R ESEARCH R EPORT

Cultural Identity, Explanatory Style, and in Navajo Adolescents

TRACI R. RIECKMANN Oregon Health and Science University MARTHA E. WADSWORTH University of Denver DONNA DEYHLE University of Utah

This study investigated the interrelationships among cultural identity, explanatory style, and depression in Navajo adolescents. A total of 332 (197 female, 135 male) Navajo adolescents completed 7 self-report measures. These data were used to create, via struc- tural equation modeling, a series of factor models and full structural models. Analyses indicated that current factor structures for explanatory style and depression are ad- equate for use with Navajo adolescents. Increased control and predictability and limited duration of stressful encounters were both predictive of decreased symptoms of depres- sion. Higher levels of Navajo cultural identity had a modest effect in terms of reducing depression. Other factors, such as perceived discrimination and urban/reservation do- micile, are important to study to provide an increased understanding of depression among Navajo adolescents. • American Indian • adolescent • depression • cultural identity

A growing body of literature exists regarding ans. We currently lack basic information the nature, causes, consequences, and cor- about symptomatology, theoretically sound relates of mental illness among European conceptual models, and interventions for American youth, but very little is known American Indian adolescents. Adverse socio- about mental illness among American Indi- demographic factors including poverty, vio-

• Traci R. Rieckmann, Department of Public Health and Preventive Medicine, Oregon Health and Science University; Martha E. Wadsworth, Department of , University of Denver; Donna Deyhle, Department of Education, Culture, and Society, University of Utah. Correspondence concerning this article should be addressed to Traci R. Rieckmann, Depart- ment of Public Health and Preventive Medicine, Oregon Health and Science University, 3181 S.W. Sam Jackson Park Road, CB 669, Portland, OR 97239. E-mail: [email protected]

Cultural Diversity and Ethnic Minority Psychology Copyright 2004 by the Educational Publishing Foundation Vol. 10, No. 4, 365–382 1099-9809/04/$12.00 DOI: 10.1037/1099-9809.10.4.365 365 366 R IECKMANN,WADSWORTH, AND D EYHLE lence, and cultural trauma (forced reloca- & Leach, 1995). For example, it has been tion and acculturation, societal prejudice, suggested that African Americans who have and systematic genocide) suggest that a strong, positive African American identity American Indians are at heightened risk for are likely to have better mental health than mental illness and chronic stress (Hawkins those whose racial identity is less positive & Blume, 2002; Paniagua, 1994; SixKiller- (Butler, 1975; Pyant & Yanico, 1991). This Clarke, 2002). Depression is the most com- relationship has not been demonstrated em- mon diagnosis among teenage girls who pirically within American Indian communi- seek services through Indian Health Service ties. Two significant factors in the absence of mental health outpatient clinics (Beiser & research related to cultural identity and Attneave, 1982). Learning problems, con- mental health in these communities are duct disorders, substance abuse, running within-group heterogeneity (i.e., differences away, and suicide attempts are also reported among individuals and tribes) and the con- at very high rates and are thought to be con- textual nature of ethnicity. Thus, ethnic sequences of depression and alienation in identity may change depending on environ- American Indian adolescents (Choney, Ber- mental conditions, specific events or experi- ryhill-Paapke, & Robbins, 1995; U.S. Office ences, individual processes, and participa- of Technology Assessment, 1990). Finally, tion in political and social institutions interpersonal stress and depression have (Stanfield & Dennis, 1993). been identified as risk factors for suicide Navajo cultural identity is understood to among American Indian youth (Dinges & incorporate the values, beliefs, and lifestyle Duong-Tran, 1993; Dinges & Joos, 1988). choices that protect Navajo individuals Although these limited data suggest that throughout their lives (White, 1998). Com- depression and its consequences may be a posed of family, spiritual, and environmen- significant mental health problem, research tal components, this identity structure on depression in American Indian adoles- guides decision making and interactions. cents suffers from two primary limitations The goal of Navajo cultural identity is to re- that prevent firm conclusions in this regard. inforce the teachings of Navajo forefathers First, studies focusing on American Indians and foremothers and to “make life unique have tended to adopt unmodified conceptu- and beautiful so that Navajo individuals may alizations of mental illness developed for Eu- see the corn pollen road of life and walk on ropean American samples and have failed to the Holy Peoples Path of Life” (Aronilth, examine cultural variables that might affect 1991, p. 20). More specifically, Navajo cul- depression in American Indian youth (Jose tural identity has been defined as including et al., 1998). Second, significant variations three key components: family, spirituality, exist among American Indian tribes, limit- and environment. The family component, ing the generalizability of findings and sug- referred to as K’ei (relations) among the Na- gesting the need for tribe-specific, locally vajo, signifies an individual’s personal iden- based projects. In the current study, we fo- tity (belonging to a family) and his or her cused on adolescents living on a reservation responsibility within the community that he in the Southwest. or she represents (White, 1998). The spiri- tual component establishes an individual’s connection to spirits through his or her clan, Indian name, involvement in traditions Cultural Identity and ceremonies, and personal relationships with Navajo deities and divine nature (Rei- Cultural identity is an important factor to chard, 1950). Finally, the environmental consider regarding the mental health status component captures the importance of re- of ethnically diverse individuals (Garza & spect for, use of, and appreciation for the Lipton, 1984; Helms, 1992; Rowe, Behrens, natural processes of the universe (Aronilth, N AVAJO E XPLANATORY S TYLE AND D EPRESSION 367

