Resource ID#: 4878

Journal of Immigrant Heallh, Vol. 2, No.3, 2000 Acculturative Stress, Anxiety, and Depressionamong Mexican Immigrant Farmworkers in theMidwest United States

Acculturative Stress, Anxiety, and Depression amongMexican Immigrant Farmworkers in the MidwestUnited States

Joseph D. Hoveyl.3 and Cristina Magaiia2

No previous studies have examined the psychological functioning of immigrant farmworkersin the midwest United States. The purposes of the present study were threefold: (1) toassess the prevalence levels of anxiety and depression in a sample of Mexican immigrantfarmworkers in the midwest United States; (2) to explore the relationships among accultura-tive stress. anxiety. and depression; and (3) to examine the variables that best predict anxietyand depression. The overall sample revealed elevated levels of anxiety and depression.Immigrant farmworkers with heightened levels of acculturative stress reported high levelsof anxiety and depression. Family dysfunction. ineffective social support, low self-esteem,lack of choice in the decision to immigrate and live a migrant farmworker lifestyle. high~ducation levels. and low levels of religiosity were significantly associated with high levelsof anxiety and depression. The overall findings suggest that Mexican immigrant farmworkerswho experience elevated levels of acculturative 5tress may be "at risk" for experiencinghigh levels of anxiety and depression. The findings highlight the importance of establishingprevention and treatment services for migrant farm workers that aim to increase levels ofemotional support, self-esteem, and coping skills.

-KEY WORDS: migrant farmworker; acculturative stress; anxiety; depression; Mexican immigrants:Hispanic.

BACKGROUND society, and a sense of anxious disorientation in re-sponse to the unfamiliar environment. Immigrantsmay furthermore feel pulled between the influenceof traditional norms, values, and customs and thevalues, norms, and experiences in the new culture(e.g., parent-child conflict due to the child's encoun-tering of the new culture through school; role conflictdue to a working mother).

Immigrants may encounter many stressors dur-ing the process of acculturating to a new country(1-2). For example. immigrants may experience thebreaking of ties to family and friends in their countryof origin, thus resulting in feelings of loss and a reduc-tion in coping resources. Immigrants may also experi-ence factors that are specific to the new environment.These include discrimination, language inadequacy,the lack of social and financial resources, stress andfrustration associated with unemployment and/orlow income, feelings of not belonging in the host

Acculturative Stress Framework

The above types of experiences are encapsulatedby the term acculturatiue stress, which refers to thestress that directly results from and has its source inthe acculturative process (2). Hovey (3-4) and Hoveyand King (5-6) presented a conceptual frameworkfor studying acculturative stress among immigrants

'Department of Psychology, The University of Tol~do, Toledo,Ohio 43606. E-mail: jhovey@utoledo.edu.IDepartment of Psychology, The University of Toledo, ToledoOhio 43606.]AuthoT to whom corTespondence should be addressed.

119

1~~S~oo.o! 19118~ C 2(xx) Plcnum Publishinl Co""ralion

no Hovey and Magana

to immigrate. and low levels of education and income

significantly predicted high levels of depression andsuicidal ideation. Hovey's overall findings suggestthat those acculturating individuals experiencing ele-vated levels of acculturative stress are "at risk" forexperiencing critical levels of psychological distress,and that buffering variables such as those above mayhelp protect against distress during the accultura-tive process.

Characteristics of Migrant Fannworkers

There are approximately 1 million migrantfarm workers in the United States (12-13). Migrantfarmworkers are individuals who annually migratefrom one place to another to earn a living in agricul-ture. This is in contrast to seasonal farmworkers, wholive in one location during the year. Migrant farm-workers generally live in the southern half of theUnited States during the winter and migrate northbefore the planting or harvesting seasons. Three mi-grant streams have been identified (12.14). The WestCoast stream is primarily compo~ed of Mexican im-migrants who return to Mexico or the southwestUnited States after the ha:vest season. The EastCoast stream is primarily composed of Puerto Ricansand African-Americans who migrate from Florida.The Midwest stream is primarily composed of Mexi-can migrants who return to Mexico or Texas afterthe agricultural season.

Several authors (12-16) have noted the difficul.ties intrinsic to a migrant farmworker lifestyle. Forexample, migrant farm workers are socially marginal.This situation is intensified by the physical isolation,discrimination, and limited opportunities experi-enced by migrants. Most migrant farm workers earnless than $6,(00 per year, making them one of themost economically deprived groups in the UnitedStates. Farm labor is strenuous. Migrant workers areoften subjected to dangerous working conditions,such as being sprayed with pesticides, and thus, notsurprisingly, farm labor has the highest incidence ofworkplace fatalities in the United States. Child laboris cornmon, and thus the average migrant worker hasa sixth-grade education. Migrant workers typicallyfind housing in labor camps provided by their em-ployers. However, the housing and sanitation are of-ten substandard. For example, one-room homes thatlack water and toilet facilities are common, and drink-ing water and toilet facilities are often not readilyavailable in the fields. Finally, although their health

and its relationship to psychological functioning.These authors extended Berry's (2, 7-8) accultura-tive stress model to include possible consequences ofelevated levels of acculturative stress, rather thanfocusing on predictors of acculturative stress as haveother researchers (9-11). The revised framework hastwo components. First, it suggests that acculturatingindividuals experience varying levels of acculturativestress, and that high levels of acculturative stress mayresult in significant levels of anxiety and depression.In other words, the model suggests that individualswho experience high levels of acculturative stress maybe at risk for the development of anxiety and depres-sion. Second, the model identifies the cultural andpsychological factors that may account for high ver-sus low levels of anxiety and depression. These in-clude social support found within the new commu-nity; support from immediate and extended familysupport networks; socioeconomic status (SES); pre-migration variables, such as adaptive functioning(self-esteem, coping ability), knowledge of the newlanguage and culture, and control and choice in thedecision to immigrate (voluntary vs. involuntary);cognitive attributes, such as expectations for the fu-ture (hopeful vs. nonhopeful); religiosity; and thenature of the larger society-that is, the degree oftolerance for and acceptance of cultural dive~itywithin the new environment. These variables mayserve as predictors of anxiety and depression. Accul-turating individuals with positive expectations for thefuture and relatively high levels of social supportmay, for example, experience less depression thanindividuals without the same expectations and

support.Hovey used the above framework to guide past

research that explored the psychological functioningof immigrants. For example, Hovey and King (5)explored the relationship among acculturative stress,depressive symptoms, and suicidal ideation in a sam-ple of adolescent Mexican immigrants. They foundthat acculturative stress was positively associatedwith depression and suicidal ideation, and that accul-turative stress, perceived family dysfunction, andnonhopeful "expectations for the future" were sig-nificant predictors of depression and suicidal ide-ation. Hovey (3-4) found the same positive relation-ship among acculturative stress, depression, andsuicidal ideation in samples of adult Mexican andCentral American immigrants. These latter two stud-ies also found that family dysfunction, ineffective so-cial support, low levels of religiosity, nonhopeful ex-pectations for the future, lack of choice in the decision

Anxiety and Depression among Immigrant Farmworkers UI

conditions are among the worst in the United States(average life expectancy: 49 years), migrant farm-workers have very limited access to health care.

