 187 Promoli, F., McCabe, J., 8 Shaw, S. (1977). Attitudes towards individuals participating in sex stereotyped sports. Paper presented at the Ninth annual symposium of the psychomotor learning and Sport psychology committee. Banff, Canada. Quarm, D. (1983). The effect of gender on sex-role attitudes. Sociological Focus, 16, 285-303. Ramirez, M., III, Taylor, C., 8 Petersen, B. (1971). Mexican-American cultural membership and adjustment to school. Developmental Psychology, 8, 141-148. Registrar, Texas Southmost College. (1984). Personal communication. Rohrbaugh, J. B. (1979). Women: Psychology's puzzle. New York: Basic Books. Romano-V, O. I. (1973). Voices. Berkeley: Quinto Sol. Rosaldo, M. A. (1980). The use and abuse Of anthropology: Reflections on feminism and cross-cultural understanding. 8ignsi Journal of Women in Culture and Society, 8, 389-417. Rubin, L. B. (1976). Worlds of pain: Life in the working class family. New York: Basic Books. Sage, G., 8 Loudermilk, S. (1979). The female athlete and role conflict. Research Quarterly, 88, 88-96. Schwartz, A. J. (1971). A comparative study of values and achievement: Mexican-American and Anglo youth. Sociology of Education, 88, 438-462. - Scimecca, J. A. (1980). Education and society. New York: Holt, Rinehart and Winston. Selby, R., 8 Lewko, (I. (1976). Children's attitudes toward females in sports: Their relationships with sex, grade and Sports participation. Research Quarterly, 81, 453-463. Sheriff, M. (1971). Girls compete? In DGWS research reports: Women in gport. Washington, DC: AAHPER. Snyder, E. E., 8 Kivlin, J. (1977). Perceptions of the sex role among female athletes and non-athletes. Adolescence, 81, 23-29. 188 Snyder, E. E., Kivlin, J. E. 8 Spreitzer, E. E. (1975). The female athlete: .An analysis of objective and subjective role conflict. In D. M. Landers, (Ed.). Psychology of sport and motor behavior 11. Pennsylvania State University. Snyder, E. E., 8 Spreitzer, E. A. (1976). Correlates of sport participation among adolescent girls. Research Quarterly, 81, 804-809. Snyder, E. E., 8 Spreitzer, E. A. (1973). Family influence and involvement in Sports. Research Quarterly, 88, 249-255. Snyder, E. E., 8 Spreitzer, E. A. (1983). Social aspects of sport (2nd ed.). Englewood Cliffs, NJ: Prentice- Hall. Spence, J. ., 8 Helmreich, R. (1972). The attitudes toward women Scale: An objective instrument to measure attitudes toward the rights and roles of women in contemporary society. JSAS Catalog of Selected Documents in Psychology, 8, 66. Spence, (1. Th, 8 Helmreich, R., 8 Stapp, J. (1975). Likability, sex-role congruence of interest, and competence: It all depends on how you ask. Journal of Applied Social Psychology, 8, 93-109. Spence, J. T., Helmreich, R., 8 Stapp, J. (1973). A short version of the attitudes toward 'women scale (AWS). Bulletin of the Psychonomic Society, 8, 219-220. Spreitzer, E., 8 Snyder, E. (1975). The psychosocial functions of Sport as perceived by the general population. International Review of Sport Sociology, 88, 87-93. Stevens, J. (1979). Comment on Olson: Choosing a test statistic in .multivariate analysis of variance. Psychological Bulletin, 88, 355-369. Sutherland, E. (1970). Colonized women: The Chicana. In E. Sutherland (Ed.), Sisterhood is powerful. New York: Random House. Thornton, A., 8 FTeedman, D. (1979). Changes in the sex role attitudes of women 1962-1977: Evidence from a panel study. American Sociology Review, 88, 831-842. Triandis, H. C. (1971). Attitude and attitude change. New York: John Wiley 8 Sons. Trotter, R. T. (1982). Ethnic and sexual patterns Of alcohol use: Anglo and Mexican American college students. Adolescence, 81, 305-325. Ulrich, C. (1979). The Significance of sport sociology to the profession. In M. L. Krotee, (Ed.). 888 dimensions of sport sociology. West Point, NY: Leisure Press. U.S. Bureau of the Census. (1984). Statistical abstract of the United States: 1984. Washington, DC: U.S. Government Printing Office. U.S. Bureau Of the Census Current Population Reports. (1980, March). 8§rsons of Spgnish origin in the United States, 88(354). Vanfossen, B. EL (1977). Sexual stratification and sex- role socialization. Journal of Marriage and the Family, 32' 563-..74. Vanfossen, E. (1979). The structure of sexual inequality. Boston: Little, Brown. Westkott, M., 8 Coakley, J. J. (1981). Women in Sport: Modalities for feminist social change. Journal of Sport and Social Issues, 8, 32-45. Williams, R. (1970). American society. New York: Alfred A. Knopf. Winer, B. J; (1971). Statistical principles in experimental design. New York: McGraw—Hill. Wittig, A. F. (1975). Attitudes towards females in Sports: In D. M. Landers, (Ed.), Psychology of motor behavior and sport II. Pennsylvania State University. Woodford, R. C., 8 Scott, W. J. (1982). Attitudes toward the participation of women in intercollegiate sports: Evidence from a metrOpolitan area survey. In A. O. Dunleavy, A. W. Miracle, 8 C. R. Rees. Studies in the sociology of Sport. Fort Worth: Texas Christian University Press. Yoder, J. D., Rice, R. W., Adams, J., Priest, R. F., 8 Prince, H. T. (1982). Reliability Of the attitudes toward women scale (AWS) and the personal attributes questionnaire (PAQ). Sex Roles, 8, 651-657. Zimbardo, P. G., Ebbesen, E. B., 8 Maslach, C. (1977). Influencing attitudes and changing behavior. Reading, MA: Addison-Wesley.