1991). Thus, the three components interact suggests that there is no direct translation to form the basis of a Navajo individual’s for depression. Navajo language and culture worldview. According to Aronilth (1994), focus on mental health or hozho nashado, lack of cultural identity among Navajo youth which refers to having the ability to “love has a significant impact on rates of depres- and work” (White, 1998). Thus, descriptions sion and anxiety. of low mood, irritability, and other symp- toms related to depression are often ob- scure, in that individuals manifest, experi- ence, and interpret their symptoms in Depression and the Navajo unique, culturally bound ways. Although research suggests higher than Defining depression is profoundly complex, average rates of suicide, drug and alcohol especially in the case of adolescents from problems, physical abuse, depressive symp- non-Anglo cultural backgrounds. According tomatology, and alienation from family and to Topper (1987), many Navajo believe that community among American Indian youth spiritual imbalance is the cause of illness. (Beauvais, 1992; LaFromboise & Howard- Thus, the etiology and understanding of de- Pitney, 1995; Manson, Beals, Dick, & Duclos, pression from a Navajo perspective may be 1989), there is an absence of current, com- quite different than is the case from a White prehensive depression prevalence rates Western perspective. Reichard (1950) noted (Manson, Ackerson, Dick, Baron, & Flem- that, among many Navajo, mental disease is ing, 1990). Among Navajo youth, there is a identified as the consequence of breaking paucity of information regarding depres- taboos. Thus, when individuals do not re- sion, substance abuse, and mental health spect restrictions and expectations, prob- status. Participative opinions, however, lems may surface. In general, for many Na- point toward high prevalence rates. For ex- vajo it is understood that illness will occur ample, most adolescents who seek mental when the clockwise rotating wind within an health treatment at Navajo area Indian individual (“wind soul”) is out of harmony Health Services facilities are diagnosed as with other spiritual forces in the universe depressed (White, 1998). Evidence also sug- (Topper, 1987). gests that depression among Navajo adoles- According to the Navajo way, a person cents contributes to suicide, suicide at- may suffer symptoms of depression because tempts, and delinquency (Grossman, he or she is out of balance and in need of Milligan, & Deyo, 1991). Depression is con- care (J. M. Martin, personal communica- sistently cited in the literature on American tion, May 11, 1999). When Navajo individu- Indian youth as a regular, almost common als neglect their responsibilities to spiritual occurrence, yet there is little empirical evi- beings, they may find themselves sick, anx- dence for this assertion and almost no re- ious, unproductive, or depressed. They may search regarding the etiology of this disor- also suffer from mood disturbance when der. In addition, little is known about they are shunned by others because of their familial, sociocultural, and other variables lack of adherence to expectations. Accord- that probably play a role in the development ing to J. M. Martin (personal communica- and maintenance of the disorder (LaFrom- tion, May 25, 1999), these expectations are boise, 1988). sets of beliefs related to how “Navajo” a per- son is, as well as his or her place in Navajo society. Thus, individuals may be shunned Explanatory Style by others if cultural and spiritual expecta- tions are violated, leading to isolation, de- pression, and anxiety. Further, interpreta- According to Bechtold, Manson, and Shore tion of the term from English to Navajo (1994), “many, if not most, Indian/Native 368 R IECKMANN,WADSWORTH, AND D EYHLE children and adolescents live in environ- with more effective coping and positive well- ments and under circumstances involving being. Control refers to individuals’ percep- significant stress” (p. 101). Rates of aggres- tions regarding their ability to influence or sion, violent crime, and accidental deaths master a stressful encounter. Perceived con- are higher on reservation lands than in the trol is associated with improved psychologi- general U.S. population (Centers for Dis- cal functioning (Deuser & Anderson, 1995; ease Control and Prevention, 2003; Robin, Jenkins & Pargament, 1988). Beliefs indi- Chester, Rasmussen, Jaronson, & Goldman, viduals have about the controllability of a 1997). American Indian adolescents also stressful encounter influence their decisions face numerous chronic stressors, including to master or change circumstances (Com- poverty, discrimination, and forced accul- pas, Banez, Malcarne, & Worsham, 1991). turation. Although significant variance sur- Novelty is simply one’s familiarity with a situ- rounds the definition and measurement of ation. Again, there is a limited amount of stress, this construct continues to resurface empirical evidence regarding the psycho- in the literature on American Indian youth. logical impact of perceptions of novelty. Du- According to Lazarus and Folkman ration is the length of exposure to a stressful (1984), a transactional model can be used to encounter. It has been established that in- understand the interaction among stressful creased exposure or duration is related to events and the perceptions or explanations poorer psychological functioning over time individuals have about these events. When (Lazarus & Folkman, 1984). Causality is an people find themselves in a stressful envi- individual’s explanation regarding what led ronment, they are in a process of constantly to or prompted a particular event. reappraising or explaining their relation- An extensive body of literature suggests ships with that environment. Explanatory that a negative explanatory style is associated style refers to attributions of what caused a with self-reported and clinical depression stressful event and appraisals made about among Anglo adolescents (Joiner & Wag- the situation and its potential relevance. Ac- ner, 1995). More depressed children are cording to Seligman and Peterson (1986), likely to explain or appraise failure experi- attributional style can be defined as the ha- ences as internal or caused by them (“It’smy bitual manner in which a person explains fault”), as long lasting, and as occurring the causes of positive and negative events. across situations (global). These same chil- The literature on attributions suggests three dren appear to attribute successful events to dimensions: internal versus external, stable external (“Someone else made it happen”) versus unstable, and global versus specific and event-specific (“It’s just this situation”) (Joiner & Wagner, 1995). causes. Appraisal is defined as a nonstatic evalu- Overall, explanations made with regard ation of an event that is taking place. The to an individual’s interaction with his or her current literature identifies six appraisal di- environment are likely to be culturally mensions: (a) salience, (b) predictability, driven. According to Deyhle and Swisher (c) control, (d) novelty, (e) duration, and (1996), cultural beliefs, attitudes, and be- (f) causality (Vitaliano, Russo, Weber, & Ce- haviors can be observed in communication lum, 1993). More specifically, salience can patterns, interrelational styles, and social val- be described as the personal importance of ues. It appears likely that cultural identity an event. If the stressful encounter is not influences an individual’s explanatory pro- salient, then there will be little or no emo- cess. These constructs also correspond to tion, and distress is unlikely. Predictability is factors proposed in the indigenous stress understood as anticipation of a stressful and coping model, which expands the vul- event, and although this dimension is not nerability hypothesis and work of Dinges well researched, it appears that decreasing and Joos (1988) to include cultural buffers uncertainty of stressful events is associated as moderators between stress and health sta- N AVAJO E XPLANATORY S TYLE AND D EPRESSION 369 tus (Walters, Simoni, & Evans-Campbell, examined as a critical construct within An- 2002). Thus, stressful experiences such as glo populations. The present study exam- loneliness or abandonment may be ap- ined the constructs of cultural identity, ex- praised or explained differently by Ameri- planatory style, and depression in Navajo can Indians than by European Americans, youth residing in the Navajo Nation. Both given their significant history of oppression the absence of literature on and the lack of and collectivistic orientation. According to operationalization of these constructs in O’Nell (1996), loss of interconnectedness American Indian communities point to the and attachment is extremely significant, be- need for explanatory and confirmatory ap- cause history has threatened these “Indian” proaches designed to provide a better un- ways of being. Therefore, Navajo youth may derstanding of their factor structures. Fur- make causal attributions or appraisals re- ther, in keeping with current theoretical lated to loss or separation that are influ- conceptualizations of stress and the indig- enced by this history. enous stress and coping model, we exam- Cultural beliefs may also influence ex- ined relationships among explanatory style, planatory processes in terms of causality and depression, and Navajo cultural identity control. According to Frisbie (1992), tradi- (Walters et al., 2002). In hypothesized par- tional Navajo religion is concerned with liv- tial and full structural models, cultural iden- ing in harmony with supernatural powers tity served as an independent variable, and (i.e., the universe is filled with living forces explanatory style and depression were de- that have innumerable powers allowing pendent variables. Thus, on the basis (in them to act both against and in favor of hu- part) of the indigenous stress and coping man beings). From this perspective, it seems model, Navajo youth with strong cultural that “powerful others” may shift control ap- identities are likely to have more positive praisals toward an external response set. In explanatory styles and should therefore ex- addition, these expectations and cultural perience fewer symptoms associated with teachings affect individuals’ perceptions re- depression. garding what led to or prompted an event (causality). According to Topper (1987) and J. M. Martin (personal communication, May Method 25, 1999), causation may be the result of neglecting one’s responsibilities or of dis- Participants harmony associated with other spiritual forces in the universe. This imbalance may Three hundred thirty-two Navajo youth (135 be connected with family, community, and male, 197 female) participated in this study. environmental factors that Navajo youth One hundred sixteen (35%) of the partici- may or may not be directly aware of without pants were in the 10th grade, 125 (38%) seeking help or consultation through a were in the 11th grade, and 91 (27%) were “medicine person” or “hand trembler” (di- in the 12th grade; all were enrolled in public viner). Unfortunately, although theoretical- school in the Navajo Nation. Participants ly driven, these relationships have not been ranged in age from 14 to 20 years, with 41% examined empirically. 14 to 16 years of age and 59% 17 to 20 years Further examination of the character of age. Participant lists were generated and origin of depression among American through a systematic random sample strati- Indians is the foremost step in understand- fied according to gender and grade. Re- ing the meaning of this concept and poten- searchers met with selected students and tial directions for intervention (Shore & sent home consent forms to be signed by Manson, 1981). Although there are several parents or guardians. Students who re- well-established determinants of depression, turned signed consent forms then com- explanatory processes have been repeatedly pleted written questionnaires at school. Par- 370 R IECKMANN,WADSWORTH, AND D EYHLE ticipants received two coupons for meals at a and it is based on the ideals, values, and local restaurant after completing their re- standards of the Hozhooji Nahagha (“Beauty search materials. Way Ceremony”). The NCIM contains six scales measuring family, spiritual, and envi- ronmental knowledge and family, spiritual, Measures and environmental attitudes. Internal con- BIOGRAPHICAL QUESTIONNAIRE. The bio- sistency estimates for the instrument ranged graphical questionnaire contained 31 self- from .82 to .95 in this sample. report items and was used to capture demo- graphic data, including age, class standing, CHILDREN’S DEPRESSION INVENTORY. The gender, religious preference, and acting-out Children’s Depression Inventory (CDI; Ko- behaviors. vacs, 1985) is a 27-item self-report question- naire designed to assess cognitive, behavior- DIMENSIONS OF STRESS SCALE. The Dimen- al, and neurovegetative signs of depression sions of Stress Scale (DSS) is a 24-item self- in children. Although the CDI is not de- report questionnaire designed to assess ap- signed as a diagnostic tool, the instrument’s praisals of personal relevance and control, author suggests a cutoff score of 11 for de- stressor properties, and self-attributions (Vi- termining depression. According to Reyn- taliano et al., 1993). Each part of the scale olds (1986), the CDI has an alpha reliability begins with a stated problem or stressful coefficient of .87. The internal consistency event that the participant is asked to imag- was .87 for the total CDI scale used in this ine while responding to statements regard- analysis. ing his or her perspective and response to the stressor. For this project, two vignettes MINNESOTA MULTIPHASIC PERSONALITY INVEN- were presented, one of a family argument TORY. Participants also completed the Min- and another prompting appraisal related to nesota Multiphasic Personality Inventory– poor academic performance. The six DSS Adolescent Form (MMPI-A; Butcher et al., scales (8 items each) are (a) duration (the 1992). In view of the primary purpose of this perceived length of the stressor), (b) pre- study, Scale 2 (Depression) and the depres- dictability (the level to which the person can sion content scale were used to assess the predict the outcome of the event), (c) con- correlation between components of Navajo trol (the extent to which the person has con- cultural identity and depression. Internal trol over the outcome), (d) causality (the consistency estimates have been shown to extent to which the stressor is perceived as range from .40 to .89 across the basic validity being attributed to the person), (e) salience and clinical scales (Butcher et al., 1992). (personal relevance), and (f) novelty (the More specifically, in the normative sample, extent to which the event is new). In the Scale 2 (Depression) had internal consis- present sample, internal consistency esti- tency coefficients of .65 for boys and .66 for mates for the DSS ranged from .44 to .70. girls. Test–retest reliability coefficients ranged from .62 to .82 across the content NAVAJO CULTURAL IDENTITY MEASURE. Navajo scales; the retest coefficient of the depres- cultural identity was assessed with the Navajo sion content scale, which was used in this Cultural Identity Measure (NCIM; White, study, was .82. Internal consistency coeffi- 1998). This instrument, developed by a Na- cients in this sample were .60 for the clinical vajo researcher, is composed of 152 items scale and .75 for the content scale. covering three dimensions of Navajo cul- tural identity: family, spirituality, and envi- DSM–IV QUESTIONNAIRE FOR DEPRESSION. ronment. The NCIM was developed through The Diagnostic and Statistical Manual of Men- personal interviews and interpretations of tal Disorders (4th ed.; DSM–IV; American Psy- Navajo philosophical and spiritual material, chiatric Association, 1994) questionnaire N AVAJO E XPLANATORY S TYLE AND D EPRESSION 371 was developed specifically for this project. to which sample variance and covariance are This 37-item true–false measure was based reproduced by the proposed model; values on DSM–IV criteria for major depression. greater than .90 indicate adequate fit The format of the questionnaire allowed stu- (Bentler, 1995; Bollen & Long, 1993). Theo- dents to simply circle true or false next to retically sound measurement models with symptoms listed after the prompt “Every day adequate statistical fit were used to generate over the last two weeks I have....” Internal partial and full structural models in an at- consistency for the scale was good, with the tempt to further explore relationships alpha coefficient calculated at .83. among the latent variables.