Given their difficult living conditions, migrantfarmworkers may be at psychological risk, and thussusceptible to problems such as anxiety and depres-sion. Very little research, however, has explored themental health of migrant farmworkers in theUnited States.

reveal whether the farmworkers are at greater riskfor anxiety or depression. for example.

Purpose of Present Study

The first purpose of the present study is to assessthe prevalence levels of anxiety and depression in asample of Mexican immigrant farmworkers in themidwest United States. Given the stressors associatedwith both immigration and migrant farmwork, it isexpected that the sample will reveal elevated levelsof anxiety and depression. The second purpose isto determine the relationships. among acculturativestress, anxiety, and depression. It is expected thatelevated levels of acculturative stress will be posi-tively associated with high levels of anxiety and de-pression. The third purpose is to determine the bestpredictors of anxiety and depression. The predictorvariables explored are acculturative stress, familyfunctioning, social support, self-esteem, religiosity,control and choice in the decision to immigrate, con-trol and choice in the decision to live as a migrantfarmworker, education, and income.

Previous Research of Mental Health amongMigrant Farmworkers

METHOD

Participants and Procedure

Vega et al. (17) examined psychological distressamong 501 Mexican-American farmworkers in cen-tral California. They measured psychological distresswith the Health Opinion Survey (HOS) (18), a mea-sure of general psychopathology. They found thathigh levels of psychological distress were related toreduced health statuses and an occurrence of envi-ronmental stressors over the previous year. In addi-tion. they found that middle-age individuals (40-59years) reported elevated levels of psychological dis-tress in comparison to other age groups. Vega et al.conjectured that middle age is an especially high-riskperiod for farmworkers because significant occupa-tional and life hazards exist to progressively degradefannworkers' health and functional capacities. Ac-cording to Vega et al.. the severe lifestyle (e.g., highfrequencies of environmental stressors, such as haz-ardous working conditions) experienced by Mexican-American farm workers places them at extraordinarypsychological risk.

Vega et al. (17) is the only study that has exam-ined predictors of mental health among migrantfannworkers in the United States. However, theirwork was limited. In their analyses, Vega et al. did notseparate migrant farmworkers from seasonal farm-workers. This distinction is important because a num-ber of authors (12-14) have suggested that, due totheir migratory and unstable lifestyle, migrant farm-workers are at greater risk for health problems thanseasonal farmworkers. Second, Vega et al. did notdirectly measure stressors that are specific to thefannworker lifestyle. Level of environmental stresswas based on one question. The participants wereasked whether they experienced a stressful life eventin the previous 12 months, such as the loss of a job,an accident. or the death of a family member orfriend. Finally. as noted, Vega et al. examined psycho-logical risk in a general fashion. Thus the data do not

Panicipants were 45 Mexican migrant farm-workers (20 females, 25 males) in the northwestOhio/southeast Michigan area. The age of the sampleranged from 17 to 65 (M = 33.53, SD = 11.03).Twenty-four percent (24.4%) of the sample were aged16-25 years; 33.3Cfo were 26-35; 26.7% were 36-45;13.3% \J,.ere 46-55; and 2.2% were 56-65. All of theparticipants were first-generation individuals. Thenumber of years living in the United States rangedfrom 1 to 35 years (M = 11.71, SD ; 8.87). Thirty-three percent (33.3%) of the sample had lived in theUnited States for 1-5 years; 22.2% of the sample hadlived in the United States for 6-10 years; and 44.5%of the sample had lived in the United States for morethan 10 years.

Sixty-two percent (62.2%) of the participantswere married; 24.5% were never married; 4.4% wereseparated or divorced; and 8.9% were in a common-law marriage or living together. Eighty-two percent(82.2%) of the participants were Catholic; 6.7Cfo were"Christian"; 8.9% reported "other" religious affilia-tions; and 22% reported no religious affiliation.

il2 Hovey and Magana

formation form asked three questions. These ques-tions were previously used (19) to assess religionamong Mexican immigrants. They were as follows:"How religious are you?" (Possible responses werethe following: 1 = not at all religious; 2 = slightlyreligious; 3 = somewhat religious; 4 = very religious.)"How much influence does religion have upon yourlife?" (Possible responses were the following: 1 =not at all influential; 2 =:= slightly influential; 3 =somewhat influential; 4 = very influential.) "Howoften do you attend church?" (Possible responseswere the following: 1 = never; 2 = once or twice ayear; 3 = once every 2 or 3 months; 4 = once amonth; 5 = two or three times a month; 6 = once aweek or more.)

The primary investigator established contactwith community agencies who have well-establishedties with migrant fannworker camps. These agencieshelped coordinate data collection by accompanyingthe present researchers to the camps and introducingthe researchers to the migrant farmworkers. The pri-mary investigator and four research assistants col-lected data from nine camps. The four research assis-tants underwent intensive training that providedinstruction on the administration of the instrumentsand focused on issues of cultural competence. Thetraining was conducted by the primary investigatorwho has extensive experience in community-basedresearch with Latin populations.

At each labor camp, the researchers recruitedone fannworker from each dwelling. In instances inwhich several unrelated families lived in the samehousehold, more than one participant was recruitedso that each family was represented. Following con-sent, each participant completed an open-ended in-terview. The purpose of these interviews was to cap-ture the phenomenology of the migrant fannworkerlifestyle. The interview data is reported in a separatepaper. After the interview, each participant com-pleted a questionnaire. Because of the low educa-tional levels among some migrant farm workers, theinterviewers offered to read and clarify, if necessary,the questionnaire items to each participant. Approxi-mately 33% of participants requested assistance. Theparticipants had the option of participating in eitherSpanish or English. Eighty-seven percent (86.7%) ofindividuals partic\pated in Spanish; 13.3% partici-pated in English. The interview and questionnairerequired approximately 1 hour to complete. Eachindividual was reimbursed $20.00 for her or his partic-ipation.

