....u ”.21., .,....»{..__.,'."".,,.. . m- ':....'..u:' . ..... n.7,.» . . . .‘..“.'.‘-'.’ .........._ . . .. v - ”1. I.” nu-r-rv-u n.- . "A w. "a“. ud-‘M- .Iy-“r:-in-u'r~ , -. .......- . .. 4II;I*C:v-.a¢. .>....A...:v......-...... .. .A ' '1‘ .. .. . ‘ .... _ m. .,‘.. . - “L ......... v mu. .0- .. «_« _.- ‘....'...'.x.-.,,..."I... . ."‘....’......,.‘. , v . . _.. ,., L“"’"’.Z'""’"" 3.,r.-.v'."........ "'.."' -r u - .. ...' r. .a ' >rrn~'. .Hn .J I. ”- r v.7 - my .uwv . '- ......:.. .- uo- .. on . any" ”or .- *v >rv u n .v n v --< . .. 1.. -.. . .1 v: ¢ 0 ”'31... . n...... .a.» . - - tr'unn—u.’ “ , A "1"" .u «.n . . 0» wmv ' a.~v- -'v n.2,: n a.’L...n-v'v.o.:ur.rr r—rvw- ,, v00 7'- w :‘ v..'7-ou.3§»'m :$-:.w~v n. :‘3 -‘l‘¢‘ .1”. u :0 :vturc' 0:1...r4‘vu' .yyra-«vtt'm u- «6;. . I-‘wv-r' ovum-v am:- .5» 'r. a. r. r‘;g'. .. u .1- . u r...- ' I.‘.' .. ...... . urn-l-‘¢i~lov. .V,.--uaun«- , 5! .I K .Pfi‘.‘ MICHIGAN sn IIII/f/III/l/Ifl/ifllifliii’iifllfiifliflil/T/il 3 129 896 3138 ll This is to certify that the dissertation entitled PSYCHOTHERAPISTS' PERCEIVED DIFFERENCES 0F LOCUS OF CONTROL, ASSERTIVENESS, SELF-DISCLOSURE, AND PSYCHOLOGICAL ADJUSTMENT FOR MEXICAN- AND ANGLO AMERICANS presented by David Robert Arreoia has been accepted towards fulfillment of the requirements for Ph.D. degreein Counseiing, Educational Psychology, and Special Education %9 L, OCJQZ‘LL-c; CL Major professor \ .“ Dam May 1991 MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771 i LIBRARY MICh'Qan State University I L; PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE “FEB 24-20%“ 11:31.; |:l__ —7 __1 :fi MSU Is An Affirmative Action/Equal Opportunity Institution chS-DJ PSYCHOTHERAPISTS’ PERCEIVED DIFFERENCES OF LOCUS OF CONTROL, ASSERTIVENESS, SELF-DISCLOSURE, AND PSYCHOLOGICAL ADJUSTMENT FOR MEXICAN- AND ANGLO AMERICANS By David Robert Arreola A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Educational Psychology, and Special Education l991 ABSTRACT PSYCHOTHERAPISTS’ PERCEIVED DIFFERENCES OF LOCUS OF CONTROL, ASSERTIVENESS, SELF-DISCLOSURE, AND PSYCHOLOGICAL ADJUSTMENT FOR MEXICAN- AND ANGLO AMERICANS By David Robert Arreola Past research has suggested there may be ethnic differences of locus of control orientation, assertiveness, and self-disclosure between Mexican- and Anglo Americans. Research efforts have focused on members of the two ethnic groups, and the conclusions have often been contradictory and controversial. This study examined the perceptions of ethnic differences of locus of control, assertiveness, and self-disclosure of those who work with both ethnic groups--psychotherapists. The subjects were 207 psychotherapists licensed in the state of Texas. Each subject was given one of six vignettes depicting a client with different levels of locus of control, assertiveness, and self-disclosure. The client described in the vignettes was either Mexican American, Anglo American, or "ethnically neutral." Psychotherapists were asked to read the vignettes and to respond to a six-page questionnaire. The questionnaire was used to assess psychotherapists’ perceptions of locus of control, assertiveness, David Robert Arreola and self-disclosure, in addition to how the different levels of these variables affect judgment of psychological adjustment. Multiple regression, analysis of variance, and a post-hoc test indicated no support for perceptual differences of locus of control, assertiveness, or self-disclosure between the two ethnic groups. Statistical evidence indicated the responding psychotherapists did not perceive ethnic differences along these variables. Further, no differences were found in how clinicians judged the level of psychological adjustment based on the level of the other dependent variables. Although the main hypotheses were not generally supported, there were a number of other findings that proved to be germane to the area of interest. Some of the dependent variables were found to be significantly correlated with the number of, percentage of, and level of experience with Mexican American clients psychotherapists reported having had. Further, there were many significant correlations of other independent variables. These curious findings, in addition to the findings of the main hypotheses, were discussed, as were the implications for further research. Dissertation Committee Chairperson: Dr. John Powell The energy, time, and expense that have been necessary to complete this research are dedicated to my parents, for their emphasis on the value of learning and education, and to my lovely wife, for her patience, understanding, and solid support. iv ACKNOWLEDGMENTS I wish to make known my appreciation for those who have provided encouragement and assistance. Special appreciation goes to Dr. John Powell, chairperson of the dissertation committee, who has given his time and has offered many fruitful suggestions. Also, special thanks are given to Dr. Dennis Simpson of Western Michigan University for his guidance and energy. Finally, I acknowledge the emotional support and editing efforts of many friends, colleagues, and family members. Thank you all! LIST OF Chapter I. II. III. IV. V. TABLE OF CONTENTS TABLES ....................... INTRODUCTION .................... REVIEW OF THE LITERATURE .............. Fatalism and Locus of Control ........... Self-Disclosure .................. Assertiveness ................... Definition of Terms ................ METHODOLOGY ..................... Subjects ..................... Procedure ..................... Cover Letter .................. Follow-Up Letters ................ Client Vignettes ................ The Questionnaire ................ Hypotheses .................... Research Design .................. Statistical Analyses ............... RESULTS ....................... Description of Survey Subjects and Vignette Distribution .................. Results of the Statistical Analyses ........ Other Pertinent Results .............. DISCUSSION ..................... Review of the Findings .............. Limitations of the Study ............. Implications for Research ............. vi 4T 55 89 93 96 Too 102 Page APPENDICES A. VIGNETTES ...................... 110 B COVER LETTER .................... 116 C FIRST FOLLOW-UP LETTER ............... 117 D. SECOND FOLLOW-UP LETTER ............... 118 E QUESTIONNAIRE .................... 119 F PERMISSION LETTER FROM UNIVERSITY COMMITTEE ON RESEARCH INVOLVING HUMAN SUBJECTS .......... 125 G. LETTER CONVEYING RESULTS OF THE STUDY TO RESPONDENTS ..................... 126 REFERENCES ......................... 127 vii Table (humane-um 10. ll. 12. l3. T4. 15. 16. 17. LIST OF TABLES Psychotherapists’ Gender .............. Psychotherapists’ Profession ............ Psychotherapists’ Primary Theoretical Orientation . . . Psychotherapists’ Ethnic/Racial Background ..... Psychotherapists’ Current Activity Level ...... Psychotherapists’ Primary Clients .......... Percentage of Mexican American Clients ....... Number of Mexican American Clients Within the Last Year ....................... Years of Experience Working With Mexican American Clients ...................... Psychotherapists’ Reported Familiarity With the Mexican American Culture ............. Psychotherapists’ Reported Fluency in Spanish . . . . Response to Vignettes by Profession ......... Total Response Percentages for Each Vignette . . . . Psychotherapists’ Response to "High" and "Low" Vignettes ..................... Psychotherapists’ Response to "Ethnic" Vignettes Correlation Coefficients: Zero-Order Correlations Among Dependent and Independent Variables ..... Multiple Regression Results: Assertiveness and Evidence by All Other Independent Variables . . . . viii 48 49 50 51 52 52 53 53 56 58 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. Results of Two-Way ANOVAs: Assertiveness by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy ............... Results of Two—Way ANOVAs: Assertiveness by Vignette and Child/Adult, Percent of MA Clients, Number of MA Clients, and Experience With MA Clients ..... Results of Two-Way ANOVAs: Assertiveness by Vignette and Race/Ethnic Background, Culture Familiarity, and Spanish Fluency ................. Results of Neuman-Keuls Post-Hoe Analysis: Assertiveness by Vignette .............. Multiple Regression Results: Assertiveness and Evidence (and Conducting Therapy) by All Other Independent Variables ................ Results of Two-Way ANOVAs: Self-Disclosure by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy ......... Results of Two-Way ANOVAs: Self-Disclosure by Vignette and Child/Adult, Percent of MA Clients, Number of MA Clients, and Experience With MA Clients ....................... Results of Two-Way ANOVAs: Self-Disclosure by Vignette and Race/Ethnic Background, Culture Familiarity, and Spanish Fluency .......... Results of Neuman-Keuls Post-Hoc Analysis: Self- Disclosure by Vignette ............... Multiple Regression Results: Locus of Control and Evidence (and Ethnicity) by All Other Independent Variables ...................... Results of Two-Way ANOVAs: Locus of Control by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy ......... Results of Two-Way ANOVAs: Locus of Control by Vignette and Child/Adult, Percent of MA Clients, Number of MA Clients, and Experience With MA Clients ....................... ix Page 61 62 63 66 67 69 70 71 74 75 76 30. 31. 32. 33. 34. 35. 36. Results of Two-Way ANOVAs: Locus of Control by Vignette and Race/Ethnic Background, Culture Familiarity, and Spanish Fluency .......... Results of Neuman-Keuls Post-Hoc Analysis: Locus of Control by Vignette ............... Multiple Regression Results: Psychological Adjustment and Evidence (and Ethnicity, Gender, and Child/Adult) by All Other Independent Variables ................ . ..... Results of Two-Way ANOVAs: Psychological Adjustment by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy ......... Results of Two-Way ANOVAs: Psychological Adjustment by Vignette and Child/Adult, Percent of MA Clients, Number of MA Clients, and Experience With MA Clients ....................... Results of Two-Way ANOVAs: Psychological Adjustment by Vignette and Race/Ethnic Background, Culture Familiarity, and Spanish Fluency .......... Results of Neuman-Keuls Post-Hoc Analysis: Psychological Adjustment by Vignette ........ Page 79 82 84 85 86 87 CHAPTER I INTRODUCTION Mexican Americans are the fastest-growing minority in the United States. According to the U.S. Bureau of the Census, there are 20.8 Inillion Hispanics, of' whom approximately 60%, or 12.7 million, are Mexican American (these figures to not include the estimated 5 to 10 million illegal aliens). As this population continues to grow in the United States, the likelihood of mental health professionals providing services to them will continue to increase. Yet what we know about the Mexican American population is almost negligible. This lack of information is due in part to the scarcity of empirical research efforts. Of' importance, too, are the controversial results of the research that has been conducted. Much of' what has been written is based on dated field research by anthropologists. Most of these efforts concentrated on the people of Mexico or those very close to the Mexican border. By extrapolation, the findings were applied to all Mexican Americans. Thus, stereotypes proliferated and continue to do so. One of the most common stereotypes of Mexican Americans that has persisted concerns the presumed tendency for fatalistic thinking. For instance, Garza (1977) noted that most 2 anthropological and ethnographic accounts have depicted Mexican Americans as passive and controlled by the "external forces of luck, fate, and chance" (p. 98). Fatalism, then, is seen as a belief in an external locus of control over the events in one’s life (Ross, Mirowsky, & Cockerham, 1983). Much has been written concerning external and internal locus of control (LOC), and most researchers and clinicians would agree that people with an internal LOC are more likely to take steps to improve their situation, whereas those with an external LOC are more likely to give up. This giving up is considered to be impaired coping with environmental demands. As Ross et al. noted, the feeling that working toward a goal is futile, feelings of being helpless to control one’s own life, and the passive orientation and impaired coping effort that accompany these feelings are associated with psychological distress. Another common stereotype attributed to Mexican Americans concerns the supposed reluctance to disclose personal, emotional information. As Acosta and Sheehan (1978) noted, the popular cultural stereotypes often picture the Mexican American male as "one who is stoic and reticent in the face of emotional strife," and the female as one "who may experience emotional stress but is still a quiet long-sufferer who does not tell others about her problems“ (p. 546). In addition, Heiman, Burruel, and Chavez (1975) indicated that strong family loyalty and general mistrust of others by Mexican Americans are probable major factors for their reluctance to participate openly in psychotherapy. They noted that discussions of feelings and open communication, especially outside the family, are not encouraged. In general, low self-disclosure not only may be indicative of poor psychotherapy candidacy but also overall mental health. Indeed, high self-disclosure has been theorized as a correlate of good mental health (Cozby, l973; Ehrlich & Graeven, 1971). The third and last stereotype of interest is that of passivity. Karno (1966) believed that Mexican Americans’ tendency for relative defense, passivity, and polite, inhibited silence is poor equipment for the successful engagement of psychotherapy. Few would argue that the general ability to assert one’s needs and desires is indicative of coping and dealing effectively with the environment. Failure to do so is, conversely, indicative of poor psychological adjustment. Because psychotherapists are part of the larger culture in the United States, it is not surprising that many are influenced by and indeed may hold racial and ethnic stereotypes. Unlike other people in general, however, psychotherapists must make accurate mental health judgments that may have tremendous implications on the life of a client. The necessity for an unbiased and nonstereotypic-based diagnosis is evident. What is not evident is whether psychotherapists make judgments of psychological adjustment based on commonly held stereotypes. Specifically, what have not been investigated are the perceptions of mental health therapists regarding internal/external LOC, assertiveness, and self-disclosure differences between the Mexican American and Anglo American populations. The question a number of researchers have attempted to answer is whether Mexican Americans are different in these three areas by examining Mexican American and Anglo American subjects themselves and not the people who work with them in therapeutic settings. As will be shown, the results of these previous investigations have been equivocal. What also has not been investigated is how clinicians’ perceptions of internal/external LOC, assertiveness, and self-disclosure influence judgments of psychological adjustment of Mexican Americans. To study these two areas was this researcher’s intention. Given the rapidly expanding Mexican American population, the importance of accurate diagnosis, the commonly held stereotypes, and the largely equivocal findings of the research, the writer deemed it important. to assess the salience of ‘these attributes and their effects on psychotherapists’ judgments of psychological adjustment. Put another way, the question to be investigated is whether clinicians perceive Mexican Americans as more fatalistic (having a greater external LOC), more passive (less assertive), and less self- disclosing than Anglo Americans. If these differences are found to exist, the next question is how these stereotypic notions affect psychotherapists’ judgments of psychological adjustment. Given the stereotype (and some supporting empirical evidence) of an external LOC, low—self-disclosure, and less assertiveness for Mexican Americans, it might be expected that clinicians will invoke these stereotypes when presented with clinical vignettes depicting behavior consistent with the stereotypes. Further, it might be expected that there is an interaction of client ethnicity, stereotypic perceptions, and judgments of psychological adjustment. Specific hypotheses are presented in Chapter III. Chapter II deals exclusively with the research concerning fatalism and external LOC, self-disclosure, and assertiveness. Included are some of the earlier anthropological accounts in addition to the findings of more recent empirical research. In Chapter III,'the methodology of the present study is presented. Chapter IV details the results of the study, including the findings of the main hypotheses. In Chapter V, a discussion of the study is presented, including a review of the findings and their implications for the field of clinical psychotherapy. Further, implications for further research and limitations of the study are outlined in the last chapter. CHAPTER II REVIEW OF THE LITERATURE Fatalism and Locus of Control Internal-external locus of control has been considered to be a major personality variable. Rotter, Seeman, and Liverant (1962) described internal LOC as "the perception of positive and/or negative events as being a consequence of one’s own actions“ and external LOC as the perception that these events are "unrelated to one’s own behaviors . . . and therefore beyond personal control" (p. 499). The direct relationship between fatalism and external LOC is intuitively evident and has been commented on by several of the authors of the following studies. In a number of studies, researchers have investigated the relationship of internal/external LOC and ethnic group membership, but few have focused on differences between Mexican Americans and Anglo Americans. As can be seen in this review, the findings of the studies that have been pursued have resulted in controversial results. Various authors have described both Mexicans and Mexican Americans as fatalistic. Perhaps one of the best-known accounts is that of Madsen (1973). Speaking of the Mexican Americans of South Texas, he wrote: Suffering is also made acceptable by a strong belief in fatalism. . . . Fatalistic philosophy produces an attitude of resignation, which often convinces the Anglo that the Latin lacks drive and determination. What the Anglo tries to control, the Mexican American tries to accept. (p. 18) It should be noted that these comments were based, not on empirical research but rather on a case study by anthropologists. A more empirically based account was offered by Jessor, Graves, Hanson, and Jessor (1968). These authors studied value differences and ”deviant behavior" of three distinct groups in a small town in southwestern Colorado. One of the groups studied were "Spanish- Americans," which was a common name for Mexican Americans at the time and location of the study. In this study, 93 Anglos, 60 Spanish, and 68 Indian adults were interviewed. In addition, 36 Spanish, 11 Indian, and 42 Anglo high school students were given group-administered Iquestionnaires to assess beliefs in internal/ external LOC. The questionnaire was based on a series of tryouts, item analyses, and revisions of a version of a forced-choice Internal-External Inventory constructed by Liverant. The results of this study indicated that there were no ethnic group differences on the I-E scale. Grebler, Moore, and Guzman (1970) also studied internal/ external LOC along with a number of other variables. Unlike others, however, the samples used in their study consisted of both Mexicans and Mexican Americans. Labeling the dimension studied "activism" (e.g., taking control over the environment by planning), the authors attempted to discover whether there were differences between groups of Mexican Americans in San Antonio and Los Angeles, and Mexicans living in Mexico City. The results showed that, for almost all items, most respondents from each city and country disagreed with the passive stance of the statements in the activism scale. Grebler et al. noted that this finding was at variance with the notion that Mexican Americans are passive. The authors concluded that the data showed no support whatsoever for the notion that Mexican Americans are culturally unique with regard to their belief in potential mastery of their environment. In 1976, Farris and Glenn conducted a study using the same pool of subjects described by Grebler et al. The same activism-mastery scale was also used. The scale consisted of four statements with which the respondents were asked to agree or disagree, an "agree" indicating a passive-fatalistic orientation and a "disagree" indicating an active-instrumental orientation. In the general analysis, the Mexican American respondents were more fatalistic than the Anglos, but as the authors noted, the differences were not extremely large. Because the Mexican Americans in San Antonio were much more concentrated at the lower levels of social economic status (SES), family income, education, and occupation of head.of household were controlled. It was found that, because fatalism varied inversely with the amount of formal education among both groups, controls for education appreciably reduced the ethnic differences. Farris and Glenn concluded that any aggregate difference in fatalism between Anglos and Mexican Americans was largely a function of differences in amount of education or of "differences in close correlates of education" (p. 398). A related study is that of Evans and Anderson (1973). The sample consisted of 87 Mexican American and 39 Anglo American junior high school students from Las Cruces, New Mexico. Subjects were administered a questionnaire designed to measure, among other things, achievement value orientations. 