Data Analysis Results Structural equation modeling (SEM) was the primary analytic strategy used in this Subscale scores and internal consistency es- study. SEM allows for multiple measures of a timates were generated for each of the in- given construct and accounts for measure- struments used in the study. Means, stan- ment error while allowing for analysis of me- dard deviations, and internal consistency diating relationships (Robitschek & Kashu- reliability estimates used are reported in beck, 1999). A confirmatory factor analysis Table 1 for the five scales from the DSS, the with a maximum-likelihood estimation CDI total scale, the DSM–IV question- method was programmed with EQS struc- naire total scale, the MMPI-A clinical and tural modeling software (Bentler, 1995). content scales for depression, and six NCIM Model fit was evaluated and adjustments subscales. made to the factor structure to determine the best model for the data. The data were examined for normality of Navajo Cultural Identity Measure Model distribution, adequacy of covariance, and minimal standards for sample size. This pro- A two-factor correlated Navajo cultural iden- cess revealed that some of the measured tity model was proposed and tested. The variables were not distributed normally, a model, presented in Figure 1, included cul- violation of the normality assumption re- tural knowledge and cultural attitude. Each quired for SEM (Tabachnick & Fidell, of these factors was based on three subscales 1996). As a result, linear z-score transforma- measuring family, environmental, and spiri- tions were used to equate the variance of the tual domains of Navajo cultural knowledge variables while retaining each indicator’s and Navajo attitudes. A modest correlation distributional qualities. Proposed measure- of .47 was observed between the knowledge ment models were tested for each factor in- and attitude factors. In the case of both cluded in the final structural model. Good- scales, the highest loadings were for the ness of fit was determined through chi- spiritual subscales (.82 for knowledge and square analyses, the Bentler and Bonett .90 for attitudes). The indexes for this (1980) nonnormed fit index (NNFI), and model suggested an excellent fit, ␹2(8, N = the Bentler comparative fit index (CFI). 332) = 43.30, p = .0000, NNFI = .92, CFI = Even though chi-square statistics have a ten- .96. The chi-square value was significant but dency to reject adequate models with large less than twice the number of degrees of sample sizes, chi-square tests comparing the freedom for the model, a common bench- covariance matrix of the specified model mark for evaluating fit with chi-square tests. with the observed variables are reported as a The other fit indexes also indicated a good matter of standard practice (Hu & Bentler, fit between the data and the proposed 1995). The CFI is an estimate of the extent model of Navajo cultural identity. 372 R IECKMANN,WADSWORTH, AND D EYHLE