Control and Choice in the Decision to Immigrate tothe United States

To assess perception of control and choice inthe decision to immigrate to the United States, theparticipants were asked the following questions (4):"Did you contribute to the decision to move to theUnited States?" (Possible responses were the follow-ing: 1 = not at all; 2 = some [a little bit]; 3 = moderate[pretty much]; 4 = very much [a great deal].) "Didyou agree with the decision to move to the UnitedStates?" (Possible responses wert the following: 1 =strongly disagreed; 2 = disagreed; 3 = agreed; 4 =

strongly agreed.)

Control and Choice in the Decision to Live as aMigrant Farmworker

To assess perception of control and choice inthe decision to live as a migrant fannworker, theparticipants were asked whether they contributed tothe decision to live as a migrant fannworker (1 =not at all; 2 = some; 3 = moderate; 4 = very much)and whether they agreed with the decision to live asa migrant farmworker (1 = strongly disagreed; 2 =disagreed; 3 = agreed; 4 = strongly agreed).

;\teasures

A self-administered battery of questionnaireswas used. A background information form assessedage, gender, marital status, ethnicity, generationalstatus, religious affiliation, influence of religion,church attendance, education, family income, lan-guage use, control and choice in the decision to immi-grate to the United States, and control and choice inthe decision to live as a migrant farmworker. Family Assessment Device

Religion Variables The General Functioning subscale of the FamilyAssessment Device (FAD) (20) was used to measurefamily functioning. The FAD is a self-report scaleconsisting of statements that participants endorse in

To assess perception of religiosity, influence ofreligion, and church attendance, the background in-

Anxiety and Depression among Immigrant Farmworkers U3

terms of how well each statement describes their fam-ily. Items are scored on a 4-point Likert scale("strongly agree" to "strongly disagree"), with scaledscores for each dimension ranging from 1.00 (healthy)to 4.00 (unhealthy). The General Functioning sub-scale consists of 12 items. Examples of items includethe following: "In times of crisis we can turn to eachother for support" and "We avoid discussing ourfears and concerns." The FAD has been found (4,20-21) to have adequate internal consistency reliabil-ity (.71-.92), test-retest reliability (.66-.76), and con-struct validity among general and Mexican-Americansamples. The Cronbach alpha for the present studywas.72, thus indicating adequate internal consis-

tency reliability.

SAFE Scale

Acculturative stress was measured with theSAFE scale (9). This scale consists of 24 items thatmeasure acculturative stress in social. attitudinal. fa-milial, and environmental contexts. in addition toperceived discrimination toward acculturating popu-lations. Participants rate each item that applies tothem on a 5-point Likert scale ("not stressful" to"extremely stressful"), Examples of items include thefollowing: "People think I am unsociable when infact 1 have trouble communicating in English"; "Itbothers me that family members I am close to do notunderstand my new values"; and "Because of myethnic background, I feel that others exclude me fromparticipating in their activities," If an item does notapply to a participant, it is assigned a score of O.The present investigators slightly revised the scaleby adding two additional items: "I feel guilty because1 have left family or friends in my home country";and "I feel that I will never gain the respect that Ihad in my home country," The scale used in thisparticular study thus consisted of 26 items, with possi-ble scores ranging from 0 to 130, Higher 'scores indi-cate higher levels of acculturative stress. The SAFEscale has been found (4, 9-10) to have adequateinternal consistency reliability (.89-.90) and con-struct validity among Mexican-American samples.The Cronbach alpha for the present study 'Nas .88.

The Personal Resource Questionnaire

The Personal Resource Questionnaire-Part 2(PRQ85) (22) was used to measure social support.This scale measures the perceived effectiveness ofsocial support and consists of 25 items rated on a 7-point Likert scale ("strongly disagree" to "stronglyagree"). Possible scores range from 25 to 175. Higherscores indicate higher levels of perceived social sup-port. Examples of items include the following: "1belong to a group in which 1 feel important"; "1 havepeople to share social events and fun activities with";..1 can't count on my friends to help me with prob-lems"; and" Among my group of friends we do favorsfor each other." The PRQ85-Part 2 has been found(4,22-24) to have adequate internal consistency re-liability (.85-.93), test-retest reliability (.72), andconstruct validity among general and Mexican-Amer-ican samples. The Cronbach alpha for the presentstudy was .92.

Personality Assessment Inventory (PAl)

The Anxiety scale of the Personality AsscssmentInventory (PAl) (27) was used to measure anxiety,This scale measures clinical features of symptomatol-ogy related to anxiety disorders and consists of 24items rated on a 4-point scale ("false, not at all true"to "very true"). Higher scores indicate higher anxietylevels. Examples of items include the f01lowing: "Iam so tense in certain situations that I have greatdifficulty getting by"; "When I'm under a lot of pres-sure, I sometimes have trouble breathing"; "I oftenhave trouble concentrating because I'm nervous";and "I usually worry about things more than Ishould," The accepted caseness threshold is 60, Ascore of 60 or more represents potentially significantanxiety, which may impair functioning. It is estimated(27) that 16% of general population individuals willreach caseness. The PAl Anxiety scale has beenfound (27-29) to have adequate internal consistencyreliability (.80-.90), test-retest reliability (.85-.88),

Adult Self-Perception Scale

Self-esteem was measured with the Global Self-Worth sub scale of the Adult Self-Perception Scale(25). The subscale consists of 6 items, each of whichis scored 1 to 4, with possible scores ranging from 6to 24. Higher scores indicate higher levels of self-esteem. The Global Self-Worth subscale has beenfound (25-26) to have adequate internal consistencyreliability. test-retest reliability, and construct valid-ity among general and Mexican-American samples.

U4 Hovey and Magana

regression analyses are presented. They were con-ducted to determine the best predictors pf anxietyand depression. For each regression analysis, the cri-teria for entering the equation was set at F = 3.84.

and construct validity among general and Mexican-American samples. The Cronbach alpha for the pres-ent study was .91.