0f importance to the present study, it was found that Mexican American students were ”far more fatalistic about their futures and were more skeptical about the value of planning ahead than their Anglo classmates" (p. 413). In l974, Garza and Ames conducted a classic study by comparing the LOC orientations of Anglo and Mexican American college students. Using Rotter’s I—E scale, data were collected from 47 Anglos and 47 Mexican Americans. In addition to studying differences on the full scale, the researchers also investigated differences within the subcategories of luck and fate, politics, respect, leadership and success, and academics. The. results of the study showed that Mexican Americans scored significantly less external than Anglo Americans on the full I-E scale. Also, there were significant differences on the respect and luck/fate dimensions, as well. Garza and Ames concluded that their findings suggested that, contrary to cultural stereotypes, Mexican American culture may actually contribute to a greater perception of internal control. Garza (1977) extended the previous study to include 244 Chicano (Mexican American) and 203 Anglo college undergraduates. The Mexican American sample used in the study was predominantly 10 bilingual and bicultural. As in the earlier study, Rotter’s Internal-External scale was administered to all students, and differences were investigated for the subcategories described previously. The results showed there was definite "cultural equivalence" for the luck/fate factor and, to a lesser extent, the leadership/success factor. The equivalence of the remaining three dimensions, however, remained questionable. College students have been used for subjects in a number of studies, which often creates problems of generalizability and covariable contamination. Scott and Phelan (1969), however, decided to investigate the LOC orientation of three groups of 60 22- to 28- year-old unemployed males who were matched for age, socioeconomic status, and scholastic aptitude. The result of the analysis was that unemployed Whites were significantly more internal in orientation than were Blacks and Mexican Americans. Most significantly, Mexican Americans showed an even greater tendency in the external direction than did the other two groups. Knight, Kagan, Nelson, and Gumbiner (1978) obtained measures of LOC and field independence from 144 Mexican American and Anglo American fourth-, fifth-, and sixth-grade children. To measure LOC, the Intellectual Achievement Responsibility Questionnaire was administered to all students. One of the findings of this study was that there were no cultural differences in the domain of LOC. The authors noted that this is contrary to "the notion that fatalism is 11 a cultural characteristic of Mexican Americans or at least that Mexican Americans are more fatalistic than Anglo Americans” (p. 94). Rotter’s Internal External Locus of Control Scale was again used in a cross-national study that also sampled Chicano students. Cole, Rodriguez, and Cole (1978) administered the instrument to male Catholic business administration students in the United States, Germany, Ireland, and Mexico. Of importance here was the comparison of LOC scores for 151 Angle and 95 Chicano high school students from three Southern California high schools. REsults of the analysis did not show a group difference. In fact, there was a high degree of similarity between the scores of the Chicanos and Anglos. Only the male Chicanos not planning to enter college showed any tendency toward a more external LOC. Interesting, too, was the finding that Mexican university students were more internally oriented than students from each of the other nations, including the United States. Ross et a1. (1983) used a sample that included both Mexicans in Mexico and persons of Mexican heritage in the United States. Data were collected by means of face-to-face interviews in El Paso, Texas, and Juarez, Mexico. The total number of subjects was 463. Based on self-reported ethnic identification ("Mexican" or "American") and language use, the sample was dichotomized in two groups: Mexican cultural identity and American cultural identity. The results indicated that fatalism was strongly associated with Mexican cultural identity and low social class. When social class was controlled, the strength of the relationship between fatalism 12 and Mexican identity' was reduced considerably, but it remained significant at the .01 level of significance. The authors concluded that, although some researchers have found that the effect of Mexican culture on fatalism was explained by social class, the findings of their study showed that Mexican culture had an independent effect on fatalism. Justin (1970) randomly selected 168 male Mexican American seniors and 209 male Anglo seniors from four urban Tucson, Arizona, high schools. A questionnaire that was developed and previously tested for reliability was administered to the students. Statistical analyses of the data pertaining to delayed gratification (future orientation) and feelings of personal control (fatalism) were conducted. The overall results showed marked contrast between the two groups. The Mexican Americans significantly lacked "feelings of personal control and concern with delayed gratification" when compared to the other group (p. 28). Justin noted that the results are even more noteworthy because the sample consisted of second-semester seniors, a rather select group of achievers in relation to their peers. Young and Shorr (1986) sampled a large, ethnically mixed group of fourth, fifth, and seventh graders that included Blacks, Whites, and Mexican Americans. Each child was given the Academic Achievement. Accountability' Questionnaire, which consists of questions that deal with the responsibility a child assumes for grades or the preconceived consequences of not studying. Also 13 administered was one of two different achievement tests. Results of this study indicated that Whites were more internal than Black and Mexican American children, who did not differ significantly from one another. The authors noted, however, that further comparison involving White and Mexican American children indicated that SES was a significantly stronger correlate of LOC than was ethnicity. Buriel (1982) investigated the relationship between Anglo and Mexican American children’s LOC orientation and their reading and math achievement. He examined 118 Angle and Mexican American fourth- and fifth-grade boys and girls with similar SES backgrounds. An Intellectual Achievement Responsibility Questionnaire was given, and teachers’ attitudes of student competence and motivation were measured. Results showed no significant differences between the children for any of the dependent measures. Although LOC orientation was originally perceived to cut across all aspects of behavior, some researchers have come to recognize the variable as domain-specific. A 47-item instrument based on such a conceptualization was subsequently developed and used in the research effort of Gaa and Shores (1979). The instrument measures perceived acceptance of responsibility for both success and failure in the domains of intellectual, physical, and social activities. A sample of 340 juniors from two state universities were given the instrument, and analysis was performed. The two researchers found that Mexican American subjects were more internal than Anglo subjects with respect to success in intellectual activities. However, no significant differences were found between the two 14 groups in the physical domain, and the Anglo subjects were significantly more internal than Mexican Americans relative to social success. Further, Anglo subjects were significantly more internal with respect to social failures. Gaa and Shores concluded that LOC generalizations across domains are inaccurate and that domain-specific LOC measures reflect "distinct but not consistent” differences between the two culturally divergent populations. Expanding on the previous study, Gaa, Williams, and Johnson (1981) attempted to determine whether similar results would be found with a sample of high school students. This time, the domain- specific instrument was administered to 204 high school students from one large metropolitan school. The overall results supported the earlier findings. In addition, Mexican Americans were generally more internal than the other groups, including Anglos. To determine specifically the influence of SES on LOC, a comparison of 86 Angle and 80 Mexican American high school adolescents was done by Buriel and Rivera (1980). Each subject was given the Rotter Internal—External LOC scale, and family incomes were tabulated. Analysis of variance indicated that Anglos were more internal on the politics dimension, whereas on the respect category Mexican Americans were more internal. Mexican Americans were found to have lower family incomes, and when controlling for income, analysis revealed the difference in politics was removed. The authors concluded that fatalism is not a cultural characteristic of Mexican Americans. 15 Chandler (1979) offered a review of Mexican and Anglo American value orientations and a study of his own. Numerous field studies conducted in Texas and New Mexico were cited and provided "convincing evidence" that Mexican Americans ”cling to values characterized by fatalism, low achievement drives [and] past and present time perspectives” (p. 153). Chandler also cited various works pointing to lower-than-Anglo scores on measures of mastery, fatalism-activism, competitiveness, and field independence. For his own research, two similar interview schedules were given to 712 Anglos and 323. Mexican Americans from Lubbock, Texas. Results showed that there were statistically significant differences between the respondents in the areas of mastery and future-time perspectives; Mexican Americans scored much lower on these dimensions than Anglo Americans. The only other studies of differences of Mexican American and Anglo American LOC found in the literature and of importance to note are a few unpublished doctoral dissertations. Cabrera (1963) synthesized data on American and Mexican American cultural values in order to improve educational programs. The basic sources of data for the study were literature and publications in psychology, education, and other social sciences. Of importance, the review indicated that American middle-class value orientations include man’s mastery over the universe. Fatalism, values in being rather than doing, and accommodation to problems, however, were found to be cultural characteristics of Mexican American value orientations. l6 Rogers (1971) sampled eighth-grade Mexican American children and found them to be less internal than a comparable sample of Anglo children. Scores on the internal/external LOC measure increased with acculturation, but differences were obscured to some extent by gender' differences. In general, acculturation and SES in combination predicted significant amounts of variance, and, overall, SES was found to be a more powerful predictor than acculturation. Garza (1977) used Rotter’s Internal-External scale to measure the LOC orientation of 76 Mexican American ninth-grade students. The scores of the subjects were contrasted with those of two comparison groups, one of which was "American" high school students. Results showed significant differences. Garza concluded that Mexican American students perceived the control of their environment to be externally located, which was a result of an orientation directed toward Mexican society. Acculturation as measured by generation was the focus of a LOC study by Mahaffey (1986). Using the Levinson (L), Powerful others (P), and Chance (C) LOC scale scores taken by 117 Mexican Americans and compared to Anglo American norms, some significant and important findings were discovered. Mahaffey found first-generation Mexican American males’ scores to be equal to Anglo males’ scores on all three LOC measures. However, second-generation Mexican American males were more internal than Anglo males yet comparable on the P and C scales. In addition, third-generation Mexican Americans scored higher on the L and lower on the P and C scales. It was also found that second- and later-generation Mexican American females did 17 not differ from Anglo females, but first-generation Mexican American females scored lower on the L scale than did their counterparts. The author concluded that acculturation was positively related to the L scale and inversely related to the C scale. In general, later-generation Mexican Americans were more internal and less chance oriented than earlier generations. While describing his research, Carballo-Dieguez (1986) pointed out that the evidence supporting the speculation that cultural background is a determinant of LOC beliefs is incomplete, often contradictory, and often biased by confounding factors. This researcher hypothesized that all possible differences in LOC scores between the two groups could be attributed to SES, religion, family’s educational level, and acculturation. Study findings confirmed a significant difference in LOC score for a subsample of high school sophomores; Mexican Americans were less internally oriented than Anglo students. However, analysis of covariance showed that, when family income level and subjects’ educational level were accounted for, the differences in LOC scores lost significance. There has also been a study on dream content as it relates to cross-cultural differences in LOC. Forty-eight Mexican American and 30 Anglo American adolescents were given a dream questionnaire and a LOC scale revised by the researcher (Kaflowitz, 1985). Comparatively, Mexican American adolescents were higher on the external-social dimension of the LOC scale but lower (Hi the 18 external-powerful others dimension. These mixed results were then interpreted in terms of different dream theorists. Finally, the last LOC research effort to be reviewed is that of Soza (1982). As part of his effort, a questionnaire was distributed to 394 high school students in the Los Angeles area. Discriminant analysis, factor analysis, and point-biserial correlations were used to evaluate the responses of Mexican and Anglo Americans. The results indicated that ethnicity was indeed more highly correlated with an external LOC than economic level. Self-Disclosure Do Mexican Americans generally differ in the amount and patterns of self-disclosure? Although self-disclosure styles of Mexican Americans have been open to significant speculation, few studies have been undertaken to investigate this question empirically. One of the few known studies regarding possible ethnic differences in self-disclosure is that of Acosta and Sheehan (1978). The study involved 187 junior college students (94 Mexican Americans and 93 Anglo Americans). All students listened to one of four audio-taped therapist introductions in which the therapists differed in ethnicity and level of professionalism. After hearing the tape, each subject was asked to indicate on a self-disclosure scale the degree of willingness to talk to the therapist he/she had just heard. Results of the study showed that Mexican Americans proved to be significantly lower in self-disclosure scores than Anglo 19 Americans. Socioeconomic status did not affect the results. However, the researchers argued that both ethnic groups indicated a willingness to disclose about themselves. Although Chicanos disclosed less than Anglo Americans, their mean disclosure scores indicated a positive tendency to disclose "in general” to the therapists. The authors concluded that, contrary to the generally held assumption that Mexican Americans are likely to show little willingness to talk of personal matters, the findings of their study showed that Mexican Americans, in general, were willing to make considerable disclosure. Littlefield (1974) sampled 300 Caucasian, Black, and Mexican American ninth-grade students. A modified version of Jourard’s Self-Disclosure Questionnaire was administered to each subject in a standardized interview. With the assumption that "self-disclosure“ referred to the total disclosure scores of the measuring instrument, it was found that the Caucasian group reported the greatest amount of self-disclosure and the Mexican American sample reported the least disclosure. Especially noteworthy was the high disclosure rate of the Caucasian females and the low disclosure rate of the Mexican American males. Also, within the Mexican American sample, females were decidedly higher disclosers than males. LeVine and France (1981) also focused on the question of whether there are differences in the amount of self-disclosure, in addition to questions of differences in what is disclosed and to whom such disclosure is made. Although Anglo Americans were compared to "Hispanics," it can be assumed that at least the 20 majority of the Hispanics were Mexican Americans because the research subjects were New Mexico State University students. The sample consisted of 155 Anglo American and 119 Hispanic undergraduate students. Jourard’s Self-Disclosure Questionnaire was administered by four doctoral students (Anglo American male and female, Hispanic female and male). Results showed that Anglo American females reported the highest mean self-disclosure, followed. by Hispanic females, Anglo males, and finally Hispanic males. In the gross analysis, Hispanics were less disclosing than Anglo Americans, but when gender and ethnicity of the administrator were varied, the trend was not maintained. Under Hispanic female examiner conditions, Hispanics reported self-disclosure comparable to or surpassing that of Anglo Americans. LeVine and Franco suggested, then, that significant intervening variables may be involved in self-disclosure that have been undetected in previous studies. Franco, Malloy, and Gonzalez (1984) replicated the research efforts of the earlier study by Levine and Franco. In addition to using the Jourard Self-Disclosure Questionnaire, an acculturation rating scale for Mexican American subjects was also used. From a subject pool of 158 Anglo Americans and 108 Mexican Americans, the researchers determined that acculturation did run: correlate significantly with reported self-disclosure under any circumstances. It was also shown that there were no ethnic group differences of self-disclosure. 21 Further expanding on the effects of examiner variables, Franco and LeVine (1985) conducted a study of self-disclosure that included different styles of administration: directive, nondirective, and neutral. The researchers again used the Jourard Self-Disclosure Questionnaire, but it was given in the three different administrative styles and to a much larger sample of New Mexico State University students (N - 811). Results of the analysis indicated that Anglos disclosed significantly' more than Mexican Americans. However, the overall main effects determined that the style of administration elicited significantly different amounts of self-disclosure. The authors strongly suggested that these results be interpreted cautiously due to the number of significant interactions. Borrego, Chavez, and Titley (1982) were also determined to cull examiner variables and their effects on self-disclosure rates of Mexican Americans. Fifty-seven Mexican American and 67 Anglo American undergraduate college students listened to one of three audiotapes of an enacted interview session and completed a modified Jourard Self-Disclosure Questionnaire. The overall results indicated that there were no diffErences across the three different intervention techniques. Most salient to this discussion was the finding that there were no ethnic differences in willingness to disclose. Finally, Molina and Franco (1986) analyzed self-disclosure patterns of 88 Anglo American and 63 Mexican American undergraduate students. Administrator ethnicity and gender were systematically 22 varied to determine the interaction of these variables with participant ethnicity and gender. Again, Jourard’s Self-Disclosure Questionnaire was used and was administered in a neutral manner to the subjects. The most important results showed than Anglo American males reported the highest mean self-disclosure, followed by Mexican American females, Anglo American females, and Mexican American males. Further, there were significant differences in preferred targets of self-disclosure, with males reporting less disclosure to their mothers than females. The researchers suggested that the data be interpreted cautiously and argued that investigations of self- disclosure in which Mexican Americans are found to be less disclosing than Anglo Americans might not be controlling for significant intervening variables. As can be seen by this review, little research has been done regarding self-disclosure differences between Mexican Americans and Anglo Americans. To date, there have been no other known attempts to investigate Mexican American/Anglo American differences in self- disclosure. With such little available empirical research, and given the findings of these studies, it would be difficult to support the position that Mexican Americans are less self-disclosing than Anglos. Thus, like the research regarding LOC, it has not been empirically' demonstrated one way or the other that there are, indeed, differences between Mexican Americans and Anglo Americans in general degree of self-disclosure. 23 Assertiveness Another frequently cited stereotype is that Mexican Americans are passive, submissive, noncompetitive, and conforming. If these characteristics are generally true, the leve1 of assertiveness would be different for Mexican Americans compared to Anglo Americans. Much like the research regarding self-disclosure and LOC, however, few' empirical studies have investigated these differences. Similarly, the results of these empirical investigations have been inconsistent. One such investigation was undertaken by Kagan and Carlson (1975). Noting that most of the descriptions of Mexican passivity have been based on comparisons of rural Mexicans with urban Anglo Americans, these investigators examined the assertiveness of ‘154 rural Mexican, semi-rural Mexican American, semi-rural Anglo American, and urban Anglo American children. An assertiveness "pull scale" was used as a measurement and was constructed so that there was an optimum level of assertiveness above which or below which children lost toys for which they were striving. Other unobtrusive measures were also used and included the number of questions and comments the subjects made and the amount of spontaneous manipulation of the experimental apparatus (the assertiveneSs pull scale). The results showed no significant differences between Mexican American and Anglo American children living in the same poor semi-rural community. As the researchers noted, despite their different cultural backgrounds, both groups behaved quite similarly. 24 Grodner (1975) investigated the effect of cultural differences and SES on levels of assertiveness and anxiety. Ninety Angle and Chicano psychiatric patients at a New Mexico Veterans Hospital were given the Adult Self-Expression Scale and the Taylor Manifest Anxiety Scale. The results showed no significant differences between Anglos and Chicanos of any socioeconomic class on levels of assertiveness. Further, it was determined that assertiveness was significantly correlated to socioeconomic class but not to ethnicity. Hall and Beil-Warner (1978) used Galassi’s College Self- Expression Scale to measure the level of assertiveness of 78 male introductory psychology students. The scale contains 50 Likert-type items categorized into "person" categories relating to friends, opposite sex, business relations, parents, strangers, authority figures, and a nonspecific category. The results of this study showed that the Anglo sample of 34 students scored significantly higher on the pencil/paper measure than did the Mexican American sample on overall assertiveness. Significant differences were found for assertiveness with parents, same-gender peers, and business relations. Mexican Americans tended to report engaging in demand behavior, refusal behavior, and expressions of annoyance less frequently than the Anglo Americans, which was interpreted in terms of the "nature of the Mexican American culture with its general emphasis on familism and the maintenance of family relations" (p. 178). 