TABLE 1 Subscales: Number of Items, Means, Standard Deviations, and Alpha Coefficients

Scale and subscale No. of items M SD ␣

Dimensions of Stress Scale Control 8 21.47 5.10 .67 Salience 8 19.86 5.62 .70 Duration 8 19.49 4.94 .61 Causality 8 19.27 4.66 .62 Predictability 8 16.63 3.83 .42 Depression CDI total 27 10.00 7.56 .87 DSM–IV 24 7.53 4.99 .83 MMPI-A clinical 50 20.29 5.05 .60 MMPI-A content 24 9.83 4.42 .75 Navajo Cultural Identity Measure Family attitude 27 106.67 16.69 .92 Family knowledge 26 12.06 6.37 .89 Environmental attitude 18 67.16 14.34 .95 Environmental knowledge 18 5.00 3.91 .82 Spiritual attitude 32 118.00 23.46 .95 Spiritual knowledge 31 8.76 5.67 .86

Note. CDI = Children’s Depression Inventory; DSM–IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); MMPI-A = Minnesota Multiphasic Personality Inventory Adolescent Form.

Explanatory Style Measurement Models planatory style is presented in Figure 2. This Confirmatory factor analysis was used to ex- model included two factors: (a) positive ex- amine the fit of several proposed measure- planatory style (consisting of the DSS dura- ment models for the explanatory style con- tion, predictability, and control scales) and struct. Factor loadings indicated that the (b) negative explanatory style (consisting of DSS novelty subscale made little or no con- the DSS causality and salience scales). Posi- tribution to the hypothesized factor; the tive explanatory style comprised stressful nonsignificant loading and the low alpha co- event perceptions relating to a shorter dura- efficient (.44) suggested that this subscale tion and a predictable encounter that could did not contribute common variance to the be influenced by an individual (control). construct of explanatory style. It was Negative explanatory style included the sa- dropped from the model. lience or increased importance of the stress- The final measurement model for ex- ful situation and personal responsibility

Figure 1. Cultural identity model: ␹2(8, N = 332) = 43.30, p = .00, nonnormed fit index = .92, comparative fit index = .96. CI Attitude = cultural attitude; CI Know = cultural knowledge. N AVAJO E XPLANATORY S TYLE AND D EPRESSION 373

Figure 2. Explanatory style model: ␹2(3, N = 332) = 4.41, p = .22, nonnormed fit index = .98, comparative fit index = 1.00.

(causality). Two observed variables, causality Depression Measurement Models and control, were allowed to load on each of Figure 3 presents a one-factor depression the factors. The highest loading for the posi- model that included four observed variables tive explanatory style factor was observed for from the DSM–IV questionnaire, total CDI, the duration subscale(.99), and the highest and MMPI-A content and clinical scales. The loading for the negative explanatory style DSM–IV questionnaire had the highest load- factor was observed for the salience sub- ing on the depression factor (.84), and the scale(.71). The cross loadings of causality MMPI-A clinical scale had the lowest loading and control in this model may indicate that (.28). The fit indexes, ␹2(2, N = 332) = 16.60, increases in personal responsibility translate p = .0003, NNFI = .87, CFI = .96, indicated an to more positive explanations for events adequate fit between the proposed model among some Navajo youth. and the data. These results also suggest that percep- tions of control influence both positive and Partial and Full Latent Structural Models negative styles for appraising situations. Al- lowing these two variables to cross load on Figure 4 presents a latent variable model of the two explanatory style factors allowed the the relationship between explanatory style model to converge and provided the best fit and depression in which it was hypothesized with the sample data. The fit indexes, ␹2(3, that individuals with positive explanatory N = 332) = 4.41, p = .22, NNFI = .98, CFI = styles would have lower depression scores. 1.00, suggested an excellent fit between the The standardized regression coefficient of data and the proposed model. −.50 indicated that an increase in positive 374 R IECKMANN,WADSWORTH, AND D EYHLE