Center for Epidemiologic Studies Depression ScaleRESULTS

The Center for Epidemiologic Studies Depres-sion Scale (CES-D) (30) was used to measure depres-sion. The CES-D assesses level of depressive symp-toms within the previous week and consists of 20items rated on a 4-point scale ("rarely or none of thetime" to "most or all of the time"). Possible scoresrange from 0 to 60. Higher scores indicate higherdepression. The accepted caseness is a score of 16 ormore, which represents the upper 18% of scores (31).A score of caseness indicates the presence of poten-tially significant depressive symptomatology. Severalstudies (4, 32-33) have found that the CES-D hasadequate internal consistency reliability (.81-.90)and construct validity among Mexican-Americansamples. The Cronbach alpha for the present studywas .80.

Descriptive Statistics

Education and Income

Table I shows the frequency distributions foreducation and income. Most individuals reported rel-atively low levels of education and extremely lowlevels of income. The median level of education was6-8 years of schooling. Thirteen percent (13.3%) ofthe sample reported high education levels, which isrepresented by high school graduation and greater.

Church Attendance, Perception of Religiosity, andInfluence of Religion

Table I shows the frequency distribution forchurch attendance. About two-thirds of individualsattended church at least 2 or 3 times per month. Themean score for perception of religiosity was 2.41

Translation

The Spanish version of the PAl (34) that wasused in the present study was translated by Psycho-logical Assessment Resources, Inc. The backgroundinformation form, the FAD. the PRQ-85, the AdultSelf-Perception Scale, the SAFE, and the CES-Dwere translated into Spanish through the double-translation procedure (35) with the help of two trans-lators.

Table t. Sample Distributions for Sociodemographic Variables-

EntireVariable Females Males sample

20.020.010.040.010.00.0

20.020.032.012.012.04.0

20.020.022.224.S11.12.2

Data Analyses

The data analyses are presented in three steps.Descriptive statistics are presented first. Second, cor-relation coefficients that were used to assess the rela-tionships among the predictor variables (i.e., accul-curative stress, family functioning, social support,self-esteem, church attendance, perception of religi-osity, influence of religion, contribution to the deci-sion to immigrate, agreement with the decision toimmigrate, contribution to the decision to live as amigrant farmworker, agreement with the decision tolive as a migrant farm worker, education, and income)and dependent variables (anxiety and depression)are presented. Finally, two forward stepwise multiple

15.842.136.85.3

33.341.720.84.2

25.641.827.94.7

0.00.00.0

20.015.065.0

2.34.56.8

18.222.745.5

4.28.3

12.516.729.229.2

Educalion0-2 years or school3-5 years or school

6-8 years or school9-11 years or school

High school graduate

Some college

Income

SO-S4,999S5,<XXJ-S14,999S15,<XXJ-S24,999Over S25,<XXJ

Church attendance

NeverOnce or twice a yearOnce every 2 or 3 monthsOnce a month

2 or 3 limes a monthOnce a week or more

'There were no significant gender differences for education. in-come. and church attendance.

Anxiety and Depression among Immigrant Famlworkers 125

(SD = 0.76). This mean represents a moderate level

of perceived religiosity. The mean score for influenceof religion was 2.93 (SD = 1.00). This mean repre-sents a relatively high level of influence of religion.

Contribution to and Agreement with the Decision toImmigrate to the United States

The mean score for contribution to the decisionto immigrate was 3.05 (SD = 1.12). The mean scorefor agreement with the decision to immigrate was3.21 (SD = 0.97). These means represent a relativelyhigh level of contribution and agreement.

(anxiety). and the CES-D (depression). The presentsample revealed a relatively high level of anxiety(M = 55.0) in comparison to the expected mean of50 (27) in general population individuals (r [44J =2.4, P < .01). Twenty-nine percent (28.9%) of theparticipants reached caseness on the PAl with a scoreof 60 or greater, compared to the expected 16% (27).The present sample revealed a relatively high levelof depression. Thirty-eight percent (37.8%) of theparticipants reached caseness with a score of 16 orgreater on the CES-D, compared to the expected18% (31).

To note, ANDY As revealed no significant maineffects for gender, generation level, age (16-25 years,26-35.36-45,46-55.56-65), and language of partici-pation on acculturative stress, anxiety, and de.

pression.

Contribution to and Agreement with the Decision toLive as a Migrant Farmworker

The mean score for contribution to live as amigrant farmworker was 3.20 (SD = 1.14). The meanscore for agreement to live as a migrant fannworkerwas 3.07 (SD = 0.97). These means represent a rela-tively high level of contribution and agreement inthe decision to live as a migrant fannworker.

Correlations among Predictor Variablesand Anxiety

Table III shows the correlations among the pre-dictor variables and anxiety. Greater education, lowlevels of perception of religiosity, low levels of influ-ence of religion. low contribution to the decision toimmigrate, low contribution to the decision to liveas a migrant farmworker, low self-esteem, ineffective

Family Functioning and Social Support

The mean score for the General Functioningsubscale of the FAD (family functioning) was 2.07(SD = 0.43). The mean score for the PRQ85 (socialsupport) was 126.09 (SD = 33.24). These two means

represent overall moderate levels of support. Table Ill. Correlations among Prcdictor Variables and Anxietyand Depression'

Anxiety DepressionSelf-EsTeem

.25**

.07

-.24**-.29*.

-.06

-.24**

.23..-.01

-.06

-.01

-.25..-.19-

The mean score for self-esteem was 18.51 (SD =3.24). This represents a moderate level of self-esteem.

Accu/turative Stress, Anxiety, and Depression

-.01 -.23--Table II lists the means and standard deviationsfor the SAFE scale (acculturative stress). the PAl

-.2988 -.13

Table II. Mean Scores and Standard Deviations on Measures ofAcculturative Stress, Anxiety, and Depression

Acculturativestress Anxiety Depression

Mean (SD) Mean (SD) Mean (SD)

-.14 -.41---

EducationIncomePerception of religiosityPerceived influence of religionChurch attendanceContribute to decision to im-

migrateAgreement with decision to

immigrateContribute to migrant farm-

workAgreement with migrant

farm workSelf-esteemSocial support

Family functioningAcculturative stress

Anxiety

-.34----.25---.03

.64

-.53 -.52 .1S.

.57 .57

55.0 (14.0)56.5 (13.9)53.8 (14.3)

14.5 (10.2)

13.8 (9.4)

15.0 (10.9)

Overall

FemalesMales

57.8 (21.4)55.3 (17.6)59.9 (24.2) 'Significance levels are based on one-tailed tests.