25 The Rathus Assertiveness Schedule (RAS) is a self-measure that is based on responses about behavior in different kinds of everyday situations. Kimble, Marsh, and Kiska (1984) used the RAS with 782 Mexican American and Anglo American female and male undergraduate students. Results showed that there was a greater difference in assertiveness between Mexican American males and females, with Mexican American females being clearly the least assertive of the four groups. Mexican American males, however, scored higher than Anglo American females. Overall, the findings were that gender and age were strongly related to assertiveness and that cultural differences in gender-role standards had a ”weaker but substantial influence on assertiveness" (p. 421). A related empirical study concerning assertiveness compared Mexican American females to Mexican American males and did not use an Anglo American group (Luian & Zapata, 1983). Using the Sixteen Personality Factor Questionnaire, Form C, the researchers found the exact opposite of the Kimble et al. study: Mexican American females were significantly more self-sufficient and assertive than Mexican American males. Citing earlier studies that had determined Mexican American children to be more cooperative and conciliatory in resolving potential interpersonal conflicts, Hoppe, Kagan, and Zahn (1977) investigated how' Mexican Americans and their Anglo counterparts dealt with maternal authority and children’s independence. It was hypothesized that Anglo American, field-independent, and male children would be more assertive than Mexican American, 26 field—dependent, and female children when there was a conflict situation with their mothers. The subjects were 32 third- and fourth-grade children and their mothers, who were interviewed and asked to participate in three tape-recorded make-believe situations. The transcripts were then analyzed by counting the fWequencies of various assertive verbal remarks and utterances. It was found that Anglo American children expressed and justified their wills more often than Mexican American children. Further, the overall results indicated that Anglo American children were more assertive and engaged in more direct conflict with their mothers than Mexican American children. Finally, Calvillo-Simpson (1983) investigated the effects of assertiveness and self-awareness skills training on 79 Anglo and Chicano high school students in an Upward Bound program. Ethnic differences in assertiveness ‘and self-esteem were studied first. The Texas Social Behavior Inventory, the Miskimins Self-Goal-Other Scale, and the RAS were administered to all subjects. Germane to this review, the results included the conclusion that there were no ethnic differences among the students, although the Chicano students had higher posttest means than Anglo students on most of the measures. Much like the research on self-disclosure and LOC, there is far too little evidence of ethnic differences regarding assertiveness. After careful review of all three areas of interest, it still cannot be asserted one way or the other whether Mexican Americans have 27 higher or lower levels of assertiveness, an internal or external LOC, or tendencies for little or more self-disclosure in comparison with Anglo Americans. Given the generally equivocal findings of the empirical research regarding these three variables, the present writer investigated whether those who work with both groups (i.e., psychotherapists) perceive ethnic differences in these areas. The question, too, is how these perceptions can affect clinicians’ judgments of psychological adjustment. Definition of Terms To examine the questions listed above, a large number of variables were involved in this study. To enhance readability and general comprehension, the terms used in the remainder of this study are defined as follows. Assertiveness (as a dependent variable)--Ability to express needs and desires, as measured by psychotherapists’ responses to Items 15 through 18 of the questionnaire. thldladult--Whether the main focus of a responding clinician’s clinical work is with children or adults, as measured by Item 39 of the questionnaire. Conducting therapy--Whether the responding psychotherapist is currently conducting therapy, as measured by Item 37 of the ques- tionnaire. Culture femilierjty--Degree of Mexican American culture famili- arity of the responding clinician, as measured by Item 42 of the questionnaire. 28 Ethniejty (as an independent variable)--Ethnic identification of the described client in the vignettes used in this study; includes Mexican American, Anglo American, and an ethnically "neutral” client. Evjdenee (as an independent variable)--Level of LOC, self- disclosure, and assertiveness, as depicted in the vignettes. Exeerieflee_yith MA cliee_s--Degree of experience the responding psychotherapist has in working with Mexican American clients, as measured by Item 44 of the questionnaire. Gender--Sex of the responding psychotherapist, as measured by Item 33 of the questionnaire. Lecgs of control (as a dependent variable)--Whether one attrib- utes consequences to chance or luck or to variables under more personal control. Measured by Items 26 through 30 of the question- naire. Member of MA clients--Number of Mexican American clients the responding clinician has seen within the last year, as measured by Item 41 of the questionnaire. Percent of MA clients--Percentage of Mexican American clients reported by the responding psychotherapist, as measured by Item 40 of the questionnaire. Profession--Professional identification of the responding clin- ician, as measured by Item 32 of the questionnaire. Psychological adjustment (as a dependent variable)--Level of healthy or impaired psychological functioning, as measured by Items 1 through 13 of the questionnaire. 29 Beeeletheie__eeekggeeng--Ethnic or racial background of the responding psychotherapist, as measured by Item 36 of the question- naire. W (as a dependent variable)--Ability to relate meaningful events and emotions, as measured by Items 21 through 24 of the questionnaire. Spanish flgency--Responding psychotherapist’s level of fluency in the Spanishlanguage, as measured by Item 43 of the question- naire. Theoretical orientation-~Main theoretical orientation of 'the responding clinician, as measured by Item 34 of the questionnaire. Vignette (as an independent variable)--One of six vignettes used in the study; also, a combination of the independent variables, ethnicity and evidence. CHAPTER III METHODOLOGY Previous research on Mexican American and Anglo American differences along the dimensions of locus of control (LOC), assertiveness, and self-disclosure has shown equivocal findings. It has not been consistently proven one way or the other that there are ethnic differences along these dimensions. In addition, previous researchers have studied subjects of both ethnic groups and have not investigated whether psychotherapists perceive ethnic differences. If, upon investigation, it is found that psychotherapists do perceive differences, an important question to consider is whether these differences affect psychotherapists’ judgments of overall psychological adjustment of' Mexican American and Anglo American clients. Specifically, the question that has not been asked by previous researchers is whether psychotherapists perceive LOC, assertiveness, and/or self-disclosure «differences between Mexican Americans and Anglo Americans. If psychotherapists do perceive ethnic differ- ences, the follow-up question to be investigated is how these different perceptions affect psychotherapists’ judgments of overall psychological adjustment. The methodology outlined in this chapter was designed to answer these questions. 30 31 sum The sample of psychotherapists in the study consisted of both psychologists and social workers. Psychologists were selected from a mailing list obtained from the Texas State Board of Examiners of Psychologists. Social workers were selected from a mailing list that originated with the Texas Department of Human Services. Both lists included only licensed personnel. Every thirteenth psychologist and every nineteenth social worker was selected from their respective lists. The total number of psychologists selected was 300, as was the total number of social workers, which resulted in a total of 600 clinicians. Procedure A packet of material was mailed to each of the 600 psychotherapists. Each packet of material contained a cover letter, a client vignette, a questionnaire, and a stamped and self-addressed envelope for the return of the questionnaire. All therapists received the same material with the exception of the cflient vignettes, which were created by the researcher. Of the six different vignettes, each clinician received only one. Subjects were randomly assigned to one of the following vignettes (see Appendix A): Vigeette MA Hiqh--Depicts a Mexican American with internal LOC, a high level of assertiveness, and high level of self— disclosing tendencies. Vignette AA High--Depicts an Anglo American with internal LOC, a high level of assertiveness, and high level of self- disclosing tendencies. 32 Vignette MA Le --Depicts a Mexican American with external LOC, a low level of assertiveness, and low self-disclosing tenden- cies. Vigeette AA Le --Depicts an Anglo American with external LOC, a low level of assertiveness, and low self-disclosing tenden- cies. Vignette NA fligh--An ethnically "neutral" vignette depicting a client with internal LOC, a high level of assertiveness, and high self-disclosing tendencies. Vignette NA to --An ethnically ”neutral” vignette depicting a client with external LOC, a low level of assertiveness, and low self-disclosing tendencies. geve: Letter The cover letter was on Michigan State University stationery and was co-signed by the researcher’s committee chairman. The letter contained a request for assistance and provided brief instructions. Further, the letter included information concerning anonymity and the meaning of an assigned code number on each questionnaire. In addition, potential respondents were informed that the true nature of the study could not be revealed at that time, and, for those who so desired, a full explanation of both the nature and results of the study would be sent to them. Finally, the cover letter included statements about the importance of the study, the voluntary nature of the research, and a fair estimate of completion time (see Appendix B). Fellow-U2 Letters After three weeks, a brief follow-up letter was sent to those clinicians who had not yet responded. Three weeks later, a second reminder was sent. To increase the likelihood of compliance, both 33 follow-up letters included an additional stamped and self-addressed envelope. Each letter encouraged potential respondents to fill out and return the questionnaires. ‘The letters also conveyed appreciation to those who might already have returned the questionnaire but whose responses had not yet been received or had been "crossed" in the mail (see Appendices C and 0). li n i n In their development, the client vignettes were modified often in an effort to convey accurately the different levels of LOC, assertiveness, and self-disclosure. Further, it was imperative that the clients depicted were as "real" as possible. As a first step, the~ original vignettes created by the researcher’ were given to various colleagues and fellow doctoral students for review and suggestions. Later, after incorporating their ideas, the researcher again gave the vignettes, on an informal basis, to other doctoral students, who were asked to comment on the extent to which each vignette conveyed the different levels desired and were discriminate from one another. After more revisions, vignettes NA High and NA Low were formally presented to ten experienced psychologists (five received NA High and five received NA Low). Responses suggested further revisions; subsequently, a final draft was again presented to the ten psychologists. This time, however, to prevent bias or boredom, the psychologists received the "opposite" vignette to the one they had received before. Each psychologist was instructed to 34 read the vignette and then rate LOC, assertiveness, and self- disclosure on three 7-point Likert scales. On the LOC scale, a numerical value of 1 represented "very internal LOC,” and a numerical value of 7 represented "very external LOC." The mean of the psychologists’ scores on the LOC scale was 6.4 for the NA Low vignette, with a standard deviation of 1.09; the mean was 2.4 for the NA High vignette, with a standard deviation of 1.09. On the self-disclosure scale, a numerical value of 1 represented "no self-disclosure," and a numerical value of 7 represented "a lot of self-disclosure." The mean of the ten psychologists’ scores on the scale was 2.4 for the NA Low vignette, with a standard deviation of 1.09; the mean was 6.0 for the NA High vignette, with a standard deviation of 1.41. On the 7-point assertiveness Likert scale, a numerical value of 1 represented "very nonassertive," and a numerical value of 7 represented A"very assertive." The mean of the scores on the assertiveness scale was 1.2 for the NA Low vignette, with a standard deviation of .9; the mean was 5.8 for the NA High vignette, with a standard deviation of 1.67. Given these figures, it was determined that there were desirable differences in how the ten psychologists responded to the two vignettes. Also, there was sufficient evidence that the vignettes did, indeed, depict accurately the different levels of the dependent variables. 35 The Qgestienneire The purpose of the questionnaire was to assess whether psychotherapists perceived a difference between the two ethnic groups along the dimensions of LOC, assertiveness, and self- disclosure. Also, the questionnaire facilitated the investigation of whether these perceptions affected judgments of psychological adjustment. Finally, the questionnaire included demographic information necessary for the completion of statistical tests regarding the independent variables. The questionnaire consisted of a series of questions to which subjects were asked to respond after reading the assigned vignette. In addition, demographic data were requested. At the very top of the questionnaire, statements were made regarding inferences of a client’s tendency toward adjustment/maladjustment. Respondents were asked for their opinions regarding the client’s inclination toward adjustment/maladjustment. There were also instructions on how to use the seven-item Likert scales. At the bottom of the questionnaire, respondents were asked whether they desired the results and an explanation of the study. Also, a blank line was left so that each questionnaire could be assigned a code number from 1 to 600. The questionnaires were coded so that complete anonymity was assured. A relative of the researcher was responsible for making sure those respondents who desired a response received a letter' explaining the results. The codes *were also useful for 36 determining which of the 600 potential respondents should receive the follow-up reminder letters. Overall, the questionnaire was divided into two main sections. The first section comprised four parts: overall psychological adjustment, LOC orientation, level of assertiveness, and level of self-disclosure. A substantial portion of the overall psychological adjustment part of the first section was derived from a previous study investigating psychological adjustment (Wolthuis, 1987). The reliability of the 20-item Wolthuis instrument was .96. The questions regarding the other three parts of the first section were developed by the researcher. The assertiveness questions were based on clinical knowledge and textbook examples of assertiveness. problems. The self-disclosure questions were also based on clinical experience and clinical knowledge. The questions regarding LOC, however, were based on questions asked in some of the LOC instruments mentioned in Chapter II. The second section of the questionnaire consisted of demographic information. Questions regarding respondents’ gender, profession, theoretical orientation, and ethnic/racial background were included. Further, questions concerning knowledge and clinical experience regarding Mexican American clients and culture were asked (see Appendix E). Efforts to insure face validity of the questionnaire included the revision of a number of drafts. The questionnaire was then given to six psychologists who had been provided an overview of this study and asked to submit verbal and written feedback. The six 37 psychologists suggested few changes other than those concerning format and design. Their feedback suggested there were few problems with the actual questions in the questionnaire. Considering the feedback, a decision to pilot test the questionnaire was made after some format and design changes were completed. The pilot testing consisted of sending the questionnaire and the two ethnically "neutral" vignettes to five social workers and _ five psychologists known to the researcher and who were not part of the earlier (vignette) pilot studies. These ten colleagues were asked to read one of the two vignettes and then respond to the questionnaire. Results of the pilot study indicated significant satisfaction. All ten clinicians indicated the questions were accurate and appropriate in regard to the dependent variables. Further, the respondents thought the questionnaire, including the demographic section, was easy to complete. In addition, there was little variability in how the respondents answered the questions regarding psychological adjustment, LOC, self-disclosure, and assertiveness. To test the variability of the responses and the overall reliability of the questionnaire, Cronbach’s coefficient alpha method was employed. Cronbach’s method is used when items are not dichotomous; it is useful when a score on each question takes on a range of values (Mehrens & Lehmann, 1984). The reliability of the psychological adjustment part of the questionnaire using the Cronbach method was .98. The reliability of both the self- disclosure and assertiveness parts was .97, and the reliability of 38 the LOC part was .96. Given the colleagues’ feedback and the reliability figures, it was determined the questionnaire was ready to be used in the study. are: lawless; The seven main hypotheses of this study, stated in null form, H t si : There is no difference in how psychotherapists perceive the level of assertiveness between Mexican Americans and Anglo Americans. Hypothesis 2: There is no difference in how psychotherapists perceive the level of self-disclosure between Mexican Americans and Anglo Americans. Hyppthesis 3: There is no difference in how psychotherapists perceive LOC orientation between Mexican Americans and Anglo Americans. Hypothesis 4: There is no difference in how psychotherapists perceive psychological adjustment between Mexican Americans and Anglo Americans, based on level of assertiveness. Hypothesis 5: There is no difference in how psychotherapists perceive psychological adjustment between Mexican Americans and Anglo Americans, based on the level of self-disclosure. Hypothesis 6: There is no difference in how psychotherapists perceive psychological adjustment between Mexican Americans and Anglo Americans, based on LOC orientation. Hypothesis 7: There is no difference in how psychotherapists perceive the NA High vignette client and the NA Low vignette client in terms of level of assertiveness, level of self- disclosure, and LOC orientation. The first six hypotheses were tested using the responses to the first four vignettes. The last hypothesis was tested using responses to the two ethnically "neutral" vignettes. The last hypothesis was included in the study to make sure psychotherapists 39 were indeed making distinctions between high and low levels of self- disclosure and assertiveness, and internal and external LOC. If these distinctions were made, it was thought that whatever differences occurred between the responses to the first four vignettes were related to the ethnicity of the described client. Research Desi n The overall research design of this study was the posttest-only control-group design. This design is used when there is a possibility that a pretest would contaminate the results (Borg & Gall, 1983). Most of the analyses used in this study were conducted with the Statistical Package for the Social Sciences (SPSS-X). The computer used was the IBM VM/CMS system at Michigan State University. Statistical Analyses A number of statistical tests were used in this study. Statistical analysis of the results first necessitated obtaining descriptive data, which included the means and standard deviations of the dependent variables and the demographic information (Arrin & Colton, 1970; Hays & Winnlon, 1971; Kirk, 1968; Meehl, 1958). Correlation data were also analyzed, which included the correlation coefficients of the demographic data and the dependent variables. The correlation coefficients were obtained to determine which demographic variables were important in their effects on the dependent variables (Galfo, 1970; Glass 8: Stanley, 1970; Lynch 8. 