Figure 3. Depression model: ␹2(2, N = 332) = 16.60, p = .00, nonnormed fit index = .87, comparative fit index = .96. CDI = Children’s Depression Inventory; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); MMPI-A = Minnesota Multiphasic Personality Inventory Adolescent Form. explanatory style is associated with a de- data and the proposed latent structural crease in self-reported depressive symptoms. model. The negative explanatory style factor was The relationship between cultural iden- also tested but was not significant (stan- tity and explanatory style was tested with the dardized regression coefficient: .02), and latent variable model presented in Figure 5. the Wald test suggested dropping this The knowledge factor had a standardized path as a result of the lack of contribution to regression coefficient of .28, whereas the at- the structural model. The fit indexes, ␹2(24, titude factor predicted the positive explana- N = 332) = 89.99, p = .00, NNFI = .87, CFI = tory style factor with a standardized regres- .91, suggested an acceptable fit between the sion coefficient of .20. The fit indexes,

Figure 4. Explanatory style (ES) and depression model: ␹2(24, N = 332) = 89.99, p = .00, nonnormed fit index = .87, comparative fit index = .91. MMPI-A = Minnesota Multiphasic Personality Inventory Adolescent Form; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); CDI = Children’s Depression Inventory. N AVAJO E XPLANATORY S TYLE AND D EPRESSION 375

Figure 5. Cultural identity and explanatory style model: ␹2(39, N = 332) = 103.41, p = .00, nonnormed fit index = .92, comparative fit index = .94. CI ATT = cultural attitude; CI KNOW = cultural knowledge.

␹2(39, N = 332) = 103.41, p = .00, NNFI = .92, standardized regression coefficient of −.11. CFI = .94, suggested an adequate fit with the The fit indexes, ␹2(83, N = 332) = 230.50, p sample data. The findings point to a signifi- = .00, NNFI = .88, CFI = .91, indicated an cant but weak relationship between attitude acceptable goodness of fit between the and situational explanatory factors; how- sample data and the latent variable model. ever, the relationship between cultural The relationship between cultural attitude knowledge and positive explanatory factors and the positive explanatory style factor was was stronger for Navajo youth. significant, as was the relationship between The final full latent structural model cultural knowledge and negative explana- shown in Figure 6 tested the relationships tory style. The relationship between cultural among (a) cultural identity, (b) explanatory attitude and depression in Navajo adoles- style, and (c) depression. In this final model, cents was significant but weak, most likely the highest standardized regression coeffi- owing to the effect of positive explanatory cient in relationship to depression was posi- style mediating this relationship. tive explanatory style, at −.47. This relation- ship suggests that as Navajo adolescents experience increased predictability and con- Discussion and Conclusion trol in regard to temporary stressful events, they are less likely to report symptoms of depression. In addition, cultural attitude In this study, we examined the factor struc- predicted positive explanatory style, with a ture and interrelationships of the constructs regression coefficient of .23. In turn, cul- of cultural identity, explanatory style, and tural attitude predicted depression, with a depression in Navajo youth. Previous work 376 R IECKMANN,WADSWORTH, AND D EYHLE