.p < .10; ..p < .05; ...p < .01; p < .001.

U6 Hovey and Magana

social support. and high levels of acculturative stresswere related to high levels of anxiety.

decision to live as a migrant farmworker. self-esteem.social support. and acculturative stress were enteredas predictors of anxiety. Significant independent pre-dictors of anxiety were acculturative stress (13 = .59.1 = 4.9. P < .(xx)1). contribution to the decision tolive as a migrant farmworker (13 = -.37. t = -3.2,p < .003), and influence of religion (13 = -.20.1 =-1.7. P < .10). As seen in Table IV. these variablesaccounted for 50% of the variance in anxiety. Theother variables added minimal variance to the equa-tion. The overall equation accounted for 53% of thevariance in anxiety.

Correlations among Predictor Variablesand Depression

Table III lists the correlations among the pre-dictor variables and depression. Greater education,infrequent church attendance, low contribution tothe decision to immigrate, low agreement with thedecision to immigrate, low agreement with the deci-sion to live as a migrant farmworker, low self-esteem,ineffective social support, family dysfunction, highlevels of acculturative stress, and high levels of anxi-ety were related to elevated levels of depression.

Multiple Regression Analysis of Depression

Table V shows a stepwise multiple regressionanalysis that was conducted to determine the bestpredictors of depression. Education, church atten-dance, contribution to the decision to immigrate,agreement with the decision to immigrate, agreementwith the decision to live as a migrant farmworker,self-esteem, social support, family functioning, accul-turative stress. and anxiety were entered as predictorsof depression. Significant independent predictors of

Multiple Regression Analysis or Anxiety

Table IV shows a stepwise multiple regressionanalysis that was conducted to determine the bestpredictors of anxiety. In this analysis, education, per-ception of religiosity, influence of religion, contribu-tion to the decision to immigrate, contribution to the

22.0617.52

13.13

(1,41)(2,40)(3,39)

.000

.000

.000

35.046.750.2

(4,38)10.10 .(XX) 51.5

8.05 (5.37)

.00)

52.1

.COO 52.76.68 (6.36)

5.65 (7.35) .00) 53.1

.(XX)

Anxiety symptoms (PAl)"SAFESAFE. Contribute to migra-,t CarmworkSAFE. Contribute to migrant farrnwork,

influenceSAFE, Contribute to migrant Carrnwork.

influence, contribute to moveSAFE. Contribute to migrant farrnwork,

influence. contribute to move, religiositySAFE. Contribute to migrant farrnwork,

influence, contribute to move. religiosity,PRO

SAFE, Contribute to migrant farrnwork,influence. contribute to move, religiosity,PRO, self.esteem

SAFE, Contribute to migrant farrnwork,influence. contribute to move, religiosity,PRO, self-esteem, education

4.83 (8.34) 53.2

'PAI = Anxiety subscaJe oc Personality A5sessment Inventory; PRQ= Persona! Resource Questionnaire(social support); SAFE = Socia!. Attitudinal. Familial, and Environmental acculturative stress scale;Inftuence = inftuence of religioa; Contribute to move = contribution to the decision to immigrateto the United States; Contribute to migrant Cannwork = contribution to the decision to live as a

migrant Cannworker.

Anxiety and Depression among Immigrant Fannworkers 127

Table V. Multiple Regression of Depression among Migrant Farmworkers

Total

percentageDependent and variance

predictor variables F Cd/) p accounted for

20.3418.7114.9113.39

(1.40)(239)(3.38)(4.37)

.000

.000

.000

.000

33.749.054.159.161.8

11.64 (5.36) .000

63.910.31 (635) .(XX)

65.09.04 (7.34) .000

65.6

7.85 (8.33) .(XX)

Depressive symptoms (CES-D)"PAlPAl, PRO.PAl, PRO, Self-esteemPAl, PRO, Self-esteem, attendancePAl, PRO, Self-esteem, attendance, agree

with migrant fannworkPAl, PRO, Self-esteem, attendance, agree

with migrant fannwork, contribute to movePAl, PRO, Self-esteem, attendance, agree with

migrant fannwork, contribute to move, SAFEPAl, PRO, Self-esteem, attendance, agree with

migrant fannwork, contribute to move, SAFE,agree with move

PAl, PRO, Self-esteem, attendance, agreewith migrant fannwork, contribute to move,SAFE, agree with move, education

PAl, PRO, Self-esteem, attendance, agreewith migrant fannwork, contribute to move,SAFE, agree with move, education, FAD

65.8

6.83 (9.32) .(XX)

65.8

6.00 (10.31) .(XX)

'CES.D = Center for Epidemiologic Studies Depression Scale; PAl = Anxiety subscale of Personality

Assessment Inventory; PRQ := Personal Resource Questionnaire (social support); FAD -FamilyAssessment Device; SAFE = Social, Attitudinal, Familial, and Environmental acculturative stress scale;Attendance = church attendance; Contribute to move = contribution to the decision to immigrate tothe United States; Agree with move = agreement with the decision to immigrate to the United States;Agree with migrant farmwork = agreement with the decision to live as a migrant farmworker.

depression were anxiety ({3 = .58, t = 4.5, p < .0001),social support ({3 = -.40, t = -3.4, P < .002), self-esteem ({3 = -.28, t = -2.3, p < .03), church atten-dance ({3 = -.24, t = -2.4, p < .03), and agreementwith the decision to live as a migrant fannworker({3 = -.21, t = -1.8, p < .10). These five variables

accounted for 62% of the variance in depression. Theremainder of the variables added little variance. Theoverall equation accounted for 66% of the variancein depression.

this context. These findings contribute critical infor-mation both to the acculturative stress literature andto the cross-cultural literature on depression andanxiety.

Because qualitative data portray a sense of indi-vidual experience that is often lacking in quantitativedata. the following discussion is highlighted with ex-amples of narrative responses from the interviews.