40 Huntsberger, 1976). Based on the correlation data, the multiple- regression method was chosen to estimate the weight of the independent variables on the dependent variables (Gorsuch, 1974; Harris, 1975; Kerlinger & Pedhazer, l973; Tatsuoka, 1971). The use of multiple regression determined the degree of significance (importance) several independent variables had on specific dependent variables. The specific analysis chosen was multiple regression (stepwise) as it shows the increment added by each predictor variable (Glass & Hopkins, 1984). The next statistical analysis involved the use of two-way analysis of variance (ANOVA). Finally, as a result of some significant findings from the two-way ANOVAs, a post-hoc multiple comparison test was used. Specifically, the Student Newmans-Keuls was incorporated to analyze further the independent variable that showed significant differences in the two- way ANOVAs and that had three or more values. The findings of ‘these statistical analyses are reported in Chapter IV. CHAPTER IV RESULTS In this chapter, the results of the statistical analyses used in the study are presented. The presentation of the results is divided into three main sections. The first section includes a description of the survey respondents and vignette distribution. The second section includes the separate results of the statistical analyses used to test each of the seven hypotheses. Finally, the third section of this chapter contains a description of other results that are pertinent to the discussion. Description of Survey Subjects and Vignette Distribution Of the 600 surveys sent out, 231 were returned (a 38.5% return rate). However, only 207 responses were used in the study as there were 24 incomplete questionnaires. Of the incomplete question- naires, seven were returned unopened due ix> an incorrect address, seven were sent back by "nonpracticing" clinicians who thought they could not answer accurately, and four were returned with no explana- tion or comment. In addition, four surveys were returned by clinicians who had written "too busy" on them, one was returned because the clinician worked exclusively with children, and one was returned because the clinician angrily disagreed with the premise 41 42 that one could make inferences based on a vignette. Of the 207 surveys used in the data analysis, none was filled out incorrectly, although a number of respondents elected not to answer every item in the demographic section. In an effort to determine why the majority of clinicians did not return the questionnaires, five nonresponding clinicians in the San Antonio, Texas, area were contacted by the administrative clerk hired to collate the results. The five responded to specific questions over the telephone. Although one of the five clinicians indicated that the materials sent were too lengthy, the general consensus of all five was that they were too busy with clinical duties to take the time to read and respond to the packet of materials. Further, another of the clinicians stated he had difficulty making judgments about clients without seeing them in person. One hundred six responding clinicians (over 51%) were female, 100 (over 48%) were male, and 1 (less than 1%) chose not to indicate gender (see Table 1). In terms of professional orientation, 108 respondents (over 52%) indicated they were psychologists, 91 (almost 44%) were social workers, 6 (3%) were "other" (presumably counselors or other professions), and 2 (1%) chose not to indicate their professional orientation (see Table 2). 43 Table l Psychotherapists’ Gender Gender Number Percent Female 106 51.2 Male 100 48.3 Unknown 1 .5 Total 207 100.0 Table 2 Psychotherapists’ Profession Profession Number Percent Psychology 108 52.1 Social work 91 43.9 Other 6 3.0 Unknown 2 1.0 Total 207 100.0 In terms of theoretical orientation, the largest group represented were those who endorsed an eclectic perspective (77 of 207 respondents or 37%). Cognitive behaviorists accounted for 58 (or 28%) of the respondents, and 21 (or 10%) indicated a strictly behaviorist orientation. Fifteen respondents (over 7%) indicated an analytic perspective, 12 (over 5%) endorsed a systemic perspective, and 8 (almost 4%) indicated a neoanalytic orientation. Of the remaining 16 respondents, 6 (2.9%) were Rogerian, 4 (1.9%) indicated 44 a Gestalt orientation, and 6 (2.9%) chose not to answer the question (see Table 3). Table 3 Psychotherapists’ Primary Theoretical Orientation Orientation Number Percent Eclectic 77 37.2 Cognitive behavioral 58 28.0 Behavioral 21 10.1 Analytic 15 7.2 Systemic 12 5.7 Neoanalytic 8 3.9 Rogerian 6 2.9 Gestalt 4 1.9 Unknown 6 2.9 Total 207 100.0 The demographic section of the survey included a question on the respondent’s ethnic/racial background. Originally, five categories were given, but because there were so few minority respondents, only two categories were maintained for data analysis. In this study, 181 responding clinicians (over 87%) were Caucasian, and 10 respondents (4.8%) indicated they were Black. Eight respondents (or 3.9%) indicated they were Mexican American, and two (or approximately 1%) were Asian/Pacific Islander: The remaining six respondents (or 2.8%) did not answer or could not be placed in any of the categories used (see Table 4). 45 Table 4 Psychotherapists’ Ethnic/Racial Background Background Number Percent Caucasian 181 87.5 Black 10 4.8 Mexican American 8 3.9 Asian/Pacific Islander 2 1.0 Other/unknown 6 2.8 Total 207 100.0 0f the survey respondents, 160 (or 77%) indicated they were currently conducting therapy with clients, and 46 (22%) indicated they were not. One respondent (about .6%) failed to answer the question (see Table 5). Many of those who reported they were not currently engaged in conducting therapy indicated on the margins of the questionnaire, they were retired or were currently engaged in administrative duties. Table 5 Psychotherapists’ Current Activity Level Active/Nonactive Number Percent Conducting therapy 160 77.2 Not conducting therapy 46 22.2 Unknown 1 .6 Total 207 100.0 46 One hundred forty-three of the 207 clinicians (69%) indicated they worked primarily with adults, and 37 (17.8%) reported their primary work was with children. Another 16 respondents (7.7%) indicated they worked with both children and adults, and another 11 (5.5%) chose not to answer (see Table 6). Table 6 Psychotherapists’ Primary Clients Primary Client Number Percent Adults 143 69.0 Children 37 17.8 Both children and adults 16 7.7 Unknown 11 5.5 Total 207 100.0 In terms of working with Mexican American clients, 114 (or 55%) indicated that 10% or less of their clients were Mexican American. Fifty-eight (28%) respondents stated that Mexican Americans comprised ll-30% of their client load, and 16 (or 7.7%) indicated Mexican Americans made up 31-50% of their client rosters. Only seven respondents (3.4%) reported 51-70% Mexican American clients, and only four (or 1.9%) indicated 71-90% Mexican American clients. One clinician, accounting for less than 1%, reported an over-90% Mexican American caseload, and seven psychotherapists (or 3.4%) did not respond to the question (see Table 7). 47 Table 7 Percentage of Mexican American Clients Client Percentage Number Percent 0-10% Mexican American 114 55.0 11-30% Mexican American 58 28.0 31-50% Mexican American 16 7.7 51-70% Mexican American 7 3.4 71-90% Mexican American 4 1.9 90% + Mexican American 1 .6 Unknown 7 3.4 Total 207 100.0 Other demographic information in the questionnaire included the number of Mexican American clients the clinician had seen within the last year. Of the respondents, 127 (61.7%) indicated ten or fewer Mexican American clients within the last year. Thirty-three (16%) indicated 11-20 Mexican American clients, and 19 (9.2%) reported 21— 40 Mexican American clients over the course of one year. Nine clinicians (4.3%) reported 41-60 Mexican American clients, six (or 2.9%) indicated 61-80, one (less than 1%) reported 81-100, and three (1.4%) of the clinicians reported seeing more than 100 Mexican American clients within one year. Finally, eight (less than 4%) of the psychotherapists chose not to answer this item (see Table 8). 48 Table 8 Number of Mexican American (MA) Clients Within the Last Year Number of MA Clients Number Percent 0-10 MA clients 127 61.7 11-20 MA clients 33 16.0 21-40 MA clients 19 9.2 41-60 MA clients 9 4.3 61-80 MA clients 6 2.9 81-100 MA clients 1 .6 100 + MA clients 3 1.4 Unknown 8 3.9 Total 207 100.0 Respondents were also asked how many years of experience they had working with Mexican American clients. Fifteen (7.4%) of the responding psychotherapists indicated no experience, 22 (or 10.6%) reported less than a year’s experience, and 32 (15.4%) reported less than 3 years’ experience working with Mexican American clients. Another 18 clinicians (8.8%) had less than 5 years’ experience, 34 (or 16.4%) of the respondents had less than 8 years, and 54 (26%) had less than 15 years’ experience. Twenty-seven (13%) clinicians reported more than 15 years of experience working with Mexican American clients, and five (2.4%) did not respond to the question (see Table 9). 49 Table 9 Years of Experience Working With Mexican American (MA) Clients Years of Experience Number Percent No experience with MA clients 15 7.4 < l year’s experience with MA clients 22 10.6 < 3 years’ experience with MA clients 32 15.4 < 5 years’ experience with MA clients 18 8.8 < 8 years’ experience with MA clients 34 16.4 < 15 years’ experience with MA clients 54 26.0 15 years + experience with MA clients 27 13.0 Unknown 5 2.4 Total 207 100.0 The final two questions of the questionnaire both used a 7-point Likert scale. For the first question, clinicians were asked to rate their familiarity with the Mexican American culture. A numerical value of 1 (Hi the Likert scale indicated "very unfamiliar," and a numerical value of 7 indicated "very familiar." 0f 'the 203 clinicians who answered this question, five (2.4%) indicated a "1." Twenty-four (11.6%) reported "2" on the scale, and 19 (or 9.2%) indicated a "3." A "4" was endorsed by 28 (13.5%) of the respondents, and 66 (almost 32%) indicated a "5" on the scale of cultural familiarity. Forty-seven respondents (22.7%) reported a "6," and 14 (or 6.8%) marked a "7" or were "very familiar" with Mexican American culture. Finally, four (1.9%) of the respondents did not answer the question (see Table 10). 50 Table 10 Psychotherapists’ Reported Familiarity With the Mexican American Culture Level of Familiarity Number Percent 1--Very unfamiliar 5 2.4 2-- 24 11.6 3-- 19 9.2 4--Familiar 28 13.5 5-- 66 31.9 6-- 47 22.7 7--Very familiar 14 6.8 Unknown 4 1.9 Total 207 100.0 Using the same 7-point Likert scale, 87 (or 42%) of the 207 responding clinicians indicated a "1" or "no fluency" in the Spanish language. Forty-six (or 22%) of the respondents endorsed a "2," and 23 (or 11.1%) of the respondents marked a "3." Fifteen clinicians (7.2%) endorsed a "4," which is the middle mark of the Likert scale, and 19 (or 9.4%) endorsed a "5." Five clinicians (2.4%) indicated a '6," and nine (4.3%) indicated they' were "very fluent" in the Spanish language or, in other words, a "7." Finally, three (1.4%) of the respondents failed to answer the question (see Table 11). 51 Table 11 Psychotherapists’ Reported Fluency in Spanish Level of Fluency in Spanish Number Percent l--No fluency in Spanish 87 42.0 2-- 46 22.2 3-- 23 11.1 4--Some fluency in Spanish 15 7.2 5-- 19 9.4 6-- 5 2.4 7--Very fluent in Spanish 9 4.3 Unknown 3 1.4 Total 207 100.0 Of the six different vignettes, 17 psychologists and 15 social workers (15.5% of responding clinicians) responded to vignette MA High. Thirteen psychologists and 14 social workers (or 13%) responded to vignette AA High, and 29 psychologists and 15 social workers (21.2%) responded to vignette MA Low. Another 15 psychologists and 19 social workers (16.4% of the total 207 respondents) responded to vignette AA Low. Responding to vignette NA High were 15.5% of the respondents, comprising 18 psychologists and 14 social workers. Finally, 19 psychologists and 19 social workers (18.4%) answered the questionnaire on the basis of vignette NA Low (see Tables 12 and 13). 52 Table 12 Response to Vignettes by Profession Vignette By Profession Number Vignette MA High 17 psychology 15 social work 32 Vignette AA High 13 psychology 14 social work 27 Vignette MA Low 29 psychology 15 social work 44 Vignette AA Low 15 psychology 19 social work 34 Vignette NA High 18 psychology 14 social work 32 Vignette NA Low 19 psychology 19 social work 38 Total 111 psychology 96 social work 207 Table 13 Total Response Percentages for Each Vignette Vignette Number Percent Vignette MA High 32 15.5 Vignette AA High 27 13.0 Vignette MA Low 44 21.2 Vignette AA Low 34 16.4 Vignette NA High 32 15.5 Vignette NA Low 38 18.4 Total 207 100.0 53 From the figures above, it is also true that, of the 207 returned questionnaires, 91 (or 44%) were in response to a "high" vignette, and 116 (or 56%) were in response to a "low" vignette (see Table 14). In addition, 76 respondents (37%) responded to vignettes with a Mexican American client, and 61 (or 29.5%) responded to the Anglo American vignettes. Finally, 70 respondents (33.5%) responded to the ethnically neutral vignettes (see Table 15). Table 14 Psychotherapists’ Response to "High" and "Low" Vignettes Vignette Number Percent All "high" vignettes 91 44.0 All "low" vignettes 116 56.0 Total 207 100.0 Table 15 Psychotherapists’ Response to "Ethnic" Vignettes Vignette Number Percent Vignette with MA client 76 37.0 Vignette with AA client 61 29.5 Ethnically neutral vignette 70 33.5 Total 207 100.0 54 The data presented thus far indicate there was little difference between the number of male and female psychotherapists who responded to the questionnaire. Although the questionnaire was distributed to equal numbers of social workers and psychologists, 17 more psychologists responded than did social workers. The difference between the responses of the professions was evident for Vignette MA Low but not particularly so for the other vignettes. The data also show that few Black and Mexican American psychothera- pists responded (ten and eight, respectively). Further, the majority of all respondents were currently involved in therapy and worked primarily with adults. Although the majority of responding psychotherapists worked with few Mexican American clients, a number of clinicians had a significant percentage of Mexican American clients and had worked with a large number of Mexican American clients within the last year. Further, more than half of the responding clinicians had six or more years’ experience working with Mexican American clients. Importantly, 60% of the respondents considered themselves more than familiar with Mexican American culture. Relatively few, however, claimed to be fluent in the Spanish language (about 25%). To summarize the data further, there were 32 to 38 responses to four of the vignettes. Vignette AA High, however, had only 27 responses, and vignette MA Low had significantly more responses than any of the others (44). Overall, there were 25 more responses to the "low" vignettes than to the "high" vignettes. Further, there 55 were 15 more responses to a vignette with a Mexican American client than with an Anglo American client. Besglts of the Statistical Analyses The statistical analyses described below were conducted to test each of the null hypotheses stated in Chapter III. The analyses produced results of significance or nonsignificance, which led to rejection of or failure to reject the null hypotheses. The following are the results of the statistical analyses used to test each of the seven hypotheses. flypothesis 1: There is no difference in how psychotherapists perceive the level of assertiveness between Mexican Americans and Anglo Americans. To test this and other hypotheses, correlation coefficients were obtained for all of the dependent and independent variables (see Table 16). Results of the analysis using the Pearson method revealed significance at the .05 level of significance for a number of variables. Pertinent to Hypothesis 1, the following variables were found to be significantly correlated with the dependent variable assertiveness: (a) psychological adjustment, (b) self- disclosure, (c) locus of control, (d) conducting therapy, (e) evidence, and (f) assertiveness with vignette. The correlation between assertiveness and self-disclosure and locus of control is apparent as all three dependent variables were presented together in the vignettes. The finding that there was a correlation between assertiveness and evidence, vignette, and psychological adjustment is also logical and of no surprise. 56 ._0.v my: .00.v. 0r - 0000. .000. 00. 00. 00.- 50. 50. 00.- 00.- 00. 50. 00. 50.- «:50. «:55. .000. 0300. ouuoco.> .0— - 00.- .0.- 00.- 50.- 00.- 50.- .5.- 0..- 00.- 00. 00. 00.- 5.. 00.- 00. «0.. 50.0.0000 .5. - 50. 0.. 50. 00. «0.. 05.. .0. 00. 00. 50.- 5.. «500. 0:00. 0000. «:50. oucou_>m .0— - «:00. 0:05.- 0000. 000—. «I00. 00.- 00. 00.- 00.- 05. 00. 05.- 50.- 00.- 500°050 00.0.00 .0. a¢05.- «5.0. «:05. 0:50. 00. mo. 00.- 00. 0.. 50. 00. _5. 0.. 50......500 0.00.00 .0— - 000..- __.- 0.55.- 50. 00.- no. 0005.. 00. .0.- 00. 50. 00. 0000000000 0.000.50000 .0. - «:00. .000. 00. 00.- 00.- 00.. .«05. 50. 50.- 00. 00. 0000.50 vucoou< .N .- ucoEumanuo —oowmo—ocoxom .— 0. 5_ 0. m_ 0. m_ 5. .p 0_ 0 0 5 0 m 0 m 5 _ o.0..a.> mo_no.co> ucoucoaopc. vco ucopcoaoo mcoE< 0co.uo_occou couco-oco~ w— upper "auc05055uoou covuo_occoo 57 However, assertiveness was also found to be significantly correlated with conducting therapy. This finding suggests that there was a correlation between the level of assertiveness and the psychothera- pists who currently were conducting therapy. Although these correlation results provided some general information, it was necessary to analyze them further to determine the proportion of variance in assertiveness. The data were subjected to further analysis using stepwise multiple regression. Stepwise multiple regression analysis was conducted for each of the dependent variables with all independent variables. For the dependent variable assertiveness, stepwise multiple regression analysis involved entering the independent variable evidence in step 1. As a result, a significant E of .0000 was found. Results showed no significant .E’s for any of the remaining independent variables. Overall, the variable evidence accounted for 47% of the total variance in assertiveness (see Table 17). The multiple regression results determined that the independent variable evidence significantly contributed to the proportion of variance of assertiveness and the other three dependent variables, but the contribution of the independent variable ethnicity was minimal. 'The independent variables ethnicity and evidence were therefore combined and labeled vignette. For the two-way analyses of variance (ANOVAs) used in this study, the four dependent variables, including assertiveness, were analyzed by the independent variable vignette and all other independent variables. 58 0000. - m 00 .000 00000.005 - m 0505m.~0 05000.55mw em, 0000000“ 0mmpm.m 00000 00000000 0000~.500~ 0000~.500~ F 00_0000m0m 00050. 000000 m umumznv< 00050. 000000 0 00000. 0 05000500 000000>m up 0000 00 0000000 00000000> mmgascm 000: mmgmscm mo 50m Hm <>oz< 005000000 00000000000 00000 550 50 mucmuw>m 000 00000>005000< "mupammm cowmmmcmmz 0000000: 50 mpnmh 59 The results of the two-way ANOVAs indicated vignette to be significant at less than the .05 level of significance in every analysis involving assertiveness. When analyzed with assertiveness and vignette, neither gender nor theoretical orientation was found to be significant. Results showed, however, that there was a two- way interaction with profession and vignette when these two variables were analyzed with assertiveness. Also, when analyzed with assertiveness, conducting therapy was significant at less than the .05 level of significance (see Table 18). The independent variables child/adult and number of MA clients were not found to be significant when analyzed with assertiveness and vignette. Further, percent of MA clients was not found to be significant, but there was a two-way interaction with vignette at the .05 level of significance when it was analyzed with assertiveness. There was also a two-way interaction with vignette when experience with MA clients was analyzed with assertiveness. This was also at the .05 level of significance (see Table 19). The remaining independent variables--race/ethnic background, culture familiarity, and Spanish fluency--were also analyzed with vignette and assertiveness. None was found to be significant, and there were no two-way interactions (see Table 20). Based on the results of the two-way ANOVAs, Neuman-Keuls post- hoc comparison tests were used to find the specific differences between the six vignettes. A Neuman-Keuls post-hoc analysis compared how the responding psychotherapists differed in their responses to the vignettes along the dimension of the dependent 60 Table 18 Results of Two-Way ANOVAs: Assertiveness by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy Source of Variation a: E Sig. of E Assertiveness by Vignette and Gender Gender 1 2.083 .151 Vignette 5 40.402 .000** Gender x vignette 5 .898 .484 Assertiveness by Vignette and Profession Profession 3 1.509 .214 Vignette 5 40.171 .000** Profession x vignette 10 1.924 .044* Assertiveness by Vignette and Theoretical Orientation Theoretical Orientation 8 1.749 .091 Vignette 5 38.979 .000** Theoretical orientation x vignette 32 1.208 .223 Assertiveness by Vignette and Conducting Therapy Conducting Therapy 2 6.446 .002* Vignette 5 40.208 .000** Conducting therapy x vignette 5 1.848 .105 *p < .05. **p < .01. 61 Table 19 Results of Two-Way ANOVAs: Assertiveness by Vignette and Child/Adult, Percent of MA Clients, Number of MA Clients, and Experience With MA Clients Source of Variation g: E Sig. 0f E Assertiveness by Vignette and Child/Adult Child/adult 3 1.338 .263 Vignette 5 39.293 .000** Child/adult x vignette 13 1.451 .140 Assertiveness by Vignette and Percent of MA Clients Percent of MA clients 6 1.053 .393 Vignette 5 39.408 .000** Percent of MA clients x vignette 17 1.846 .026* Assertiveness by Vignette and Number of MA Clients Number of MA clients 7 1.511 .165 Vignette 5 37.995 .000** Number of MA clients x vignette 21 .919 .566 Assertiveness by Vignette and Experience With MA Clients Experience with MA clients 7 .825 .568 Vignette 5 40.571 .000** Experience with MA clients x vignette 32 1.767 .012* *p < .05. **p < .01. 62 Table 18 Results of Two-Way ANOVAs: Assertiveness by Vignette and Race/Ethnic Background, Culture Familiarity, and Spanish Fluency Source of Variation at E Sig. of E Assertiveness by Vignette and Race/Ethnic Background Race/ethnic background 1 1.570 .212 Vignette 5 37.607 .000* Race/ethnic background x vignette 5 1.084 .371 Assertiveness by Vignette and Culture Familiarity Culture familiarity 7 .870 .532 Vignette 5 37.833 .000* Culture familiarity x vignette 32 1.285 .159 Assertiveness by Vignette and Spanish Fluency Spanish fluency 7 1.537 .158 Vignette 5 37.430 .000* Spanish fluency x vignette 30 1.078 .369 *p < .01. 63 variable assertiveness. Results determined there was a significant difference in how the psychotherapists responded to the MA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. Further, psychotherapists responded significantly differently to the AA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette in terms of overall level of assertiveness. Finally, there were significant differences in how the psychotherapists responded to the NA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. No other significant differences were found in this analysis (see Table 21). Table 21 Results of Neuman-Keuls Post-Hoc Analysis: Assertiveness by Vignette Mean Group Grl Gr2 Gr5 Gr3 Gr6 Gr4 19.5000 Grl 20.0370 Gr2 21.3438 Gr5 26.4773 - Gr3 * * * 27.0000 Gr6 * * * 27.5588 Gr4 * * * ote: Grl denotes MA High vignette. Gr2 denotes AA High vignette. Gr3 denotes MA Low vignette. Gr4 denotes AA Low vignette. Gr5 denotes NA High vignette. Gr6 denotes NA Low vignette. *Denotes pairs of groups significantly different at the .05 level of significance. 