Figure 6. Full latent structural model: ␹2(83, N = 332) = 230.50, p = .00, nonnormed fit index = .88, comparative fit index = .91. CI Attitude = cultural attitude; CI Know = cultural knowledge; ES = explanatory style; CDI = Children’s Depression Inventory; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); MMPI-A = Minnesota Multiphasic Personality Inventory Adolescent Form. suggested that depression, substance abuse, sive symptoms” (p. 405). As reported by the learning problems, suicide, and other self- U.S. Office of Technology Assessment destructive behaviors are significant prob- (1990), depression and alienation are sig- lems for American Indian youth. Although nificant problems for American Indian ado- sound prevalence rates of depression are lescents, and several studies have suggested not available for Navajo youth, research sug- high rates of self-reported depression (Bach- gests that 2%–5% of European American man et al., 1991). In the present sample of adolescents suffer from major depressive 135 male and 197 female Navajo adolescents disorder (Lewinsohn, Hops, Roberts, Seeley, from four different high schools, 14% of the & Andrews, 1993) and many of these youth participants exceeded the CDI cutoff score experience recurring symptoms throughout of 20 suggested for use with unselected or their adolscence. (Lewinsohn, Clarke, See- general samples (Kovacs, 1992). ley, & Rohde, 1994). According to Nolen- Further analyses of the depression con- Hoeksema, Girgus, and Seligman (1992), struct for the structural modeling portion of “At any given time, approximately 10%–15% this study included observed variables from of children in the general population will the MMPI-A (clinical and content scales), report a moderate to severe level of depres- CDI, and DSM–IV questionnaire. The final N AVAJO E XPLANATORY S TYLE AND D EPRESSION 377 depression construct supported a one-factor wherein adapting to the situation is model based on these four observed vari- required. ables that corresponded with the hypoth- Furthermore, there is an extensive litera- esized model. Although results were consis- ture suggesting that a negative attributional tent with expectations, it is significant to style including internal, stable, and global note that the European American under- attributions is associated with self-reported standing of the etiology and experience of and clinically diagnosed depression among depression employed in this study may not Anglo adolescents (Joiner & Wagner, 1995). have fully captured additional symptoms or The self-responsibility or causality variable unique conceptualizations relevant to our used in this study may have captured the sample. For example, symptoms of depres- internal locus and the increased salience of sion may manifest as stomachaches, other the stressful encounter that are likely to re- medical ailments, or even spiritual discon- sult in elevated levels of distress (Vitaliano et tent and lack of harmony and balance. Fur- al., 1993; Weiner, 1980). Thus, these ob- ther, there is no direct translation in the served variables correspond with a negative Navajo language for “depression.” Conse- explanatory set. In addition, causality is de- quently, some of the meanings inherent in scribed by theory as an explana- descriptions of symptoms may also be tion made by an individual regarding re- unique to this population. sponsibility for an event, as well as the The final explanatory style construct re- individual’s ability to influence the event. sulted in two factors: (a) positive explana- According to Vitaliano and colleagues (1993), among European Americans, ap- tory style and (b) negative explanatory style. praisals of causality and control influence In this model, the duration variable from one another. The cross loading of these two the DSS had the highest loading. Further, constructs also corresponds with Weiner’s the control and causality variables loaded on (1980) assertion that causality itself may be both positive and negative explanatory sets, separated into locus, duration, and control- suggesting their significance in the explana- lability in the explanatory process. Navajo tory process. These findings and the associ- youth may experience increased personal ated positive explanatory factor coincide responsibility as positive because they per- with the available literature on depression in ceive greater influence or control over the which perceived control, shorter exposure event without seeking traditional healing to stressful events, and increased predictabil- through a medicine person or diviner, ity have been associated with more effective which would draw the problem to their fami- coping and positive well-being among Euro- ly’s attention and demand more resources. American youth (Deuser & Anderson, 1995; Thus, the results of this study suggest that Lazarus & Folkman, 1984). These results the relationships among control, causality, further correspond with the available litera- and explanatory style are also critical for ture on European American youth, accord- Navajo youth. ing to which perceived control plays a sig- In addition, a significant relationship was nificant role in decision making around found between positive explanatory style stressful encounters and is associated with and depression. More specifically, Navajo improved psychological functioning (Com- adolescents who perceive higher levels of pas, Orosan, & Grant, 1993; Jenkins & Par- control, shorter event durations, and more gament, 1988). According to Compas, Con- predictability in stressful encounters may be nor-Smith, Saltzman, Thomsen, and less likely to suffer from depression. The re- Wadsworth (2001), coping incorporates lationship between the positive style factor both primary control, wherein efforts are di- and depression accounted for a significant rected toward changing the situation or portion of the variance (25%) in the model, one’s emotions, and secondary control, suggesting that this factor is important in 378 R IECKMANN,WADSWORTH, AND D EYHLE conceptualizing explanatory processes and control, predictability, and duration of their associations with depression among stressful events. According to French and Navajo youth. This finding corresponds with Picthall-French (1998), the emphasis among literature on European American youth the Navajo on “oneness” with nature and the showing that attributional style is correlated traditional “corn pollen road of life” or with depression in the case of both depres- “beauty way” may serve to insulate those who sion diagnoses and self-reported depressive are fully enculturated from the “harsh reali- symptoms (Joiner & Wagner, 1995; Lewin- ties of their environment” (p. 104). Thus, as sohn, Joiner, & Rohde, 2001). Further, the the present findings suggest, increasing a results of this study correspond with previ- Navajo youth’s sense of cultural awareness ous studies showing that the interaction be- and identity may serve to decrease isolation tween negative events and explanatory style and increase skills and factors associated is predictive of depression among older ado- with resiliency and mental well-being. Ac- lescents but may not hold in younger popu- cording to Deyhle (1995), youth who main- lations (Nolen-Hoeksema et al., 1992). tain strong Navajo cultural ties are more suc- In this sample, the salience, or impor- cessful among their own society and in the tance, associated with a stressful encounter Anglo world. was less predictive of experiences of depres- These findings also suggest that in- sive symptomatology. This may have been creased knowledge of the Navajo belief sys- related to structural issues, including pov- tem may be associated with more personal erty, racism, substance abuse, depression, responsibility in relation to stressful events. and alienation, faced by American Indian For many Navajo, this knowledge is gleaned youth (U.S. Office of Technology Assess- over time through experience, interactions ment, 1990). Their history of facing mul- with cultural leaders, and participation in tiple stressors may lead Navajo youth to be- traditional events. Further, it is up to the come somewhat desensitized in terms of individual to live within cultural expecta- distinguishing how one stressful encounter tions once he or she has attained the knowl- differs from another. Further, it is possible edge (Farella, 1993). The Navajo youth in that the stressful situations used in this study this sample may have been responding in to prompt responses (family argument and terms of their individual developmental poor academic performance) were not as level; that is, they were not fully educated meaningful to Navajo youth as would have regarding all of the “ways” of the Navajo, yet been Navajo conceptions of stress such as an they were aware of the importance of obtain- “imbalance . . . precipitated by indulging in ing this knowledge. Perhaps Navajo youth excesses, having improper contact with dan- perceive the role of diagnosing and address- gerous powers, deliberately or unwittingly ing spiritual disturbances as the responsibil- breaking other rules, or by malicious inten- ity of their Navajo forefathers and foremoth- tions of unpredictable deities or evil humans ers, where they may turn when experiencing [witches]” (Frisbie, 1992, p. 460). mood disturbance or anxiety. These findings correspond with the re- Several limitations of the present study sults of the full structural model including warrant attention. For example, very little cultural identity, explanatory style, and de- literature is available regarding explanatory pression factors. The correlation between style, cultural identity, and depression across the cultural knowledge factor and negative ethnic groups, and there is a complete lack explanatory style was .27, whereas the cul- of literature regarding appraisals and Na- tural attitude factor had a correlation of .23 vajo youth. In addition, a portion of this with positive explanatory style. project was exploratory in nature; that is, Again, a decrease in self-reported de- culturally appropriate definitions of the con- pressive symptomatology was seen with in- structs described have not been clearly es- creased positive explanations regarding the tablished in the literature. Similarly, existing N AVAJO E XPLANATORY S TYLE AND D EPRESSION 379 instruments designed to capture these con- cultural identity so that these constructs may structs have little or no available data from a be appropriately operationalized. Finally, fu- Navajo sample, making the potential for cul- ture research should explore the sequence tural bias a significant factor in interpreting and interaction of explanations for, and re- findings (Dana, 1998). sponses to, stress as generated among Methodological limitations of this study American Indian adolescents. included the cross-sectional design and the self-report nature of the data. Although nec- essary to gain access to the sample popula- References tion, participants were identified only by number, eliminating the potential for track- American Psychiatric Association. (1994). Diag- ing participants over time. Consequently, nostic and statistical manual of mental disorders potential interpretations and findings are (4th ed.). Washington, DC: Author. limited; that is, a cross-sectional study can- Arnoilth, W. (1991). Foundation of Navajo culture. not determine whether increased Navajo Navajoland, AZ: Dine’ Community College. cultural identity is a consequence of or a Arnoilth, W. (1994). Dine’ bi bee ohoo’aah ba sila: predisposing factor for depression. Further, An introduction to Navajo philosophy. Tsaile, AZ: the self-report measures used may have re- Navajo Community College. flected response bias stemming from socially Bachman, J. G., Wallace, J. M., O’Malley, P. M., desirable responding, a lack of awareness of Johnston, L. D., Kurth, C. L., & Neighbors, H. W. (1991). Racial/ethnic differences in the mental and emotional processes partici- smoking, drinking, and illicit drug use pants were asked to identify, and attempts to among American high school seniors, respond in ways confirming or disconfirm- 1976–1989. American Journal of Public Health, ing what participants believed was the pro- 81, 372–377. posed hypothesis. Longitudinal prospective Beauvais, F. (1992). Characteristics of Indian studies that incorporate multidimensional youth and drug use. American Indian and representations of constructs (behavioral Alaska Native Mental Health Research, 5, 51–67. observations and ratings by significant oth- Bechtold, D. W., Manson, S. M., & Shore, J. H. ers) are necessary to confirm whether or not (1994). Psychological consequences of stress the relationships identified here are consis- among Native American adolescents. In tent across situations and developmental J. Yager (Ed.), Stress in psychiatric disorders (pp. stages. 101–116). New York: Springer. Overall, the findings of this study illus- Beiser, M., & Attneave, C. L. (1982). Mental dis- trate that there are important relationships orders among Native American children: among the constructs of cultural identity Rate and risk periods for entering treatment. and explanatory style and their potential in- American Journal of Psychiatry, 139, 193–198. fluence on depression in Navajo youth. Bentler, P. M. (1995). EQS structural equations Given the current understanding that his- program manual. Encino, CA: Multivariate Software. torical factors and cultural abuse may be re- lated to intergenerational mental health Bentler, P. M., & Bonett, D. G. (1980). Signifi- cance tests and goodness of fit in the analysis struggles for American Indians (McNeil, of covariance structures. Psychological Bulletin, Kee, & Zvolensky, 1999), further research 88, 588–606. regarding these dynamic processes is war- Bollen, K., & Long, J. (1993). Testing structural ranted. Subsequent research should exam- equation models. Newbury Park, CA: Sage. ine additional instruments to determine Butcher, J. N., Williams, C. L., Graham, J. R., their applicability to Navajo and American Archer, R. P., Tellegen, A., Ben-Porath, Y. S., Indian populations. Empirically sound stud- et al. (1992). Minnesota Multiphasic Personality ies are also necessary to provide an under- Inventory–Adolescent: Manual for administration, standing of the unique manifestation of ex- scoring, and interpretation. Minneapolis: Uni- planatory processes, depression, and Navajo versity of Minnesota Press. 380 R IECKMANN,WADSWORTH, AND D EYHLE