Acculturative Stress in Relation to Anxietyand Depression

DISCUSSIONIn the present study, migrant farm workers expe-

riencing elevated levels of acculturative stress alsoreported high levels of anxiety and depression. Theseacculturating fannworkers may feel caught betweencultures. That is, these individuals may feel pulledbetween the influence of traditional customs, values,and norms and the values, nonns, and customs foundin the mainstream society. Experiences of economichardship, language difficulties, and discrimination

The major theme of this study is that the immi-gration experience in conjunction with the migrantfarmworker lifestyle may put an individual at psycho-logical risk. As mentioned, no previous studies haveaddressed the psychological functioning of immigrantfarmworkers in the midwest United States. There-fore, a purpose of the present study was to examineacculturative stress, anxiety, and depression within

us Hovey and Magana

may further contribute to distress during the accultur-ative process. Many farrnworkers reported experi-encing discrimination and exploitation. For example,the following narrative, reported by a 26-year-oldmale, captures such experiences.

fore, the present findings have relevance for clinicalwork and research among immigrant farmworkers.

Predictors of Anxiety and Depression

There are lots of thieves. On one occasion I caughtone guy who was trying to steal from our home. Itook him to the police. The police said we couldpress charges. so we did. but the thief was releasedon the third day because he was a U.S. citizen. Thebottom line is that they let him go free becausewe are not U.S. citizens. so the police did not payattention to our charges.

Social Support and Self-Esteem

They are supposed to pay us weekly or every other

week, but they take longer to pay us. They make

excuses such as they don't have the checks, or theymay say to come back another day. We need the

money right away but they still don't pay us. Usuallythe contractors take advantage of the people whoare new and who know nothing about being a farm-

worker.

The present study measured the perceived effec-tiveness of social support rather than access to socialsupport networks. Several authors (39-41) havenoted that larger social networks do not ensure thatthe support will be of higher quality or more effective,and therefore the perceived quality of social supportmay be a more accurate predictor of psychologicaldistress than is quantity of social support. The presentfindings indicated that ineffective social support wasstrongly related to heightened levels of anxiety anddepression. These findings thus lend support to theidea that social support of high quality may helpimmigrant farm workers cope against anxiety and de-pression.

Mexican culture traditionally emphasizes collec-tivist values and affiliation (42). Mexican immigrantsmay thus feel particularly vulnerable when they lacksocial support. Because social support helps provideindividuals with a sense of belonging and identity,ineffective social support may lead acculturating indi-viduals to feel undervalued and contribute to lowself-esteem (43). Moreover, given that self-esteemmay help buffer against distress during the accultura-tive process (44), low self-esteem may place an indi-vidual at greater risk for distress. Not surprisingly, thepresent findings indicated a very strong relationshipbetween low self-esteem and elevated levels of anxi-ety and depression.

Levels of Anxiety and Depression amongMigrant Farmworkers

Relatively high levels of anxiety and depressionwere found in the present sample. As noted, 38% ofthe sample reached depression caseness with a CES-D score of 16 or greater. This percentage appears tobe high. As a comparison, about 18% of individualsfrom general population samples reach the casenessthreshold (30, 36). As a further comparison, Vega etat. (31) noted the very high prevalence of depressivesymptoms found within their sample of Mexican im-migrants. They found that 42% of their sample scored16 or greater. It is important to note that the highoverall rate of anxiety and depression found in thepresent sample does not imply that all immigrantfarmworkers, per se, are highly anxious and/or de-pressed, but that the experiences that go into beingan immigrant farmworker (e.g., discrimination, lan-guage inadequacy, reduced self.esteem, financialstressors, lack of family and social support) poten-tially influence psychological status.

The present study measured depression as a con-stellation of symptoms and did not obtain specificclinical information about the onset, duration, andseverity of the symptoms. Although the CES-D isnot a diagnostic instrument, it was found (37) to havea concordance of 85% for current major depressionusing the Diagnostic Interview Schedule (38). There.

Control and Choice in the Decisions to Immigrate

and to Live as a Migrant Farmworker

Salgado de Snyder (45) and Vega er ai. (46),in their respective studies of depression risk factorsamong Mexican immigrants, found that those individ-uals who voluntarily immigrated ("wanted to") to theUnited States revealed significantly less depressivesymptoms than those individuals who involuntarilyimmigrated ("had to"). These findings suggest thatindividuals who are willing to immigrate may be atless risk for depression than those who are not willing.In other words, greater depression among those whodo not choose to immigrate may be due to the effects

Anxiety and Depression among Immigrant Farmworkers 129

of the lack of empowerment to control their liveswhen migration occurs. This notion has relevance forthe present study. The present study assessed bothinternational migration and the participants' migra-tion as farm workers. The farmworkers were askedwhether they contributed to and agreed with movingto the United States, and whether they contributedto and agreed with the decision to live as a migrantfarmworker, or whether they were involved in fann-work due to the desire of others. Not surprisingly,it was generally found that those farmworkers whowillingly migrated reported less anxiety and depres-sion than those who did not.

Education

Berry et al. (47) noted that education may helpprovide acculturating individuals with the resourcesto cope with the larger society. They believed thatthose individuals with more education may havegreater cognitive, economic, and social resources withwhich to deal with changes. The direction of educa-tion as a predictor in the present study was thereforesurprising. High education was associated with de-pression. This finding may depend partly on the ques-tion of comparison. Some fannworkers may comparetheir current situation to a lower socioeconomic expe-rience in Mexico. However, farmworkers who aremore educated may be more sensitive to the discrep-ancy between their current life conditions and thoseof other individuals in the United States. Those whoare more educated may also have set life and careergoals other than migrant farmwork, and may havefelt that they have failed to reach these goals. Thefollowing narrative is from a relatively educated (highschool graduate) 39-year-old male who seems awareof the disparity between his current socioeconomicsituation and those of others in the United States:

We receive such miserable pay as migrants. I believethat migrants are a resource. We are a very importantpart of the growth and feeding of this country. andI believe we have a right to be recognized for ourhard work. either by the government or the labordepartment. They pay no more than minimum wageand that is too little to get by. Everything is so expen-sive. We should have better pay. It is our right.

Suggestions for Prevention and Treatment

immigrant farmworkers who experience psychologi-cal problems. This situation may also exist in otherareas of the United States. The present findings, how-ever, suggest the need for prevention. assessment,and treatment services for immigrant farmworkers.

It is crucial that prevention efforts be directedtoward those farmworkers who are at risk for anxietyand depression. These include farmworkers who areisolated, lack emotional support and self-esteem, andwho experience elevated levels of acculturativestress. Possible preventive strategies include the es-tablishment of support groups, at the camps or localcommunity centers, \1I11ere migrant workers can talkabout their difficult experiences and the ways inwhich they can cope with these difficulties. Supportgroups would provide emotional support and in-crease self-esteem. Several participants in the presentstudy expressed interest in the establishment of sup-port groups.