64 In conclusion, after testing for a relationship between ethnicity and level of assertiveness, it was found that no significant relationship existed. Neither the stepwise multiple regression analysis nor the subsequent two-way ANOVAs or Neuman- Keuls analysis produced results of significance which would indicate psychotherapists perceived a different level of assertiveness between Anglo Americans and Mexican Americans. As such, Hypothesis 1 (null form) cannot be rejected. Failure to reject this hypothesis indicates that psychotherapists in this study did not perceive a difference in the level of assertiveness between Mexican Americans and Anglo Americans, as measured by the instruments used in the study. Hyppthesis 2: There is no difference in how psychotherapists perceive the level of self-disclosure between Mexican Americans and Anglo Americans. As noted earlier, correlation coefficients were obtained for all dependent and independent variables. Pertinent to Hypothesis 2, the following variables were found to be significant with the dependent variable self-disclosure (see Table 16): (a) locus of control, (b) conducting therapy, (c) evidence, and (d) vignette. As with Hypothesis 1, the correlations between self-disclosure and locus of control, evidence, and vignette are intuitively logical. However, the data suggested there was a relationship between self-disclosure and psychotherapists who were currently conducting therapy, a finding that was not expected. 65 These data were subjected to further analysis using stepwise multiple regression. For the dependent variable self-disclosure, stepwise multiple regression analysis involved entering the independent variable evidence in step 1. As a result, a significant E of .0000 was found. In addition, significant E’s were found for the independent variables conducting therapy and number of MA clients. Step 2 then involved entering the independent variable conducting therapy. As a result, a significant E of .0000 was found, but no significant E’s were found for the remaining independent variables. The overall results of the multiple regression analysis indicate that evidence accounted for 25% of the variance in self-disclosure. Also, less significantly, conducting therapy accounted for only 2% of the variance in self-disclosure (see Table 22). As noted earlier, the four dependent variables including self— ' disclosure were analyzed by the independent variable vignette and all other independent. variables. The results of these two-way ANOVAs indicated vignette to be significant at less than the .05 level of significance in every analysis involving self-disclosure. When analyzed with self-disclosure and vignette, neither gender, profession, nor theoretical orientation was found to be significant. However, conducting therapy was found to be significant at the .05 level of significance (see Table 23). 66 0000. u 0 00 .000 05000.00 - 0 05000.0 05055.0050 000 00000000 00000.0 00000 00000000 00000.000 00500.000 0 0000000000 00000. 000000 0 00000000 00050. 000000 0 00000. 0 00000000 000000» 0000000000 "0 000m 00 0000000 00000000> 0000. - 0 00 .000 00000.00 - 0 00005.0 00000.0050 000 00000000 00000.0 00000 00000000 00000.000 00000.000 0 0000000000 00000. 000000 0 00000000 00000. 000000 0 00000. 0 00000000 00:000>m 00 0000 :0 0000000 00000000> 0000000 000: 0000000 00 E00 00 <>oz< 00000000> 00000000000 00:00 00< 00 00000000 0000000000 0000 00000000 000 00000>000000< 00000000 0000000000 0000000: mm 00000 67 Tabie 23 Resu1ts of Two-Hay ANOVAs: Se1f-Disc1osure by Vignette and Gender, Profession, Theoretica1 Orientation, and Conducting Therapy Source of Variation Qfi F S1'9. 0f E Se1f-Disc1osure by Vignette and Gender Gender 1 3.176 .076 Vignette 5 15.498 .000** Gender x vignette 5 .497 .778 Se1f-Disc1osure by Vignette and Profession Profession 3 .109 .955 Vignette 5 14.516 .000** Profession x vignette 10 .969 .472 Se1f-Disc1osure by Vignette and Theoretica1 Orientation Theoretica1 orientation 8 1.492 .164 Vignette 5 15.570 .000** Theoretica1 orientation x vignette 32 1.305 .145 Se1f-Disc1osure by Vignette and Conducting Therapy Conducting therapy 2 3.552 .031* Vignette 5 15.027 .000** Conducting therapy x vignette 5 1.698 .137 *p < .05. **p < .01. 68 The resu1ts of the two-way ANOVAs invo1ving se1f-disc1osure, vignette, and the other independent variab1es indicated no significance for chi1d/adu1t, percent of MA c1ients, or experience with MA c1ients. However, the independent variab1e number of MA clients was found to be significant at the .05 1eve1 of significance. There were no two-way interactions for any of these variab1es, however (see Tab1e 24). Fina11y, the resu1ts of the two-way ANOVAs that inc1uded race/ethnic background, cu1ture fami1iarity, and Spanish f1uency indicated no significance except for cu1ture fami1iarityu Again, there were no two-way interactions (see Tab1e 25). The 1ast statisticai ana1ysis to test Hypothesis 2 was the Neuman-Keuis post-hoc comparison test. This ana1ysis was used to compare how the responding psychotherapists differed in their responses to the vignettes a1ong the dimension of se1f-disc1osure. The resu1ts of this post-hoc ana1ysis were the same as those from the Neuman-Keu1s described ear1ier. Results determined there was a significant difference in how the psychotherapists responded to the MA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. Further, psychotherapists responded significant1y different1y to the AA High vignette versus the MA Low vignette, and AA Low vignette, and the NA Low vignette in terms of overa11 1eve1 of se1f—disc1osure. Fina11y, there were significant differences in how the psychotherapists responded to the NA High vignette versus the MA Low vignette, the AA Low vignette, and 69 Tab1e 24 Resu1ts of Two-Nay ANOVAs: Se1f-Disc105ure by Vignette and Chi1d/Adu1t, Percent of MA C1ients, Number of MA C1ients, and Experience With MA C1ients Source of Variation g: E Sig. of E Se1f—Disclosure by Vignette and Chi1d/Adu1t Chi1d/adu1t 3 1.373 .252 Vignette 5 15.253 .000** Chi1d/adu1t x vignette 13 1.410 .158 Se1f-Disc1osure by Vignette and Percent of MA c1ients Percent of MA c1ients 6 1.447 .199 Vignette 5 15.183 .000** Percent of MA c1ients x vignette 17 1.241 .237 Se1f—Disc1osure by Vignette and Number of MA c1ients Number of MA c1ients 7 2.625 .013* Vignette 5 16.967 .000** Number of MA c1ients x vignette 21 1.121 .331 Se1f—Disc1osure by Vignette and Experience Hith MA CTients Experience with MA c1ients 7 1.094 .369 Vignette 5 15.188 .000** Experience with MA c1ients x vignette 32 1.178 .251 *p < .05. **Q < .01. 70 Tab1 Resu1ts of Two-Nay ANOVAs: e 25 Se1f-Disc1osure by Vignette and Race/Ethnic Background, Cu1ture Fami1iarity, and Spanish F1uency Source of Variation g: E Sig. of E Se1f-Disc1osure by Vignette and Race/Ethnic Background Race/ethnic background 1 .240 .625 Vignette 5 15.405 .000** Race/ethnic background x vignette 5 2.156 .061 Se1f—Disc1osure by Vignette and Cu1ture Fami1iarity Cu1ture fami1iarity 7 2.249 .033* Vignette 5 15.850 .000** Cu1ture fami1iarity x vignette 32 1.180 .249 Se1f-Disciosure by Vignette and Spanish F1uency Spanish f1uency 7 1.307 .250 Vignette 5 15.193 .000** Spanish f1uency x vignette 30 1.426 .085 *2 < .05. **p < .01. 71 the NA Low vignette. No other significant differences were found in this ana1ysis (see Tab1e 26). Tab1e 26 Resu1ts of Neuman-Keu1s Post-Hoc Ana1ysis: Se1f-Disc1osure by Vignette Mean Group GrZ Gr5 Gr1 Gr3 Gr6 Gr4 13.6667 Gr2 13.9063 Gr5 14.4063 Gr1 17.5682 Gr3 * * * 17.6053 Gr6 * * * 18.6053 Gr4 * * * Note: Gr1 denotes MA High vignette. Gr2 denotes AA High vignette. Gr3 denotes MA Low vignette. Gr4 denotes AA Low vignette. Gr5 denotes NA High vignette. Gr6 denotes NA Low vignette. *Denotes pairs of groups significant1y different at the .05 1eve1 of significance. In conc1usion, the statistica1 analyses described above indicate that no significant re1ationship between ethnicity and se1f-disc1osure was found. Neither the stepwise mu1tip1e regression ana1ysis nor the subsequent two-way ANOVAs and Neuman-Keu1s post-hoc analyses produced resu1ts of significance that wou1d indicate psychotherapists perceived different 1eve1s of se1f-disc1osure between Mexican Americans and Ang1o Americans. As such, Hypothesis 2 (nu11 form) cannot be rejected. 'The fai1ure to reject this 72 hypothesis indicates that responding psychotherapists did not perceive a difference in the level of self-disclosure between Mexican Americans and Anglo Americans, as measured by the instruments used in this study. Hypothesis 3: There is no difference in how psychotherapists perceive LOC orientation between Mexican Americans and Anglo Americans. Review of the correlation coefficient table (Table 16) indicates a number of variables that were significant with the dependent variable locus of control. These variables were (a) conducting therapy, (b) percent of MA c1ients, (c) evidence, and (d) vignette. The data point to the logical relationship between locus of control and evidence and vignette. Of special interest is the relationship between locus of control and therapists who were currently engaged in conducting therapy. Further, the data suggest that a significant relationship existed between self-disclosure and the percentage of Mexican American clients to whom the responding psychotherapists had been providing services. As with the testing of the previous hypotheses, the data were subjected to stepwise multiple regression. For the dependent variable locus of' control, the procedure involved entering the independent variable evidence in step 1. As a result, a significant E of .0000 was produced. Step 1 also produced a significant E for the independent variable ethnicity. Step 2 of the analysis involved entering the independent variable ethnicity, and the result 73 indicated a significant 5 of .0000 but no significant {’5 for any of the other remaining independent variables. Overall, results of the multiple regression analysis indicate that evidence accounted for 32% of the variance in locus of control, but ethnicity accounted for only about 1% of the variance in the same variab1e (see Table 27). Two-way ANOVAs were used to analyze the data further. The results of the two-way ANOVAs involving locus of control, vignette, and the other independent variables indicated vignette to be significant at less than the .05 level of significance in every analysis. 'The results also showed no significant findings for gender or profession. There were, however, significant findings for theoretical orientation and conducting therapy. Both of these findings were significant at the .05 level of significance. No significant two-way interactions were found (see Table 28). Results of other two-way ANOVAs also determined no significance for the independent variables child/adult, percent of MA clients, number of MA clients, and experience with MA clients. Further, there were no two-way interactions with vignette for any of these variables (see Table 29). Finally, results of the last set of two-way ANOVAs using vignette and locus of control indicated no significant findings for race/ethnic background, culture familiarity, and Spanish fluency. In addition, there were no two-way interactions for these variables (see Table 30). 74 coco. u m co .mwm mmeom.o¢ a a epmmm.mp Amm-.memm mm, Pasupmag oapme.e coccm ccmucaum GNPoe.mom ammom.opmp N cowmmmcmmm 05mm”. acaaam m nocmahu< ~ommm. mcmzam m hmmfim. m «Parapzz xpwu_cgum "N amum cw umgmucm mmpnmwgm> coco. . u co .mwm unemo.om . m “memm.mp mcmom.F~mm amp Pazuwmam mo~m¢.¢ coccm ucaucaum mem~¢.~.NP mNmN¢.~FN. _ cosmmacmmm mmapm. acmzcm m Dagmaflu< mompm. acaaam m mmamm. m mpawapaz mucmcw>m up amum cw vmcmucm mmpnmwgm> mmgmaam cam: mmcmacm we saw mu <>oz< mm_n~_c~> “cavemamucm cacao ._< »n Axa_umccpm nemv mucmu_>m new Pocacou we mzuod "mapzmmm cowmmmcmmm mpawupzz KN mpnm» 75 Table 28 Results of Two-Nay ANOVAs: Locus of Control by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy Source of Variation g: E Sig. of E Locus of Control by Vignette and Gender Gender 1 3.241 .073 Vignette 5 20.317 .000** Gender x vignette 5 .532 .752 Locus of Control by Vignette and Profession Profession 3 .934 .425 Vignette 5 19.447 .000** Profession x vignette 10 .971 .470 Locus of Control by Vignette and Theoretical Orientation Theoretical orientation 8 2.320 .022* Vignette 5 24.536 .000** Theoretical orientation x vignette 32 1.464 .066 Locus of Control by Vignette and Conducting Therapy Conducting therapy 2 3.950 .021* Vignette 5 19.817 .000** Conducting therapy x vignette 5 .848 .517 *p < .05. **p < .01. 76 Table 29 Results of Two-Hay ANOVAs: Locus of Control by Vignette and Child/Adult, Percent of MA Clients, Number of MA Clients, and Experience With MA Clients Source of Variation g: E Sig. of E Locus of Control by Vignette and Child/Adult Child/adult 3 .316 .813 Vignette 5 20.800 .000* Child/adult x vignette 13 1.769 .051 Locus of Control by Vignette and Percent of MA Clients Percent of MA clients 6 1.185 .316 Vignette 5 19.039 .000* Percent of MA c1ients x vignette 17 .962 .503 Locus of Control by Vignette and Number of MA Clients Number of MA clients 7 .781 .604 Vignette 5 17.646 .000* Number of MA c1ients x vignette 21 .749 .777 Locus of Control by Vignette and Experience Nith MA Clients Experience with MA clients 7 .792 .595 Vignette 5 18.020 .000* Experience with MA c1ients x vignette 32 .865 .677 *p < .01. 77 Table 30 Results of Two-Nay ANOVAs: Locus of Control by Vignette and Race/Ethnic Background, Culture Familiarity, and Spanish F1uency Source of Variation d: E Sig. of E Locus of Control by Vignette and Race/Ethnic Background Race/ethnic background 1 .237 .627 Vignette 5 19.987 .000* Race/ethnic background x vignette 5 1.117 .352 Locus of Control by Vignette and Culture Familiarity Culture fami1iarity 7 1.807 .089 Vignette 5 19.726 .000* Culture familiarity x vignette 32 1.180 .249 Locus of Control by Vignette and Spanish Fluency Spanish f1uency 7 1.055 .395 Vignette 5 20.991 .000* Spanish f1uency x vignette 30 1.215 .220 *p < .01. 78 To test Hypothesis 3 further, the Neuman-Keuls post-hoc comparison test was incorporated. The results of this post-hoc test were the same as those found in previous hypothesis testing. Results determined there was a significant difference in how the psychotherapists responded to the MA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. Further, psychotherapists responded significantly differently to the AA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette in terms of overall locus of control orientation. Finally, there were significant differences in how the psychotherapists responded to the NA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. No other significant differences were found in this analysis. Overall, it has been shown that, after testing for a relationship between ethnicity and locus of control orientation, no significant relationship was found. Neither the stepwise multiple regression analysis nor the subsequent two-way ANOVAs or the Neuman- Keuls analyses produced results of significance that would indicate psychotherapists perceived different levels of locus of control orientation between Mexican Americans and Anglo Americans. As such, Hypothesis 3 (null form) cannot be rejected. Failure to reject this hypothesis indicates that psychotherapists did not perceive a difference in the level of locus of control orientation between Mexican Americans and Anglo Americans. 79 Table 31 Results of Neuman-Keuls Post-Hoc Analysis: Locus of Control by Vignette Mean Group Grl Gr2 Gr5 Gr3 Gr6 Gr4 18.2500 Grl 19.5556 Gr2 19.8750 Gr5 24.5455 Gr3 * * * 26.1579 Gr6 * * * 26.5588 Gr4 * * * Note: Grl denotes MA High vignette. Gr2 denotes AA High vignette. Gr3 denotes MA Low vignette. Gr4 denotes AA Low vignette. Gr5 denotes NA High vignette. Gr6 denotes NA Low vignette. *Denotes pairs of groups significantly different at the .05 level of significance. H at is 4: There is no difference in how psychotherapists perceive psychological adjustment between Mexican Americans and Anglo Americans, based on level of assertiveness. Hypothesis 5: There is no difference in how psychotherapists perceive psychological adjustment between Mexican Americans and Anglo Americans, based on the level of self-disclosure. H ot is 6: There is no difference in how psychotherapists perceive psychological adjustment between Mexican Americans and Anglo Americans, based on LOC orientation. Hypotheses 4, 5, and 6 were, to a large extent, contingent on the findings of the first three hypotheses. If psychotherapists had perceived ethnic differences of assertiveness, self-disclosure, and/or locus of control orientation, the next question was whether these differences had an effect on the psychotherapists’ judgments 80 of psychological adjustment. Because there were no significant findings for the first three hypotheses, it was not expected there would'be significant findings for Hypotheses 4, 5, and 6. Nonetheless, the statistical analyses used to test the first three hypotheses were done simultaneously and were also used to test Hypotheses 4, 5, and 6. The following results of testing Hypotheses 4, 5, and 6 are presented together for practical purposes and for ease of understanding. Correlation coefficients were obtained for the dependent variable psychological adjustment using the Pearson method. Results of this analysis revealed significance at the .05 level of significance for a number of variables. These variables were (a) assertiveness, (b) self-disclosure, (c) locus of control, (d) evidence, (e) ethnicity, and (f) vignette. The procedure for the stepwise multiple regression analysis involving the dependent variable psychological adjustment inc1uded entering the independent variable evidence in step 1. As a result, a significant E of .0000 was found. In addition, significant E’s for the independent variables gender, child/adult, and ethnicity were also found. For step 2 of the analysis, the independent variable ethnicity was entered in the formula and produced a significant E of .0000. Significant E’s were again discovered for the independent variables gender and adult/child. Step 3 of this analysis entered the independent variable gender and produced a significant F of .0000 for the independent variable adult/child. For the fourth and last step, the independent variable adult/child 81 was entered. As a result, a significant E of .0000 was found, but no significant E’s were found for the remaining independent variables. Overall results of this analysis indicated that evidence and ethnicity accounted for approximately 35% of the variance in psychological adjustment. The results also indicated that the variables gender and adult/child accounted for approximately 1.5% of the variance in the same variable (see Table 32). The results of the two-way ANOVAs indicated vignette to be significant at less than the .05 level of significance in every analysis involving psychological adjustment. Nhen analyzed with psychological adjustment and vignette, neither profession, theoretical orientation, nor conducting therapy was found to be significant. Gender, however, was found to be significant at the .05 level of significance. For these independent variables there were no two-way interactions (see Table 33). In addition, no other independent variables were significant when analyzed with psychological adjustment and vignette, nor were there any two-way interactions (see Tables 34 and 35). The last statistical analysis to test Hypotheses 4, 5, and 6 was the Neuman Keuls post-hoc comparison test. This analysis was used like the other post-hoc analyses, and the results were the same. Resu1ts determined there was a significant difference in how the psychotherapists responded to the MA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. Further, psychotherapists responded significant1y differently to the 82 coco. n m to .mmm mmmso.om . a om~m~.P~P Nuom¢.ompmm map P~=u_ma¢ mNmpo.P_ cocca ucmucapm oh-~.¢~oo ~mmm¢.mep~P N =o_mmmcmam woman. acaaam m caumaflc< thmm. mamas“ m seamm. m «Papaya: zupumcgum “N amum a? vmgmucm mmpnawcm> coco. u u co .m_m esoo~.~m u m “Noum.¢~_ mmomo.m~m- amp P~=u_mmm ~mm¢_._F Loccm newscaam amm~m.mmapp mmm~¢.moa__ P cowmmmcmmm amomm. acazam m umumzfiu< ommmm. mcaaam m mmehm. m m_a_upzz mucmu_>m up amam cw umgoucm mm—nmwg~> mmgaacm cam: mmcmscm we saw Hw <>oz< maFn~LL~> acaccaaaucfi cagpo _P< xn Au_=u<\upcgu ucm .Laucmu .au_upcsam ucmv mucmcm>m uca acmsumanc< pmuwmopoguxmm mm mpnmp "mapammm cowmmmcmwm mpawu—az 83 coco. n m co .mwm Noo-.m~ . a mwoom.opp Pm~m~.~mppm Pmp Pazcwmmm Neopm.op soggm ugaccmum Pvmmo.mm~m mommp.~mpmp v cowmmmgmmm omonm. menacm m umumznu< chmm. wgmsam m oempo. m «Pawppaz uP=n<\vpm;u "a amum cw vmgmucm mmpnm_cm> coco. u m co .mwm asmmm.mm u m m~eoo.m~_ popsm.mmmpm Nap Pmacwmmm mc~o¢.e Logcm ucavcmam msmwe.upxp ~me~o.mmo~p m cowmmmcmmm mammm. mcaaam m vmumsnc< nmmmm. menacm m mssoo. m «papupaz cmucmu um amam c? nmcmpcm mmpnm_c~> mmgaacm cam: mmcmsam mo Sam .Hu <>oz< emacwucou--~m apnah 84 Table 33 Results of Two-Hay ANOVAs: Psychological Adjustment by Vignette and Gender, Profession, Theoretical Orientation, and Conducting Therapy Source of Variation g: E Sig. of E Psychological Adjustment by Vignette and Gender Gender 1 5.493 .020* Vignette 5 24.724 .000** Gender x vignette 5 1.620 .157 Psychological Adjustment by Vignette and Profession Profession 3 1.238 .297 Vignette 5 23.477 .000** Profession x vignette 10 1.295 .236 Psychological Adjustment by Vignette and Theoretical Orientation Theoretical orientation 8 1.798 .081 Vignette 5 23.130 .000** Theoretical orientation x vignette 32 .816 .747 Psychological Adjustment by Vignette and Conducting Therapy Conducting therapy 2 .866 .422 Vignette 5 22.838 .000** Conducting therapy x vignette 5 .939 .457 *p < .05. **p < .01. 85 Table 34 Results of Two-Hay ANOVAs: Psychological Adjustment by Vignette and Child/Adult, Percent of MA C1ients, Number of MA Clients, and Experience Nith MA C1ients Source of Variation g: E Sig. of E Psychological Adjustment by Vignette and Child/Adult Child/adult 3 1.308 .273 Vignette 5 22.289 .000* Child/adult x vignette 13 .577 .871 Psychological Adjustment by Vignette and Percent of MA Clients Percent of MA clients 6 1.516 .332 Vignette 5 21.509 .000* Percent of MA clients x vignette 17 .831 .656 Psychological Adjustment by Vignette and Number of MA Clients Number of MA clients 7 .779 .605 Vignette 5 22.011 .000* Number of MA c1ients x vignette 21 .861 .641 Psychological Adjustment by Vignette and Experience Hith MA Clients Experience with MA clients 7 .415 .892 Vignette 5 21.552 .000* Experience with MA c1ients x vignette 32 .907 .614 *p < .01. 