Butler, D. O. (1975). Psychotherapy: Implications Issues Among American Indians and Alaska of a Black-consciousness model. Psychotherapy: Natives, 1, 8–64 Theory, Research, and Practice, 12, 407–411. Farella, J. R. (1993). The main stalk: A synthesis of Centers for Disease Control and Prevention. Navajo philosophy. Tucson: University of Ari- (2003). Injury mortality among American In- zona Press. dian and Alaska Native children and youth— French, L. A., & Picthall-French, N. (1998). The United States, 1989–1998. Morbidity and Mor- role of substance abuse among rural youth by tality Weekly Report, 52, 697–699. race, culture, and gender. Alcoholism Quar- Choney, S. K., Berryhill-Paapke, E., & Robbins, R. terly, 16, 101–108. R. (1995). The acculturation of American In- Frisbie, C. J. (1992). Temporal change in Navajo dians. In J. M. Ponterotto, J. M. Casas, L. A. religion: 1868–1990. Journal of the Southwest, Suzuki, & C. M. Alexander (Eds.), Handbook 34, 457–514. of multicultural counseling (pp. 73–93). Thou- sand Oaks, CA: Sage. Garza, R. T., & Lipton, J. P. (1984). Foundations for a Chicano . In J. L. Mar- Compas, B. E., Banez, G. A., Malcarne, V., & Wor- tinez Jr. & R. H. Mendoza (Eds.), Chicano psy- sham, N. (1991). Perceived control and cop- chology (2nd ed., pp. 335–365). San Diego, ing with stress. Journal of Social Issues, 47, CA: Academic Press. 23–34. Grossman, D. C., Milligan, B. C., & Deyo, R. A. Compas, B. E., Connor-Smith, J. K., Saltzman, H., (1991). Risk factors for suicide attempts Thomsen, A. H., & Wadsworth, M. E. (2000). among Navajo adolescents. American Journal Coping with stress during childhood and of Public Health, 81, 870–874. asolescence: Progress, problems, and poten- tial in theory and research. Psychological Bul- Hawkins, E. H., & Blume, A. W. (2002). Loss of letin, 127,87–127. sacredness: Historical context of health poli- Compas, B. E., Orosan, P. G., & Grant, K. E. cies for indigenous people in the United (1993). Adolescent stress and coping: Impli- States. In P. D. Mail, S. Heurtin-Roberts, cations for psychopathology during adoles- S. E. Martin, & J. Howard (Eds.), Alcohol use cence. Journal of Adolescence, 16, 331–349. among American Indians and Alaska Natives (pp. 25–46). Bethesda, MD: National Insti- Dana, R. H. (1998). Understanding cultural identity tute on Alcohol Abuse and Alcoholism. in intervention and assessment. Thousand Oaks, CA: Sage. Helms, J. (1992). Race is a nice thing to have: A guide to being a White person or understanding the Deuser, W. E., & Anderson, C. A. (1995). Con- White persons in your life. Topeka, KS: Content trollability, attributions, and learned helpless- Communication. ness: Some methodological and conceptual problems. Basic and Applied Social Psychology, Hu, L., & Bentler, P. (1995). Structural equation 16, 297–318. modeling: Concepts, issues, and applications. In Hoyle (Ed.), Evaluating model fit (pp. Deyhle, D. (1995). Navajo youth and Anglo rac- 76–99). Thousand Oaks, CA: Sage. ism: Cultural integrity and resistance. Har- vard Education Review, 65, 403–444. Jenkins, R. A., & Pargament, K. I. (1988). Cogni- Deyhle, D., & Swisher, K. (1996). Research in tive appraisals in cancer patients. Social Science American Indian and Alaska Native educa- and Medicine, 26, 625–633. tion: From assimilation to self-determination. Joiner, T. E., & Wagner, K. D. (1995). Attribu- Review of Research in Education, 22, 113–194. tional style and depression in children and Dinges, N. G., & Duong-Tran, Q. (1993). Stress- adolescents: A meta-analytic review. Clinical ful life events and co-occurring depression, Psychology Review, 15, 777–798. substance abuse and suicidiality among Jose, P. E., D’Anna, C. A., Cafasso, L. L., Bryant, American Indian and Alaska Native adoles- F. B., Chiker, V., Gein, N., et al. (1998). Stress cents. Culture, Medicine and Psychiatry, 16, and coping among Russian and American 487–502. early adolescents. Developmental Psychology, 34, Dinges, N. G., & Joos, S. K. (1988). Stress, coping, 757–769. and health: Models of interaction for Indian Kovacs, M. (1985). Children’s Depression Inven- and native populations. Behavioral Health tory. Psychopharmacology Bulletin, 21, 995–998. N AVAJO E XPLANATORY S TYLE AND D EPRESSION 381