Second, educational workshops and presenta-tions (48-49) can be conducted by health profession-als. Because for utmost prevention it is importantthat these educational workshops and presentationsbe accessible to migrant workers, they should be es-tablished at easily accessible locations such as migrantcamps, community centers, or local schools. Theselectures and workshops can address specific topicssuch as risk factors for anxiety and depression, sub-stance abuse, and how to cope with the stress of amigratory lifestyle. These educational programswould be preventive in that active participationwould help thwart future problems in these areas.

The church can be another possible site for pre-vention. Several characteristics of the church may bepreventive (19). Religious organizations foster socialnetworking and thereby reduce risk through socialsupports. Moreover, church attendance providesgreater exposure to basic religious beliefs thought toincrease coping. As expected, the present findingsindicated a negative relationship betWeen religiosityand distress. Church members may also use theirministers or priests as sources of support in times ofdistress. In addition to such supportive roles, clergymembers may disseminate information to farmwork-ers regarding the availability of other community ser-vices. Because the cultural importance of the churchextends beyond the scheduled religious services, out-reach programs that are sponsored by the church-but not necessarily held at the church-are likely tohave the respect of migrant workers.

Finally, preventive efforts can be incorporatedinto lay health-worker programs (50-51). Lay health-

Currently, in the area sampled, there is little tono prevention and treatment options available for

130 Hovey and Mag:ciia

aspects, religious practices, and spiritual dimensionsof religion.

Further research should concentrate on increas-ing the study's generalizability. This includes researchof a representative nature that examines the specificpathologies found among migrant farm workers, ex-plores the mental health differences between migrantand seasonal farmworkers, and examines the psycho-logical functioning of migrant farm workers in othermigrant streams. Qualitative research is needed toidentify those stressors specific to the migrant farm-worker lifestyle and the coping mechanisms that areemployed in response to these stressors. This infor-mation will be useful in establishing preventive ser-..;ces for migrant farmworkers.

worker programs use individuals who are former orcurrent migrant farmworkers who are trained to pro-vide health information to migrant farmworkers. Thelay health-workers organize and run educational andpreventive workshops (sample topic: HIV / AIDS pre-vention) and act as liaisons between communityagencies/health services and migrant farmworkers.Lay health-worker programs have been shown to beeffective preventive resources (51) and to be veryempowering for migrant farmworkers (50). In addi-tion to being educational, these programs may helpprovide social contacts and increase self-esteemamong migrant workers.

For the farm worker who may be experiencingacculturative stress, anxiety, and/or depression, thefindings highlight the importance of assessment andtreatment within a cultural context. In other words,clinical evaluation and treatment should carefully ad-dress the stress related to farmwork; the stress relatedto acculturation; family and social support; the farm-worker's sense of self; the farmworker's hopes andexpectations for the future; and past and present cop-ing strategies, including religion. Treatment for mi-grant workers should be short term in focus becauseof the migratory nature of their lifestyle. Moreover.the clinician should be aware of mental health ser-vices that are available in the farmworker's otherareas of residence.

ACKNOWLEDGMENTS

This work was supported in part by a SummerResearch Fellowship and Research Grant, UniversityResearch Awards and Fellowship Program, Univer-sity of Toledo. The authors thank Migrant HealthPromotion, Inc., and Rural Opportunities, Inc., fortheir help in arranging data collection.

REFERENCES

Limitations and Directions for Future Research

This study should be considered preliminary be-cause of its relatively limited sample size, its self-report method, and its cross-sectional design. In ad-dition. the homogeneity of the sample in terms ofethnicity and area sampled suggests that these find-ings should not be generalized to the West Coast andEast Coast migrant farmworker streams. Similarly,the findings should not be generalized to migrantfarrnworkers of other ethnicities in the Midweststream. Although the instruments used were shownto be reliable in the present study and have pre-viously been validated on Mexican immigrants, thesescales have yet to be fully validated on Mexicanmigrant farmworkers. The present study assessedthe influence of religiosity on anxiety and depression.It is unclear how the findings would have differedif spirituality was also measured. Future researchshould thus use a more comprehensive measure ofreligion that is able to distinguish between the social

1. Hovey JD: Psychosocial prediclo~ of acculluralive stress inCentral American immigrants. J Immigrant Health 1999;1:187-194

:. Williams CL. Berry JW: Primary prevention of acculturativestress among refugees: Application of psychological theoryand practice. Am Psycho I 1991; 46:632-641

3. Hovey JD: Acculturative stress. depression. and suicidal ide-ation among Central American immigrants. Suicide LifeThreat Behav 2roo; 30:125-140

4. Hovey JD: Acculturative stress, depression, and suicidal ide-ation in Mexican immigrants. Cultural Diversity and Ethnic~iinorily Psychology 200>; 6:134-151

5. Hovey JD, King CA: Acculturative stress. depression. andsuicidal ideation among immigrant and second generation Lat-ino adolescents. J Am Acad Child Adolesc Psychiatry 1996:35:1183-1192

6. Hovey JD. King CA: Suicidality among acculturating Mexican-Americans: Current knowledge and directions for research.Suicide Life Threat Behav 1997; 27:92-103

7. Berry JW: Psychology of acculturation. In: Berman JJ, ed.:..'ebraska Symposium on Motivation: Vol. 37. Cross-CulturalPe~pectives. Lincoln: Unive~ity of Nebraska Press; 1990:201-234

8. Berry JW, Kim U: Acculturation and mental health. In: DasenP. Berry JW, Sartorius N. eds. Health and Cross-Cultural Psy-~bology: Towards Application. London: Sage; 1988: 207-236

9. Mena FJ. Padilla AM. Maldonado M: Accullurative stress andspe~jfic coping strategies among immigrant and later genera-tion college students. Hispani~ J Behav ~i 1987; 9:207-225

10. Padilla AM. Alvarez M. Lindholm KJ: Generalional status

Anxiety and Depression among Immigrant Fannworkers 131

Americans and non-Hispanic whites. Psychol Assessment1989; 1:163-168

33. Golding 1M. Aneshensel CS. Hough RL Responses to depres-sion scale items among Mexican-Americans and non-Hispanicwhites. 1 Clin Psycholl991; 47:61-75