86 Table 35 Results of Two-Hay ANOVAs: Psychological Adjustment by Vignette and Race/Ethnic Background, Cu1ture Fami1iarity, and Spanish Fluency Source of Variation Qfi E 519- 0f E Psychological Adjustment by Vignette and Race/Ethnic Background Race/ethnic background 1 .227 .634 Vignette 5 22.681 .000* Race/ethnic background x vignette 5 .703 .622 Psychological Adjustment by Vignette and Culture Familiarity Cu1ture fami1iarity 7 .954 .467 Vignette 5 21.510 .000* Culture familiarity x vignette 32 .899 .627 Psychological Adjustment by Vignette and Spanish Fluency Spanish fluency 7 1.787 .093 Vignette 5 23.876 .000* Spanish fluency x vignette 30 1.334 .131 *p < .01. 87 AA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette in terms of overall psychological adjustment. Finally, there were significant differences in how the psychotherapists responded to the NA High vignette versus the MA Low vignette, the AA Low vignette, and the NA Low vignette. No other significant differences were found in this analysis (see Table 36). Table 36 Results of Neuman-Keuls Post-Hoc Ana1ysis: Psychological Adjustment by Vignette Mean Group Grl Gr5 Gr2 Gr3 Gr4 Gr6 50.7188 Grl 55.0625 Gr5 56.5185 Gr2 66.5682 Gr3 * * * 70.9118 Gr4 * * * 72.1316 Gr6 * * * Note: Grl denotes MA High vignette. Gr2 denotes AA High vignette. Gr3 denotes MA Low vignette. Gr4 denotes AA Low vignette. Gr5 denotes NA High vignette. Gr6 denotes NA Low vignette. *Denotes pairs of groups significantly different at the .05 level of significance. In summary, the results of the statistical analyses described above indicate the failure to reject the null forms of Hypotheses 4, 5, and 6. Neither the stepwise multiple regression analysis nor the subsequent two-way ANOVAs and Neuman-Keuls analyses provided results 88 of significance that would indicate psychotherapists perceived a difference in psychological adjustment between Mexican Americans and Anglo Americans based on level of assertiveness, self-disclosure, or locus of control orientation. : There is no difference in how psychotherapists perceive the NA High vignette client and the NA Low vignette client in terms of level of assertiveness, level of self- disclosure, and LOC orientation. Hypothesis 7 was included in the study to ensure that psychotherapists were making distinctions between the high and low levels of self-disclosure, assertiveness, and internal and external locus of control. It was thought that, if these distinctions were made, whatever differences occurred between the responses to the first four vignettes were related to the ethnicity of the described client. Given the results of the Neuman-Keuls post-hoc analyses described earlier, it was found that clinicians were indeed able to discriminate the different levels of assertiveness, self-disclosure, and locus of control when responding to the NA High vignette and the NA Low vignette. Therefore, Hypothesis 7 (null form) was rejected. The rejection of this hypothesis indicates the responding psychotherapists were able to discriminate between the two ”ethnically neutral" vignettes. Aggregate results of' the statistical analyses indicate the failure to reject Hypotheses 1 through 6. However, Hypothesis 7 was rejected. The failure to reject and the rejection of these hypotheses indicate that in) differences were found in the 89 perceptions of responding psychotherapists in the level of assertiveness, self-disclosure, or locus of control orientation between the Mexican American and Anglo American clients. Further, no differences were found in how responding clinicians perceived overa11 psychological adjustment between the two ethnic groups based on assertiveness, self-disclosure, and locus of control orientation. Finally, it was found that psychotherapists were able to discriminate between the two "ethnically neutral" vignettes in terms of the different levels of three of the dependent variables. In addition to these findings, the results of the statistical analyses also determined other pertinent findings, which are outlined in the next section. cher Pertinent Results .In addition to some of the pertinent findings described in the Description of Survey and Vignette Distribution section, the data lent themselves to a number of variables that are important in understanding the topic of interest. In particular, the results of the correlation coefficient analysis revealed significance for a number of variables not presented thus far. The variables found to be significant with one another are: 1. Gender with profession Gender with number of MA clients Gender with experience with MA c1ients Profession with conducting therapy (II-#00“) Profession with race/ethnic background 90 Theoretical orientation with percent of MA clients Conducting therapy with percent of MA clients Percent of MA clients with number of MA clients «:0on Percent of MA clients with experience with MA clients 10. Percent of MA clients with race/ethnic background 11. Percent of MA clients with culture fami1iarity 12. Percent of MA clients with Spanish f1uency 13. Percent of MA clients with evidence 14. Number of MA clients with experience with MA clients 15. Number of MA clients with culture familiarity 16. Number of MA clients with Spanish f1uency 17. Number of MA clients with evidence 18. Experience with MA clients with race/ethnic background 19. Experience with MA clients with culture fami1iarity 20. Experience with MA clients with Spanish f1uency 21. Race/ethnic background with culture familiarity 22. Race/ethnic background with Spanish f1uency 23. Culture familiarity with Spanish f1uency 24. Evidence with vignette 25. Ethnicity with vignette As the list above and previous results of’ the correlation coefficients indicate, all dependent and independent variables except child/adult had at least one significant correlation result with another dependent or independent variable. A large number of these correlations are logical, given the design of the study. 91 However, a number of these correlations provide some additional information not presented thus far. Much like the data described for Hypotheses 1, 2, and 3, a finding of a significant relationship between conducting therapy and another variable was found. In this case, the percentage of Mexican American clients was significantly correlated with whether a psychotherapist was currently engaged in direct therapy. In addition, percent of MA clients, which was found earlier to be related to locus of control, was also related to the theoretical orientation of the responding clinicians. Other significant relationships considered to be important include those between percent of MA clients with culture familiarity, Spanish fluency, and race/ethnic background. These data also suggest that race/ethnic background was related to the amount of experience psychotherapists reported they had in working with Mexican American clients. In addition, the level of experience with Mexican American c1ients was found to be related to gender, cu1ture familiarity, and Spanish f1uency. The last correlations of interest to the discussion concern the independent variable number of MA clients. The data indicate a significant relationship existed between the number of Mexican American clients a psychotherapist had seen within the last year and whether the clinician was female or male. Also, the number of Mexican American clients was related to the level of locus of control, self-disclosure, and assertiveness iri the vignettes 92 (evidence). These findings, in addition to those directly related to the hypotheses, are discussed in Chapter V. CHAPTER V DISCUSSION There has been little research effort in the past to determine the existence of differences between Anglo Americans and Mexican Americans. The review of the literature outlined in Chapter II indicated that research efforts regarding locus of control, assertiveness, and self-disclosure were mostly from the 19705 and early 1980s. Few efforts to elucidate potential ethnic similarities or differences are currently being conducted. This general neglect and decrease of research over the years is unfortunate as the world in which we live becomes increasingly multidimensional and influenced by so many different perspectives and cultural influences. The paucity of research is also particularly disturbing in light of the rapidly increasing growth of the Mexican American population residing in the United States, Mexican American underutilization rates of mental health treatment services, the overall adverse socioeconomic condition of many Mexican Americans, and the excessively high dropout rate of Chicano high school and college students. In addition to the effects this void of information has on the Mexican American population and the general American public, psychotherapists in particular are faced with the predicament of 93 94 service delivery to this population but with little data base on which to rely. Approximately half of the research cited in this study indicated there were significant differences between ethnic groups, whereas the other half determined there were no differences. Equivocal findings such as these tend to create confusion for individual psychotherapists. This lack of information has many other ramifications, but perhaps a most significant one is the absence of information that can counteract the stereotypic notions that have developed over the years. Stereotypic behaviors are all too often ascribed to members of both ethnic groups. We know that Mexican American stereotypes have been influenced by the early literature, which often was based on unrepresentative populations and whose findings were generalized to the entire ethnic group. Without a current, solid foundation of data, it is extremely difficult to dispel the myths and preconceived notions we may have of any culturally dissimilar individual. Because psychotherapists are human and part of the "greater culture” of American society, it would not be surprising if they are also influenced by stereotypic perceptions. Despite efforts to be objective and neutral in their clinical judgments, it is presumed they may be influenced as others are by unproven behavioral traits ascribed to different ethnic groups. More specifically, it would not be surprising if psychotherapists perceive Mexican Americans as more passive or unassertive than Anglo Americans, which is one stereotypic notion that has prevailed over the years. Nor would it be surprising if psychotherapists perceive differences of locus of 95 control orientation or self-disclosure tendencies, which have also been said to differ significantly between the two ethnic groups. 0f' course, psychotherapists are different from the general American public in what they do and, most important, how they can affect individual clients. Diagnosis, disposition, and recommendations from a psychotherapist may have a tremendous effect on the life, job, and future of a client. How a psychotherapist judges the psychological adjustment of a particular individual client will have a large effect on such diagnoses and dispositions. A fascinating question to consider is whether judgments made by psychotherapists are influenced by the same stereotypic perceptions many others may hold. Previous researchers in the area of interest have investigated ethnic differences by studying members of the two ethnic groups. To date, no investigators have researched ethnic differences as seen through the eyes of psychotherapists. The current study was designed to determine ethnic differences as seen by psychotherapists and to determine whether potential differences influence judgments of psychological adjustment. It is hoped that this research will encourage further examination of ethnic differences and perceptions of ‘those people» who have a large influence on their clients-- psychotherapists. The rest of this chapter is devoted to a review of the findings of this study and their implications. Further, limitations of the 96 study are presented. Fina11y, implications for research are discussed. Review of the Fin in s The results of the statistical analyses described in Chapter IV were the data-based foundation for the drawing of conclusions. No differences were found in Hypothesis 1 testing to indicate that psychotherapists in this study perceived an ethnic difference in the level of assertiveness between the two ethnic groups. This conclusion appears to support and extend previous research indicating no significant differences between Mexican Americans and Anglo Americans regarding assertiveness (Calvillo-Simpson, 1983; Grodner, 1975; Kagan & Carlson). This conclusion failed, however, to support previous research indicating significant differences between the same two groups (Hall & Beil-Harner, 1978; Hoppe et al., 1977; Kimble et al., 1984). As such, this conclusion has many implications for the research and therapeutic domains, which are discussed later. The conclusions drawn from testing of Hypothesis 2 suggest that psychotherapists perceived no differences of self-disclosure between Mexican Americans and Anglo Americans. This conclusion tends to support and extend previous research indicating no significant differences between the two groups in terms of self-disclosure (Borrego et al., 1982; Franco et al., 1984). The conclusion does not support the research in which differences were found between the 97 two groups (Acosta & Sheehan, 1978; Franco & Levine, 1985; Littlefield, 1974). The conclusion based on testing of Hypothesis 3 suggests no difference in how psychotherapists perceived locus of control orientation between Mexican Americans and Anglo Americans. This conclusion appears to support and extend previous research indicating no significant differences between the two groups in terms of locus of control (Buriel, 1982; Buriel & Rivera, 1980; Cole et al., 1978; Garza, 1977; Garza & Ames, 1974; Jessor et al., 1968; Knight et al., 1978). The conclusion fails, however, to support the findings of previous research that concluded there are differences between Mexican Americans and Anglo Americans (Chandler, 1979; Evans & Anderson, 1973; Farris & Glenn, 1976; Gaa et a1., 1981; Garza, 1977; Justin, 1970; Rogers, 1971; Ross et al., 1983; Scott & Phelan, 1969; Soza, 1982; Young & Shorr, 1986). The findings based on testing of Hypothesis 4 indicated that psychotherapists did not perceive a difference in psychological adjustment based on the level of assertiveness between Mexican Americans and Anglo Americans. As there does not exist a body of research in this area, this finding may have a number of implications for the therapeutic/treatment domain. The conclusions drawn from testing of Hypothesis 5 indicate that psychotherapists did not perceive a difference in psychological adjustment between Mexican Americans and Anglo Americans based on the level of self-disclosure. This, too, appears to be an important conclusion as there is no known body of research in this area. 98 The conclusion drawn from Hypothesis 6 testing suggests no differences in how psychotherapists perceived psychological adjustment between Mexican Americans and Anglo Americans based on locus of control orientation. Because there are no research data to refute~ or support this conclusion, it. may be another benchmark conclusion that has implications for future research. The last hypothesis, Hypothesis 7, was not rejected. Failure to reject this hypothesis indicates that psychotherapists were able to discriminate the different levels of locus of control orientation, self-disclosure, and assertiveness described in the vignettes used in this study. This conclusion supports the six previous conclusions. Further, this conclusion negates any contamination of the previous results in that the lack of differences in results relative to Hypotheses 1 through 6 could be attributed to lack of discriminant ability of the responding psychotherapists. As such, the lack of statistical significance for Hypotheses 1 through 6 can be attributed to perceived differences based on the known criteria. Responding psychotherapists appear to have responded to the different levels of the dependent variables but did not respond differently to the two ethnic groups. In addition to the conclusions based on testing of the specific hypotheses, some other significant findings resulted from this study. As outlined in Chapter IV, a significant number of psychotherapists (about 41%) indicated that more than 10% of their clients were Mexican American. Additionally, approximately 35% of 99 the responding clinicians indicated they had more than 11 Mexican American clients in the past year. Further, more than half of the respondents had six or more years’ experience working with the Mexican American population. What these figures indicate is that a large number of Mexican American clients are seen by Texan psychotherapists. This wealth of experience is in contrast to the outdated ”information" we have about the population and the dearth of research in the subject area. Also in contrast to the lack of information, the majority of respondents (60%) considered themselves more than familiar with Mexican American culture. Other findings of interest include the large number of variables found to be correlated with one another. As outlined in Chapter IV, it was found that all the dependent and independent variables had at least one significant correlation result with another variable with the exception of the independent variable child/adult. Most important, significant relationships were found between therapists who were currently conducting therapy and the level of assertiveness, level of self-disclosure, locus of control orientation, and percentage of Mexican American clients. Apparently, clinicians who were currently engaged in therapy responded differently to three of the dependent variables and differed in the total percentage of Mexican American clients on their caseload. Other important findings also involved the independent variable percent of MA clients. Significant correlations were found between the overall percentage of Mexican American clients the responding 100 psychotherapists reported having and locus of control, theoretical orientation, Mexican American culture familiarity, Spanish fluency, and the race or ethnic background of the responding psychotherapists. Apparently, relationships between these variables existed, but the nature of the relationships is unknown at this time. Also unknown is the nature of the significant correlations involving the level of experience one had working with Mexican Americans. As noted before, experience with MA c1ients was significantly correlated with Mexican American culture familiarity and Spanish fluency. A significant relationship also was found between level of experience and one’s gender and race or ethnic background. Last, the number of Mexican American clients a psychotherapist saw appeared to have some significant relationship with three of the dependent variables. Apparently, psychotherapists differed in their perceptions of locus of control, assertiveness, and self-disclosure, depending on the number of Mexican American clients they had seen within the last year. There was also a relationship between this independent variable and whether the psychotherapist was male or female. Again, the nature of these relationships is unclear, but they certainly raise a number of questions. Limitations of the Study A number of considerations must be taken into account when assessing the conclusions of this research. First, the findings are 101 limited to the confines of the geography of where the research was done; the psychotherapists in the sample were licensed in Texas. A serious consideration is that it may be that Texan psychotherapists are different from psychotherapists in other parts of the country. One possibility is that, by virtue of their proximity to the Mexican border, Texas psychotherapists may be more sensitive to stereotypic notions than others and therefore less likely to perceive the ethnic differences described in this research. Dr Texas psychotherapists may be more sensitive to Mexican American stereotyping, whether they themselves hold particular stereotypes or not“ Another considera- tion is that the responding clinicians had had contact with Texan Mexican Americans, who may also be different from Mexican Americans from other areas of the country. Of course, it may also be true that the Mexican American clients who had been seen by the Texan psychotherapists were different from Mexican Americans in Texas in general. It is quite possible that the experience of the Texan respondents was based on more assimilated Mexican Americans and/or those of a higher socioeconomic status than their ethnic peers. Of course, the time frame in which this research was conducted needs to be considered. Another important limitation for considera- tion is the fact that the conclusions are based on and therefore limited by the responses of the psychotherapists who responded. Although an attempt was made to determine potential differences, it may well be that the responding psychotherapists were different in some way from the nonrespondents. 102 An important factor, too, is the questionnaire used in the studyu Although the questionnaire underwent pilot testing and appeared to have good reliability and validity, it was an unproven instrument. Further, although the vignettes were also pilot tested, they were unproven, as well. Implications for Research Given the numerous implications of this study, this section is divided into three main parts. The first part involves some general implications of all the findings. This is followed by implications of the conclusions based on the testing of specific hypotheses. The last part outlines potential research ramifications of other pertinent findings. Nhen analyzing the conclusions of this study and their general implications for further research, an important consideration must be made. It is important to consider that the ethnicity in the Anglo American and Mexican American vignettes was based on the name of the client and an ethnic identification label. Communication of ethnicity was accomplished by the printed word and not by other variables that may be associated with ethnic cultural identity (e.g., body language, physical characteristics, and/or language f1uency). The changing of a name and the identification of ethnicity with a label may not have been enough to have had the desired effect on psychotherapists. Even though the statistical data supported the conclusion that there were no ethnic differences, this does not offer conclusive proof that there were no differences. 