Kovacs, M. (1992). Children’s Depression Inventory: O’Nell, T. D. (1996). Disciplined hearth: History, Manual. North Tonawanda, NY: Multi-Health identity, and depression in an American Indian Systems. community. Berkeley: University of California LaFromboise, T. D. (1988). American Indian Press. mental health policy. American Psychologist, Paniagua, F. A. (1994). Assessing and treating cul- 43, 388–397. turally diverse clients: A practical guide. Thou- LaFromboise, T. D., & Howard-Pitney, B. (1995). sand Oaks, CA: Sage. The Zuni Life Skills Development Curricu- Pyant, C. T., & Yanico, B. J. (1991). Relationship lum: Description and evaluation of suicide of racial identity and gender-role attitudes to prevention program. Journal of Counseling Psy- Black women’s psychological well-being. Jour- chology, 42, 479–486. nal of Counseling Psychology, 38, 315–322. Lazarus, R. S., & Folkman, S. (1984). Stress, ap- Reichard, G. A. (1950). Navajo religion: A study of praisal and coping. New York: Springer. symbolism. New York: Bollingen Foundation. Lewinsohn, P. M., Clarke, G. N., Seeley, J. R., & Reynolds, W. M. (1986). A model for screening Rohde, P. (1994). Major depression in com- and identification of depressed children and munity adolescents: Age at onset, episode du- adolescents in school settings. Professional ration, and time to recurrence. Journal of the School Psychology, 1, 117–129. American Academy of Child and Adolescent Psy- Robin, R. W., Chester, B., Rasmussen, J. K., chiatry, 33, 809–818. Jaronson, J. M., & Goldman, D. (1997). Preva- Lewinsohn, P. M., Hops, H., Roberts, R., Seeley, lence, characteristics, and impact of child- J. R., & Andrews, J. A. (1993). Adolescent psy- hood sexual abuse in a southwestern Ameri- chopathology: I. Prevalence and incidence of can Indian tribe. Child Abuse and Neglect, 21, depression and other DSM-III-R disorders in 769–787. high school students. Journal of Abnormal Psy- Robitschek, C., & Kashubeck, S. (1999). A struc- chology, 102, 133–144. tural model of parental alcoholism, family Lewinsohn, P., Joiner, T., & Rohde. (2001). functioning, and psychological health: The Evaluation of cognitive diathesis-stress mod- mediating effects of hardiness and personal els in predicting major depressive disorder in growth orientation. Journal of Counseling Psy- adolescents. Journal of Abnormal Psychology, chology, 46, 159–172. 110, 203–215. Rowe, W., Behrens, J. T., & Leach, M. M. (1995). Manson, S. M., Ackerson, L. M., Dick, R. W., Racial/ethnic and racial consciousness: Baron, A. E., & Fleming, C. M. (1990). De- Looking back and looking forward. In J. M. pressive symptoms among American Indian Ponterotto, J. M. Casas, L. A. Suzuki, & C. M. adolescents: Psychometric characteristics of Alexander (Eds.), Handbook of multicultural the Center for Epidemiologic Studies Depres- counseling (pp. 218–235). Thousand Oaks, sion Scale (CES-D). Psychological Assessment: A CA: Sage. Journal of Consulting and Clinical Psychology, 2, Seligman, M. E. P., & Peterson, C. (1986). A 231–237. perspective on child- Manson, S. M., Beals, J., Dick, R. W., & Duclos, C. hood depression: Theory and research. In M. (1989). Risk factors for suicide among Indian Rutter, C. E. Izard, & P. B. Read (Eds.), De- adolescents at a boarding school. Public pression in young people: Developmental and clini- Health Reports, 104, 609–614. cal perspectives. (pp. 223–249). New York: Guil- McNeil, D. W., Kee, M., & Zvolensky, M. J. ford Press. (1999). Culturally related anxiety and ethnic Shore, J. H., & Manson, S. M. (1981). Cross- identity in Navajo college students. Cultural cultural studies of depression among Ameri- Diversity and Ethnic Minority Psychology, 5, can Indians and Alaska Natives. White Cloud 56–64. Journal, 2, 5–12. Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. SixKiller-Clarke, A. (2002). Social and emotional E. P. (1992). Predictors and consequences of distress among American Indian and Alaska childhood depressive symptoms: A 5-year lon- Native students: Research findings. Retrieved gitudinal study. Journal of Abnormal Psychology, December 2002, from http://www 101, 405–422. .indianeduresearch.net 382 R IECKMANN,WADSWORTH, AND D EYHLE

Stanfield, J. H., & Dennis, R. M. (1993). Race and (1993). The Dimensions of Stress Scales: Psy- ethnicity in research methods. New York: Sage. chometric properties. Journal of Applied Social Tabachnick, B. G., & Fidell, L. S. (1996). Using Psychology, 23, 1847–1878. multivariate statistics. New York: HarperCollins. Walters, K. L., Simoni, J. M., & Evans-Campbell, T. (2002). Substance abuse among American Topper, M. D. (1987). The traditional Navajo Indians and Alaska Natives: Incorporating medicine man: Therapist, counselor, and culture in an “indigenist” stress-coping para- community leader. Journal of Psychoanalytic digm. Public Health Reports, 117, S104–S117. Anthropology, 10, 217–249. Weiner, B. (1980). Human motivation. New York: U.S. Office of Technology Assessment. (1990). Holt, Rinehart & Winston. Indian adolescent mental health. Washington, White, K. (1998). Navajo adolescent cultural identity DC: U.S. Government Printing Office. and depression. Unpublished doctoral disserta- Vitaliano, P. P., Russo, J., Weber, L., & Celum, C. tion, University of Utah, Salt Lake City.