34. Morey LC: Personality Assessment Inventory: Spanish Trans-lation. Odessa: Psychological Assessment Resources; 1992

35. Brislin RW: Translation and content analysis of oral and writ-ten material. In: Triandis HC. Berry JW. eds. Handbook ofCross-Cultural Psychology: Vol. 2. Methodology. New York:Wiley: 1980: 389-444

36. Weissman M. Meyers J: Rates and risks of depressive symp-toms in a United States urban community. Acta PsychiatrScand 1978: 57:219-231

37. Hough. R: Comparison of psychiatric screening questionnairesfor primary care patients. Final Report prepared for AD-AMHA. Division of Biometry and Epidemiology. NationalInstitute of Mental Health (Cantract No. 278-81-0036): 1983

38. Robins LN. Helzer JE. Croughan J. Ratcliff K: Nationallnsti-tute of Mental Health Diagnostic Interview Schedule: Its his-tory. characteristics. and validity. Arch Gen Psychiatry1981; 38:381-389

39. Golding 1M. Burnam MA: Stress and social support as pre-dictors of depressive symptoms in Mexican Americans andnon-Hispanic whites. J Sac Clin Psychol 1m-. 9:268-287

40. Hovey JD: The moderating influence or socia! support onsuicidal ideation in a sample of Mexican immigrants. PsycholRep 1999: 85:78-79

41. Sarason IG. Levine HM. Basham RB. Sarason BR: Assessingsocial support: The Social Support Questionnaire. J Pers SacPsychoI1983; 44:127-139

42. Alvarez RR: Familia. Berkeley: University of CaliforniaPress; 1987

43. Smart IF. Smart DW: Acculturative stress of Hispanics: Lossand challenge. J Counsel Devell995; 73:390-396

44. Espin OM: Psychological impact of migration on Latinas. Psy-chol Women Q 1987; 11:489-503

45. Salgado de Snyder VN: Factors associated with acculturativestress and depressive symptomatology among married Mexi-can immigrant women. Psychol Women Q 1987; 11:475-488

46. Vega WA, Kolody B. Valle lR: Migration and mental health:An empirical test of depression risk factors among immigrantMexican women. Inl Migration Rev 1987; 21:512-530

47. Berry lW. Kim U. Minde T, Mok D: Comparative studies ofacculturative stress. Int Migration Rev 1987; 21:491-511

48. Davis 1M, Sandoval J, Wilson MP: Strategies for the primaryprevention of adolescent suicide. School Psychol Rev 1988;17:559-569

49. Sandoval J: Crisis counseling: Conceptualizations and generalprinciples. School Psychol Review 1985; 14:257-265

50. Booker VK, Robinson JG, Kay BJ. Gutierrez-Najera L. Stew-art G: Changes in empowerment: Effects of panicipation ina lay health promotion program. Health Educ Behav 1997;24:452-464

51. Harlan C. Eng E. Watkins E: Migrant lay health advisors: Astrategy for health promotion. In: McDuffie HH. Dosman JA.Semchuk KM. Olenchok SA, eds. Agricultural Heallh andSafety: Workplace, Environment, Suslainability. New York:CRC Press; 1995:499-502

and personality factors as predictors of stress in students. His-panic J Behav Sci 1986; 8:275-288

11. Dona G. Berry JW: Acculturation attitudes and acculturativestress of Central American refugees. International Journal ofPsychology 1994; 29:57-70

12. Barger WK. Reza EM: The Fann Labor Movement in theMidwest. Austin: University of Texas Press; 1994

13. Rothenberg D: With These Hands: The Hidden World ofMigrant Farmworkers Today. New York: Harcourt Brace &Company; 1998

14. Goldfarb RL: Migrant Farm Workers: A Caste of Despair.Ames: Iowa State University Press; 1981

15. Farm Labor Organizing Committee: Fannworkers and FannLabor Conditions [online]. Available: http://www.iupui.edu/-fioclfws.htm; 1998

16. Valdes DN: AI Norte. Austin: University of Texas Press; 199117. Vega W. Warheit G, Palacio R: Psychiatric symptomatology

among Mexican American fannworkers. Sac Sci Med 1985;20:39-45

18. MacMillan A: The Health Opinion Survey: Technique forestimating prevalence of psychoneurotic and related types oCdisorders in communities. Psychol Rep 1957; 3:325-329

19. Hovey JD: Religion and suicidal ideation in a sample of LatinAmerican immigrants. Psycho I Rep 1999; 85:171-177

20. Epstein NB. Baldwin LM. Bishop DS: The McMaster FamilyAssessment Device. J Marital Fam Ther 1983; 9:171-180

21. Halvorsen JG: Self-report family assessment instruments: Anevaluative review. Family Practice Res J 1991; 11:21-55

22. Weinert C: A social support measure: PRO85. Nun Res1987; 36:273-277

23. Weinert C. Brandt PA: Measuring social support with thePersonal Resource Questionnaire. Western J Nurs Res1987: 9:589-602

24. Weinert C. TildenVP: Measures of social support: Assessmentof validity. Nurs Res 1990; 39:212-216

25. ~1esser B. Harter S: Manual: Adult Self-Perception Scale. Den-ver: University of Denver Press: 1986

26. Knight GP. Virdin LM. Ocampo KA. Roosa M: An examina-tion of the cross-ethnic equivalence of measures of negativelife events and mental health among Hispanic and Anglo.American Children. Am J Community Psychol 1994;22:767-783

27. Morey LC: Personality Assessment Inventory: ProCessionalManual. Odessa: Psychological Assessment Resources; 1991

28. Fantoni-Salvador P. Rogers R: Spanish versions of the MMPI.2 and PAl: An investigation of concurrent validity with His.panic patients. Assessment 1997; 4:29-39

29. Rogers R. Flores J. Ustad K, Sewell KW: Initial validation ofthe Personality Assessment Inventory-Spanish version withclients from Mexican American communities. J Pers Assess-ment 1995; 64:340-348

30. Radloff LS: The CES-D scale: A self-report depression scalefor research in the general population. Appl Psycho I Meas1977; 1 :385-401

31. Vega W A, Kolody B. Valle R, Hough R: Depressive symptomsand their correlates among immigrant Mexican women in theUnited States. Sac Sci Med 1986; 22;645-652

32. Golding JM, Aneshensel CS: Factor structure of the Centerfor Epidemiologic Studies Depression scale among Mexican