103 Because the perceptions of the therapists were based on the vignettes, the psychotherapists responded to a specific client described in a specific vignette. Even though the vignettes were found to be adequate before they were sent out, the ethnic stimuli offered in the vignettes might not have been sufficient to "pull" the perception of ethnic differences that psychotherapists might indeed have had. It is also noted that the conclusions from Hypotheses 1 through 6 were different from those in the previously mentioned studies as the focus was on psychotherapists’ perceptions and not members of the two ethnic groups themselves. As such, this research has other important implications for the therapeutic/treatment domain. However, as is the case with the other findings of this study, it is noted that the conclusions supported by this study may be valid only at the point of initial impression. The lack of significant findings in terms of ethnicity can be seen in terms of the level of referral and before seeing a client. Apparently, most psychothera- pists in this study did not discriminate at the point of a written description of a client. Hhether psychotherapists are influenced at the point of face—to-face encounter remains to be investigated. A significantly viable extension of this study would be to investigate the perceptual differences of clinicians beyond the scope of the printed word or referral process. Future research could emphasize face-to-face encounters between clinicians and Mexican American clients to assess possible ethnic differences. This could be done with the use of videotape, "trained clients," and/or "live" 104 interviews. The advantage of such approaches is that the percep- tions of psychotherapists would be based on numerous other variables often associated with ethnicity. Psychotherapists would have a chance to respond to color of skin, hair color, language, and many other potential ethnic/stereotypic indicators. Another advantage concerns the limitation of an analogue study. Information to the therapists could be standardized, and the clinician could experience the different levels of the dependent variables. A number of research implications were drawn from the conclusions of hypothesis testing. One such implication is that, upon reading a problem of a Mexican American client, psychothera- pists might not generate age-old perceptions that Mexican Americans are less assertive, less self-disclosing, and/or more external in locus of control orientation than Anglo Americans. These perceptions may be present even with the absence of socioeconomic information. This would suggest that socioeconomic factors are noncontributory regarding perceptions of ethnic variab1es along the dimensions of the dependent variables. This conclusion argues for a position that has been the center of much controversy. Some researchers have argued that, when taking socioeconomic background into account, any' ethnic. differences along these dimensions are nonsignificant. These researchers are represented by individuals such as Casavantes (1970), who argued that any ethnic differences attributed to Mexican Americans are due to membership in the "culture of poverty" rather than any ethnic membership. It appears 105 that another question for future research is whether this is true. One possible approach is to examine members from a number of "poverty cultures" and to contrast them with those from different socioeconomic levels. In the past, there have also been suggestions that ethnic differences' are due to an urban versus rural background. The implications of this research suggest that this may not be a confounding variable as there were no indications of the background of the clients in the vignettes. Again, a novel approach may be called for in order to answer this question. One possibility is to study various members of both backgrounds without taking ethnicity into account at all. Another implication of this study is based on the finding that there were no ethnic differences of self—disclosure. This finding does not lend credence to the belief that a low level of self- disclosure for Mexican Americans is a variable in participation in therapy and underutilization of mental health services. If psychotherapists do not perceive a difference in level of self- disclosure for Mexican Americans, it could be argued that they do not think self-disclosure is a variable of the lower rates of mental health utilization. Another important implication of the finding of Hypothesis 2 is the therapeutic technique used in therapy with Mexican American clients. Although technique has been found to affect the self- disclosure of Anglo American c1ients (Highlen & Baccus, 1977; Powell, 1968), it may be that clinicians do not expect to use 106 different techniques with Mexican Americans than with Anglo Americans, at least at the point of referral. The implications of no ethnic differences regarding locus of control are many. One such implication is that psychotherapists may assume Mexican American clients are just as willing or unwilling to seek help as Anglo American clients. If Mexican Americans were more external in their LOC orientation, they would be more likely not to take responsibility to change their circumstances and thus much more unlikely to seek assistance. Another implication is that Mexican Americans might not be perceived as being in any more distress than Anglo Americans. It has been noted that an external LOC orientation contributes to psychological distress as people with such an orientation tend to give up and may have impaired coping skills. Because psychotherapists may not perceive an ethnic difference, they may not perceive a difference in the level of distress and/or coping skills for Mexican Americans. These last few implications would argue against those who suggest different treatment strategies and modalities when working with Mexican American clients. It appears this is another important area that needs further study. If the results of this study are replicated and found to be true, there are a number of treatment assumptions and strategies that will require reexamination. Of course, ethnicity' and ethnic stereotyping are multidimensional constructs. As such, the research implications are 107 rather staggering. Possible confounding variables include interac- tions of socioeconomic factors, acculturation, language usage, influence of ethnic peers, religion, family educational level, generational standing, and family cohesion. Other confounding factors may be geography, influence of teachers and parents, and cultural identification. In this study alone, a number of signifi- cant relationships were found but have not yet been explained. Why do clinicians who currently conduct therapy answer differently from those who do not? What is the nature of the relationship between percentage of Mexican American clients and theoretical orientation? Are psychotherapists finding a particular treatment modality more effective than others? The relationship between percentage of Mexican American clients and Spanish fluency and culture fami1iarity is also puzzling. Does it mean Mexican American clients are more likely to seek services from those who are more culturally aware and able to converse in Spanish? Other unanswered questions generated from this study include the relationships found between experience with Mexican American clients and race/ethnic background of psychotherapists. Is a disproportionate number of Mexican American clients seen by minority providers? If this is true, is it due to the choice of the client, or is it a reflection of the psychotherapists’ professional interests? And in what way is the gender of the clinician related to his/her experience with Mexican American clients? Are Mexican American clients drawn to one gender or the other, or is this again a reflection of professional interests? 108 Many' other implications were generated from this research. Important questions remain concerning why the number of Mexican American c1ients seen within the last year was related to perceptions of locus of control, assertiveness, and self-disclosure. Did clinicians who had seen more Mexican American clients have more stereotypic perceptions? Or did the same group of clinicians perceive no ethnic differences and answer differently from those with less experience? The research possibilities are indeed staggering. To assess such multiple variables, a number of research strategies might be necessary. One such strategy is to take advantage of the knowledge base of clinicians who work with Mexican Americans. The results of this study indicated that psychotherapists were seeing a substantial number' of' Mexican American c1ients. Further, these clinicians indicated a knowledge of Mexican American culture. It would appear prudent to tap into this knowledge base in an effort to enlarge the data base regarding ethnic differences. Psychotherapists are in a position where they work, usually intimately, with others of various nationalities and cultures. Future researchers could examine possible ethnic differences by studying the perceptions of clinicians who work with such groups. In addition to the utility of examining a number of possible interactions, researchers could focus on any one variable mentioned here or in other studies. By concentrating on one variable, a more in-depth analysis might be accomplished. Also, nontraditional 109 research efforts might be required. It might be advantageous to have a small number of subjects or even to learn from a single case study. Although generalizability would be limited, any research data are needed. Although there is little research information on Mexican Americans, there has been much debate concerning stereotypic behaviors. In general, it cannot be said one way or the other whether the ethnic stereotypes presented here do, indeed, exist. As this study and the results of previous efforts have shown, there are many unanswered questions. It is only by answering such questions that we can distinguish fact from fiction. Through future research efforts, including continued assessment (H’ psychotherapists’ perceptions, we will be in a better position to dispel unwarranted stereotypic myths. Knowledge of Mexican Americans’ similarities and differences will also enable us to provide quality psychotherapeutic intervention without the presence of inaccurate treatment assumptions that might be based on cultural stereotypes. APPENDICES APPENDIX A VIGNETTES 110 MA HIGH VIGNETTE Jose Garcia (not his real name) is a 26-year-old male Mexican American client who has come to your office because of problems he has with his job and people at work. During the first session, he appears somewhat anxious but maintains direct eye contact with you. There are no signs of depression. Throughout the interview, he is cooperative and speaks politely. He states that he has never been to therapy before because he has been able to deal with past problems on his own. Near the beginning of the interview, Jose indicates that part of the problem is his relationship with his boss. Jose feels competent as a worker but believes his boss unfairly gives him harder assignments than those assigned to his co-workers. Jose relates of feeling angry at his boss. He is particularly angry about a recent evaluation he received. Jose states that, in some ways, he can understand how his boss arrived at the conclusions of the evaluation but that overall his boss is underestimating his worth as a worker. Sometime during the interview, Jose is questioned as to what he has done so far about his situation. Jose indicates that he has tried on numerous occasions to confront his boss. Jose states that he has directly expressed his anger to his boss but feels he is met with contempt and disdain. Jose freely volunteers that he is proud to have been able to work out most interpersonal difficulties, including those with his wife, Consuela, but the situation with his boss has been problematic. Nithout any prompting, Jose continues to talk about his frustration with work and begins to discuss the option of changing jobs. Generally, he considers himself to be a hard and dedicated worker and feels he could find another job with little difficulty, despite the current job market. After talking about it, he speaks less about changing jobs and more about staying in his current position. By the end of the session, he is determined to go immediately back to work and talk to his boss about his anger and frustration. Jose feels confident that if he can do things differently, some kind of change will occur. Jose asks to see you for another appointment. 111 AA HIGH VIGNETTE John Smith (not his real name) is a 26-year-old male Caucasian client who has come to your office because of problems he has with his job and people at work. During the first session, he appears somewhat anxious but maintains direct eye contact with you. There are no signs of depression. Throughout the interview, he is cooperative and speaks politely. He states that he has never been to therapy before because he has been able to deal with past problems on his own. Near the beginning of the interview, John indicates that part of the problem is his relationship with his boss. John feels competent as a worker but believes his boss unfairly gives him harder assignments than those assigned to his co-workers. John relates of feeling angry at his boss. He is particularly angry about a recent evaluation he received. John states that, in some ways, he can understand how his boss arrived at the conclusions of the evaluation but that overall his boss is underestimating his worth as a worker. Sometime during the interview, John is questioned as to what he has done so far about his situation. John indicates that he has tried on numerous occasions to confront his boss. John states that he has directly expressed his anger to his boss but feels he is met with contempt and disdain. John freely volunteers that he is proud to have been able to work out most interpersonal difficulties, including those with his wife, Susan, but the situation with his boss has been problematic. Hithout any prompting, John continues to talk about his frustration with work and begins to discuss the option of changing jobs. Generally, he considers himself to be a hard and dedicated worker and feels he could find another job with little difficulty, despite the current job market. After talking about it, he speaks less about changing jobs and more about staying in his current position. By the end of the session, he is determined to go immediately back to work and talk to his boss about his anger and frustration. John feels confident that if he can do things differently, some kind of change will occur. John asks to see you for another appointment. 112 MA LOH VIGNETTE Jose Garcia (not his real name) is a 26-year-old male Mexican American client who has come to your office because of problems he has with his job and with people at work. During the first session, he appears somewhat anxious and does not maintain direct eye contact with you. There are no signs of depression. Throughout the interview, he is cooperative and speaks politely. He states that he has never been to therapy before because he feels that, in the past, most things have just worked out by themselves. Throughout the interview, Jose answers questions with brief responses. After much prompting, you have come to find out that part of the problem is Jose’s relationship with his boss. Jose feels that his boss does not like him and unfairly gives him harder assignments than those of his co-workers. Jose indicates that there is little he can do to make his boss like or respect him. Though he does not express it directly, you come to realize that Jose is very angry at his boss. Jose is particularly angry about a recent evaluation he received. He states that he does not understand how his boss arrived at the conclusions of the evaluation. Jose feels that his worth as a worker is being underestimated but tempers this by saying that most people’s worth is underestimated no matter what they do or how hard they try. After being silent much of the time, Jose is questioned as to what he has done so far about his situation. Jose indicates that he has done very little and alludes to being afraid of confronting his boss. Prompted further, Jose admits he cannot tell the boss much of anything because he is met with contempt and disdain. Eventually, you come to realize that Jose avoids confrontation with others, including Consuela, his wife, and other members of his family. Hhen prodded further, Jose talks about his general frustration with work and discusses briefly the option of changing jobs. Generally, he considers himself to be a good worker but feels getting another job would be a matter of luck. By the end of the session, he states that he is tempted to "let nature take its course" and not do anything different. He feels the situation will probably get better and comments that it is helpful to just talk about his problems. He does not ask for another appointment, but when offered, he agrees to one. 113 AA LOH VIGNETTE John Smith (not his real name) is a 26-year-old male Caucasian client who has come to your office because of problems he has with his job and with people at work. During the first session, he appears somewhat anxious and does not maintain direct eye contact with you. There are no signs of depression. Throughout the interview, he is cooperative and speaks politely. He states that he has never been to therapy before because he feels that, in the past, most things have just worked out by themselves. Throughout the interview, John answers questions with brief responses. After much prompting, you have come to find out that part of the problem is John’s relationship with his boss. John feels that his boss does not like him and unfairly gives him harder assignments than those of his co-workers. John indicates that there is little he can do to make his boss like or respect him. Though he does not express it directly, you come to realize that John is very angry at his boss. John is particularly angry about a recent evaluation he received. He states that he does not understand how his boss arrived at the conclusions of the evaluation. John feels that his worth as a worker is being underestimated but tempers this by saying that most people’s worth is underestimated no matter what they do or how hard they try. After being silent much of the time, John is questioned as to what he has done so far about his situation. John indicates that he has done very little and alludes to being afraid of confronting his boss. Prompted further, John admits he cannot tell the boss much of anything because he is met with contempt and disdain. Eventually, you come to realize that John avoids confrontation with others, including Susan, his wife, and other members of his family. Hhen prodded further, John talks about his general frustration with work and discusses briefly the option of changing jobs. Generally, he considers himself to be a good worker but feels getting another job would be a matter of luck. By the end of the session, he states that he is tempted to "let nature take its course" and not do anything different. He feels the situation will probably get better and comments that it is helpful to just talk about his problems. He does not ask for another appointment, but when offered, he agrees to one. 114 NA HIGH VIGNETTE A 26-year-old male client has come to your office because of problems he has with his job and people at work. During the first session, he appears somewhat anxious but maintains direct eye contact with you. There are no signs of depression. Throughout the interview, he is cooperative and speaks politely. He states that he has never been to therapy before because he has been able to deal with past problems on his own. Near the beginning of the interview, your client indicates that part of the problem is his relationship with his boss. He feels competent as a worker but believes his boss unfairly gives him harder assignments than those assigned to his co-workers. Your client relates of feeling angry at his boss. He is particularly angry about a recent evaluation he received. He states that, in some ways, he can understand how his boss arrived at the conclusions of the evaluation but that overall his boss is underestimating his worth as a worker. Sometime during the interview, you question him as to what he has done so far about his situation. He indicates that he has tried on numerous occasions to confront his boss. He states that he has directly expressed his anger to his boss but feels he is met with contempt and disdain. He freely volunteers that he is proud to have been able to work out most interpersonal difficulties, including those with his wife, but the situation with his boss has been problematic. Nithout any prompting, your client continues to talk about his frustration with work and begins to discuss the option of changing jobs. Generally, he considers himself to be a hard and dedicated worker and feels he could find another job with little difficulty, despite the current job market. After talking about it, he speaks less about changing jobs and more about staying in his current posi- tion. By the end of the session, he is determined to go immediately back to work and talk to his boss about his anger and frustration. He feels confident that if he can do things differently, some kind of change will occur. He asks to see you for another appointment. 115 NA LOH VIGNETTE A 26-year-old male client has come to your office because of problems he has with his job and with people at work. During the first session he appears somewhat anxious and does not maintain direct eye contact with you. There are no signs of depression. Throughout the interview, he is cooperative and speaks politely. He states that he has never been to therapy before because he feels that, in the past, most things have just worked out by themselves. Throughout the interview, your client answers questions with brief responses. After much prompting, you come to find out that part of the problem is his relationship with his boss. He feels that his boss does not like him and unfairly gives him harder assignments than those of his co-workers. He indicates that there is little he can do to make his boss like or respect him. Though he does not express it directly, you come to realize that your client is very angry at his boss. He is particularly angry about a recent evaluation he received. He states that he does not understand how his boss arrived at the conclusions of the evaluation. He feels that his worth as a worker is being underestimated but tempers this by saying that most people’s worth is underestimated no matter what they do or how hard they try. After being silent much of the time, you question him as to what he has done so far about his situation. He indicates that he has done very little and alludes to being afraid of confronting his boss. Prompted further, he admits he cannot tell the boss much of anything because he is met with contempt and disdain. Eventually, you come to realize that he avoids confrontation with others, including his wife and other members of his family. When prodded further, your client talks about his general frustration with work and discusses briefly the option of changing jobs. Generally, he considers himself to be a good worker but feels getting another job would be a matter of luck. By the end of the session, he states that he is tempted to "let nature take its course" and not do anything different. He feels the situation will probably get better and comments that it is helpful to just talk about his problems. He does not ask for another appointment, but when offered, he agrees to one. APPENDIX B COVER LETTER 116 MICHIGAN STATE UNIVERSITY COUNSELING CENTER EAST LANSING 0 HICHIGAN 0 “Ill-Ill} DIVISION OI" STUDENT AFFAIRS AND SERVICES April 1989 9 Dear Collegue: We would appreciate your help. We are conducting a study involving the assessment of clients. The full nature of the study cannot be revealed at this time but generally involves variables affecting client vulnerability to maladjustment. To do this research, we are dependent on psychotherapists like yourself to answer the brief questionaire enclosed in this packet. Also enclosed is a brief description of a client. Please read the description and the instructions to the questionaire. Then, using your clinical expertise and intuition, please answer all questions and return the questionaire in the stamped envelope provided. Please be assured that your responses will remain anonymous. A full explanation of the nature and results of the study will be available for those who indicate a desire for such. At the end of the questionaire you will be asked if you would like this information sent to you. Ybu will also find an assigned code number at the bottom of the last page of the question- aire. This number will be utilized by a clerk, someone other than the re- searchers, to ensure that study results and explanations are sent to those who request it. we believe this to be an important research effort and expect that what we learn can be beneficial to both therapists and clients. Of course, your participation is voluntary and you may withdraw your participation at any point. We hope you can help us by giving your most serious attention to this brief exercise which should take some 10 to 15 minutes to complete. Thank you in advance for your cooperation. \Sincerely, / 3 Powell, Ph.D. ' fessor, Counseling Center and Co 'ng Psychology David . Arreo , Doctoral Candidate Counseling Psychology Captain, USAF Enclosures MSU is an Affirmative Action/Equal Opportunity Institutio- APPENDIX C FIRST FOLLOW-UP LETTER 117 May 1989 Dear Colleague, Just a reminder! If you have already mailed your completed questionnaire which was sent to you recently, thank you very much. You can be sure your responses will be very useful. If you have not found time to do so, at your earliest conveni- ence, please fill out the questionnaire and mail it. Enclosed is another SASE for your convenience. This research is important, and your response will be very helpful. Thank you in advance! Sincerely, David R. Arreola, Doctoral Candidate Michigan State University, Counseling Psychology Captain, USAF APPENDIX D SECOND FOLLOW-UP LETTER 118 May 1989 Dear Colleague, Just another friendly reminder. There is no way to determine if you have filled out the questionnaire sent to you recently. If you have, great! Thank you very much. Your time and effort are appreciated, and your responses will be useful. If you have not returned the questionnaire, we urge you to do so. We are quite dependent on your responses, which will be of use to both clients and therapists. Please respond as soon as possible. An additional SASE has been included for your convenience. Thank you again in advance. Sincerely, David R. Arreola, Doctoral Candidate Michigan State University, Counseling Psychology Captain, USAF APPENDIX E QUESTIONNAIRE 119 Having just read the enclosed vignette, you have begun to form impressions of the client. Therapists are generally able to make reliable and consistent inferences concerning a client's general tendency toward adjustment/maladjustment. By design, you were given a limited amount of information about the client. However, we are asking you to form a clinical opinion as to whether this client is more inclined toward adjustment or maladjustment in general. Please note that the questions are not asking for your perceptions of the client's current clinical condition. For example, question #3 is asking if you think the client is vulnerable to feelings of depression and is not asking if you think the client is depressed now. Please answer all questions by circling one of the seven numbers along the following scale: 1 2 3 4 5 6 7 Strongly Slightly Slightly Strongly Disagree Disagree Agree Agree Moderately Neutral Moderately Disagree Agree 1. This client probably will not show for the next appointment. 1 2 3 4 5 . 6 7 2. This client has difficulties with people other than those at work and at home. 1 2 3 4 5 6 7 3. This client is vulnerable to feelings of depression. ‘1 1 2 3 4 . 5 6 7 4. This client is vulnerable to feelings of despair and hopelessness. 1 2 3 4 5 6 7 5. This client is vulnerable to having suicidal thoughts. 1 2 3 4 5 6 7 6. This client is vulnerable to feelings of worthlessness. 1 2 3 4 5 6 7 120 1 2 3 4 5 6 7 Strongly Slightly Slightly Strongly Disagree Disagree Agree Agree Moderately Neutral Moderately Disagree Agree 7. This client is vulnerable to experiencing psychological problems later in life. 1 2 3 4 5 6 7 8. This client is vulnerable to periods of low energy. 1 2 3 4 5 6 7 9. This client is vulnerable to feelings of lonliness and alienation. 1 2 3 4 5 6 7 10. This client is vulnerable to feelings of anxiety. 1 2 3 4 5 6 7 11. This client is likely to cope with stress unsatisfactorily. 1 2 3 4 5 6 7 12. This client is vunerable to worrying excessively. 1 2 3 4 5 6 7 13. This client has been vulnerable to psychological problems and maladjustment in the past. 1 2 3 4 . 5 6 7 \ 14. Using the continuum below, how would you rate the overall psychological adjustment of this client? 1 2 3 4 5 6 7 Very poorly Very well- adjusted adjusted Research has shown that there are many variables therapists consider when judging psychological adjustment. Among these is the ability to assert oneself when dealing with others. With this in consideration, please answer the following: 15. This client has difficulty with conflict. 1 2 3 4 5 6 7 121 1 2 3 4 5 6 7 Strongly Slightly Slightly Strongly Disagree Disagree Agree Agree Moderately Neutral Moderately Disagree Agree 16. This client has difficulty letting others know about his needs and desires. 1 2 3 4 5 6 7 17. This client avoids interpersonal tension. 1 2 3 4 5 6 7 18. This client has difficulty expressing opinions that differ with others. 1 2 3 4 5 6 7 19. Looking only at the ability to be assertive and using the continuum below, how would you rate this client? 1 2 3 4 5 6 7 Very low Very high assertiveness assertiveness 20. To what extent is this client's level of assertiveness contributing to your judgement of overall psychological adjustment? 1 2 3 4 5 6 7 No Contributing contribution very much Another variable that affects our‘perceptions of psychological adjustment is the ability to disclose personal information. With this in consideration, please answer the following: 21. If he returns, this client will probably take a long time to divulge personal/emotional information. 1 2 3 4 5 6 7 22. This client is reluctant to talk about his problems with a therapist. 1 2 3 4 5 6 7 122 1 2 3 4 5 6 7 Strongly Slightly Slightly Strongly Disagree Disagree Agree Agree Moderately Neutral Moderately Disagree Agree 23. This client probably will spend much of the time during sessions in silence. 1 2 3 4 5 6 7 24. Looking only at the ability to self—disclose and using the continuum below, how'would you rate this client? 1 2 3 4 5 6 7 Very low Very high self-disclosure self-disclosure 25. To what extent is this client's level of self-disclosure contributing to your judgement of overall psychological adjustment? 1 2 3 4 5 6 7 No Contributing contribution very much The last variable to consider is the belief that one has an impact on one's environment (an internal Locus of Control orientation). With this in consideration, please answer the following: 26. This client does not feel he has much impact on his situation. 1 2 3 4 5 6 7 27. This client attributes his situation more to luck than anything he has or has not done. 1 2 3 4 5 6 7 28. This client is vulnerable to giving up rather than persevering. 1 2 3 4 5 6 7 29. This client is not likely to plan ahead and would rather let events occur spontaneously. 1 2 3 4 5 6 7 30. 31. 123 Looking only at the concept of Locus of Control (LOC) and using the continuum below; how would you rate this client? 1 2 3 4 5 6 7 External LOC Internal LOC (no impact on environment) (great impact on environment) To what extent is this client's Locus of Control contributing to your judgement of overall psychological adjustment? 1 2 3 4 5 6 7 No Contributing contribution very much Thank you for filling out the previous section! So that we may compare your answers with those of other psychotherapists, please fill out the following information. 32. 33. 34. 35. 36. 37. Your profession (check one): ___Psychology ___Social Work ___Other, please specify: Your gender: Female Male Your main theoretical orientation (please check one only): ___Analytic ___Neoanalytic ___Behavioral ___Cognitive Behavioral ___Cestalt Rogerian Eclectic Other, please specify: Your age: Your ethnic/racial background (please check your main identification). ___Black (Afro—American) ___Mexican-American Caucasian Asian or Pacific Islander Other, please specify: Are you currently conducting therapy? Yes No 124 38 Please indicate the average number of hours per week you spend providing therapy: 0-5 hrs 11-20 hrs 6—10 hrs 31—40 hrs 11-20 hrs Over 40 hrs 39 Do you work primarily with adults or children? Children Adults 40 On average, what percentage of your clients have been Mexican American? - 0-1og 51—708 11-30% 71—90% 31-508 Over 90% 41 Within the last year, approximately how many Mexican American clients have you provided therapy for? 0-10 clients 61-80 clients 11-20 clients 81—100 clients 21-40 clients Over 100 41-60 clients 42 Using the continuum below, how'would you rate yourself on familiarity with Mexican American culture? 1 2 3 4 5 6 7 Very Very unfamiliar familiar 43. Using the continuum below, how would you rate your fluency of the Spanish language? 1 2 3 4 5 6 7 No fluency Very fluent 44. Overall, how'much experience do you have working with Mexican American clients? . None Less than 8 yrs Less than 1 yr Less than 15 yrs Less than 3 years More than 15 yrs Less than 5 yrs Please place this questionnaire in the envelope provided and mail it at your earliest convenience. Thank you for taking the time to complete and return it! If you would like a full explanation of this research and/or the results, please fill out the bottom of this page. Otherwise, the coded number at the bottom is used for clerical purposes only and in no way violates respondent anonymity. I would like an explanation of this study Yes No I would like a summary of results Yes No APPENDIX F PERMISSION LETTER FROM UNIVERSITY COMMITTEE ON RESEARCH INVOLVING HUMAN SUBJECTS 125 MICHIGAN STATE UNIVERSITY UNIVEISITY COW ON RESEARCH INVOLVING EAST IANSING 0 MICHIGAN 0 0324-111! HUMAN SUBJECTS (UQIHS) 206 m HALL (317) 353-973. July 27, 1988 mm Ca tain David Arreola 50 Karen Devine, TX 78016 Dear Captain Arreola: Subject: "PSYCHOTHERAPISTS' PERCEPTIONS OF LOCU S OF CONTROL ASSERTIVENESS, AND SELF-DISCLOSURE DIFFERENCES OF MEXICAN- AND ANGLO-AMERICAN S: RELATIONSHIP TO JUDGMENTS OF CLIENT ADJUSTMENT W" The above project is exempt from full UCRIHS review. I have reviewed the proposed research protocol and find that the rights and welfare of human 1subjects appear to be protected. You have approval to conduct the researc . You are reminded that UCRIHS approval is valid for one calendar year. If you plan to continue this pr%ect beyond one year, please make provisions for obtaining appropriate U RIHS approval ' Any chafiifes in procedures involving human sub'ects must be reviewed by the UC HS prior to initiation of the change. CRIHS must also be notified promptly of any problems (unexpected side effects, complaints, etc.) involving human subjects during the course of the work. Thank on for bringing this project to our attention. If we can be of any future elp, please do not hesitate to let us know. Sincerely, Ema/AA Daniel A. Bronstein, SJ.D. Vice Chair, UCRIHS DAB/sat cc: J. Powell MSU all ma Affirmative Action/Equal Opportunity Institution APPENDIX G LETTER CONVEYING RESULTS OF THE STUDY TO RESPONDENTS 126 MICHIGAN STATE UNIVERSITY COUNSHJNG enrm EAST LANSING 0 MICHIGAN 0 43324-1113 DIVISION 0! STUDENT AflAIIS AND SERVICES Dear Colleague: During the summer of last year, you were considerate enough to return a brief questionnaire after reading a description of a client. You had indicated on the questionnaire that you desired an explanation of the nature and results of the study. After much delay, the results of the study have been compiled and analyzed. It is the intent of this letter then to provide you with the requested information. The study was conducted to assess psychotherapists’ perceptions of locus of Control (LOC), self-disclosure, and assertiveness differences between Mexican American and Anglo American clients. Six hundred clinicians such as yourself received the same questionnaire. However, client descriptors were varied and included Anglo American and Mexican American clients with different levels of LOC, self-disclosure, and assertiveness. Two hundred and seven questionnaires were returned. Statistical analysis revealed no significant differences between the two ethnic groups. In other words, the results of the study showed that of the clinicians who responded, there were no perceived differences of LOC, self-disclosure, or assertiveness between Mexican American and Anglo American clients. We apologize for the delay in complying with your request. We hope this information answers some questions you may have had regarding the material you received and filled out. Thank you for your assistance, patience, and cooperation. Sincerely, Counseling Psy ology Captain, USAF a”? 0 CW Fatwa/CC— hn Powell, Ph.D Professor, Counseling Center and MS U is an Allin-aim Action/Equal Opportunity Institution REFERENCES REFERENCES Acosta, F. X., & Sheehan, J. G. (1978). Self-disclosure in relation to psychotherapist expertise and ethnicity. American Journal of Community Psychology, E, 545-553. Arrin, H., & Colton, R. R. (1970). Statistical mgghpds. New York: Barnes & Noble. Borg, N. R., & Gall, M. D. (1983). Educational research. New York: Longman. Borrego, R. L., Chavez, E. L., & Titley, R. H. (1982). Effect of counselor technique on Mexican-American and Anglo-American self-disclosure and counselor perception. Journal of Counsel- ing Psychology, gs, 538-541. Buriel, R. (1982). Mexican- and Anglo-American children’s locus of control and achievement in relation to teachers’ attitudes. Journal of Genetic Psychology, 159, 131-143. Buriel, R., & Rivera, L. (1980). The relationship of locus of control to family income and familism among Anglo- and Mexican- American high school students. Journal of Social Psychology, 111, 27-34. Cabrera, Y. A. (1964). A study of American and Mexican-American culture values and their significance in education. Dissergs- tion Abstracts, gs, 309. Calvillo-Simpson, S. (1983). Cross-cultural study of assertiveness and self-esteem among high school students. Dissertation Abstracts, 1E, 344-A. Carballo-Dieguez, A. (1986). Effects of Hispanic background on locus of control in Mexican-American sophomores. Dissertation Abstracts, 11, 3511-8. Casavantes, E. (1970, Winter). Pride and prejudice: A Mexican American dilemma. Civil Rights Diggst, 1, 22-27. Chandler, C. R. (1979). Traditionalism in a modern setting: A comparison of Anglo- and Mexican—American value orientations. Human Organization, ss, 153-159. 127 128 Cole, 0., Rodriguez, J., & Cole, S. (1978). Locus of control in Mexicans and Chicanos: The case of the missing fatalist. qurua! pf gpusulting and Clinipal Esycholpgy, as, 1323-1329. Cozby, P. C. (1973). Self-disclosure: A literature review. E§19h919919i1_flulleiin. 12. 73-91. Erliich, H. J., & Graeven, D. B. (1971). Reciprocal self- disclosure in a dyad. qurpal of Esperimenia! Susial Esychology, 1, 389-400. Evans, F. B., & Anderson, J. G. (1973, Fall). The psychocultural origins of achievement motivation: The Mexican American family. Sociology of Edupation, 1E, 396-416. Farris, B. E., & Glenn, N. D. (1976). Fatalism among Anglos and Mexican Americans in San Antonio. Spgiplpgy and Spcia! Research, 99, 393-402. Franco, J. N., & LeVine, E. (1985). Effects of examiner variables on reported self-disclosure: Implications for group personality testing. Journal of Bahavioral Spiencas, 1, 199- 210. Franco, J. N., Malloy, T., & Gonzalez, R. (1984). Ethnic and acculturation differences in self-disclosure. Jo n of Social Esychology, 122, 21-32. Gaa, J. P., & Shores, J. H. (1979). Domain specific locus of control among Black, Anglo, and Chicano undergraduates. Journal of Social Psychplpgy, 191, 3-8. Gaa, J. P., Hilliams, R. E., & Johnson, S. H. (1981). Domain- specific locus of control orientations of Anglo, Black, and Chicano adolescents. Journal of Psychology, 191, 185-190. Galfo, A. J. (1975). Interprgting edupational rgseapch. Dubuque, IA: William C. Brown. Garza, E. D. (1977). Adolescent Mexican-American student attitudes of self-concept, locus of control and family ideology in the lower Rio Grande Valley of Texas. Dissertation Abstrasts, 3835-A. Garza, R. T. (1977). Personal control and fatalism in Chicanos and Anglos: Conceptual and methodological issues. In J. L. Martinez, Jr. (Ed.), thcano psychology (pp. 97-108). San Francisco: Academic Press. 129 Garza, R. T. & Ames, R. E. (1974). A comparison of Anglo- and Mexican- American college students on locus of control. Journal pf Dpnsul ti ng and Dlinipal Psypnplpgy, _2, 919. Glass, G. V., & Hopkins, K. D. (1984). Statistisal naihpds in fidupation and psysnplogy. Englewood Cliffs, NJ: Prentice- a . Glass, G. V., & Stanley, J. C. (1970). Stat'stic l metho in aducation and psychology. Englewood Cliffs, NJ: Prentice- Hall. Gorsuch, R. L. (1974). factor analysis. Philadelphia: H. B. Saunders. Grebler, L., Moore, J. H., & Guzman, R. C. (1970). The Magisan Amerisan people. New York: Free Press. Grodner, B. S. (1975). Assertiveness and anxiety among Anglo and Chicano psychiatric patients. Dissertation Abstrasts, 3042-8. Hall, J. R., & Beil-Harner, D. (1978). Assertiveness of male Anglo and Mexican-American college students. Journal pf Sosial Esychology, 1D§, 175-178. Harris, R. J. (1975). A primar of multivariani statisiiss. New York: Academic Press. Hays, N. L., & Hinnlow, R. L. (1971). Siatistics: Prppapiliiy, inference, and dapision. New York: Holt, Rinehart & Ninston. Heiman, E. M., Burruel, G., & Chavez, N. (1975). Factors determin- ing effective psychiatric outpatient treatment for Mexican- Americans. Hospiial and Community Psychiatry, as, 515-517. Hoppe, C. M., Kagan, S. M., & Zahn, G. L. (1977). Conflict resolution among field-independent and field-dependent Anglo- American and Mexican-American children and their mothers. Developmental Psychology, 11, 591-598. Jessor, R., Graves, T. D., Hanson, R. C., & Jessor, S. L. (1968). o i t ersonalit and evi t h vior. New York: Holt, Rinehart & Winston. Justin, N. (1970). Culture conflict and Mexican American achieve- ment. Sphopl and Sosiety, DD, 27-28. Kagan, S., & Carlson, H. (1975). Developmental and adaptive assertiveness in Mexican and United States children. Develop- mental Psyphology, 11, 71-78. 130 Karno, M. (1966). The enigma of ethnicity in a psychiatric clinic. Archives pf Genaral Psypniatry, 15, 516-520. Kerlinger, F. N., & Pedhazur, E. J. (1973). Multipla regression in behavioral rassarch. New York: Holt, Rinehart & Winston. Kimble, C. E., Marsh, N. B., & Kiska, A. C. (1984). Sex, age, and cultural differences in self-reported assertiveness. Esypnp; logipal Reports, ED, 419-422. Kirk, R. E. (1968). Experimental design: Propeduras for the pehavipral sciences. Belmont, CA: Brooks/Cole. Knight, G. P., Kagan, S., Nelson, W., & Gumbiner, J. (1978). Acculturation of second- and third-generation Mexican-American children. Journal of Cross-Cultural Psycholpgy, 2, 87-97. LeVine, E., & Franco, J. N. (1981). A reassessment of self- disclosure patterns among Anglo-Americans and Hispanics. Journal of Counseling Psyphology, ZD, 522-524. Littlefield, R. P. (1974). Self-disclosure among some Negro, White, and Mexican-American adolescents. ,Journal of Counseling Psyghology, 21, 133-136. Lujan, S., & Zapata, J. T. (1983). Personality differences among Mexican-American college freshmen. Journal of Collega Studeni Personnel. Lynch, M. D., & Huntsberger, D. V. (1976). em t f t i l inferanca for avaluation and psychplogy. Boston: Allyn & Bacon. Madsen, W. (1973). lhe Mexican-Americans of South Texas. San Francisco: Holt, Rinehart & Winston. Mahaffey, M. B. (1986). Locus of control in Mexican, Central, and South American adults. Dissertation Abstracts, 11, 5059-B. Meehl, P. E. (1958). Clinical versus statistisal predictions. Minneapolis: University of Minnesota Press. Mehrens, W. A., & Lehmann, I. J. (1984). Measurangnt and avalua- Linn. New York: Holt, Rinehart & Winston. Molina, R. A., & Franco, J. N. (1986). Effects of administrator and participant sex and ethnicity on self-disclosure. qurna1 pf Counssling and Development, ED, 160-162. 131 Rogers, 0. P. B. (1971). Personality traits and academic achieve- ment among Mexican-American students. ' r i a , 2180-A. Ross, C. E., Mirowsky, J., & Cockerham, W. C. (1983). Social class, Mexican culture, and fatalism: Their effects on psychological distress. WWW 223751121991. 1.1. 383-399. Rotter, J. B., Seeman, M., & Liverant, S. (1962). Internal versus external control of reinforcement: A major variable in behavior therapy. In N. F. Washbourne (Ed.), Dssisipnsy valuas, ang grpups (pp. 473-516). London: Pergamon. Scott, J. D., & Phelan, J. G. (1969). Expectancies of employable males regarding source of control of reinforcement. Psycho- logigal Reporis, 15, 911-913. Soza, A. M. (1982). A cross- and intra-cultural perspective of the attitudes of the Mexican-American. Dissartaiipn Abstrasts, 5;, 2046-A. Tatsuoka, M. M. (1971). Multivarieni analysis; Teshniguss fpr t'ona ch 1 ' r . New York: John Wiley & Sons. Wolthuis, R. L. (1987). li e nd de re f en n . l t'on chot er 's s’ ' d em n lie - ngnr. Unpublished doctoral dissertation, Michigan State University. Young, T. W., & Shorr, D. N. (1986). Factors affecting locus of control in school children. G n ' S ’ 1 and Gen W. uz. 407-417. "Illllllllllllllll