A cross-cultural study of depressive symptomology

WEINING C. CHANG

A C R O S S - C U L T U R A L S T U D Y O F

D E P R E S S I V E S Y M P T O M O L O G Y

ABSTRACT. Black (N = 26), white (N = 26) and Overseas Chinese (N = 32) college students were surveyed with Zung's Self-rating Depression Scale (SDS). With the exception of two items where marginally significant differences were found, there were no quantitative differences in the results across these ethinc groups. Inter-item analyses, factor analyses and comparison of factor profiles indicated that the items clustered in different configurations for the three groups. The factors contributing the most variances differed in their item contents across the ethnic groups: a mixture of affective and somatic complaints characterizes the black group; existential and cognitive concerns characterize the white group and somatic complaints characterize the Overseas Chinese group.

Discussion was focused on issues concerning cultural conception of psychological difficulties, communication styles used to communicate these difficulties and situational variables affecting performances on psychological diagnostic tests.

INTRODUCTION

In the United States, depression is considered a serious mental health problem.

Cross-cultural studies of mental health, however, have revealed a very different

situation about depression in people of non-Western cultures (e.g., Al-Issa 1970;

Benedict and Jacks 1954; Fabrega 1974; Marsella 1980; Pfeiffer 1968). Many

of these reports observed that the cluster o f symptoms characterizing depression

in the West seemed to be rare or absent in non-Western cultures. As a matter of

fact, some o f the reports have indicated that the incidences of depression in

some of these societies were related to the degree of "Westernization" of the

culture groups in question. These observations have prompted some researchers

to conclude that "depression" is perhaps a "Western" problem (see Marsella

1980 for a review of cross-cultural studies of depression).

Aside from the absence of emotional uproar (there is certainly no pride to

be taken in being more or less depressed), the situation is reminiscent of the

controversy over intelligence testing in different cultural groups. For a while it

was considered that many of the non-Western populations, including some o f

the ethnic minorit ies in the United States, were deficient in mental capacities

because o f the lower mean test scores these groups typical ly obtained in standard

intelligence tests. Only recently have researchers looked into the validity of these

standard measurements and at the situational variables involved in psychological

testing to provide alternative interpretat ions of the performance differences in

the standard intelligence tests (e.g., Cole and Brunet 1971 ; Cole and Scribner

1974). These scholars questioned the adequacy of using the same psychological

tests on subjects of different cultures and contended that before any quantitative

Culture, Medicine and Psychiatry 9 (1985), 295-317. 0165-005X/85.10 © 1985 by D. Reidel Publishing Company.

296 WEINING C. CHANG

comparison can be undertaken, it is mosf important to look into the possible

cultural differences in the structure and content o f the psychological trait in question (Cole and Scribner 1974). Though studies on culture and human cogni-

tion are still inconclusive, important questions have been generated by these

studies about psychological testing on people of different cultures (Irvine and Carroll 1980).

Some of the basic problems in testing subjects of different cultures were summarized by Florence Goodenough (1936) almost forty years ago:

Examination of the literature in the past twenty years shows that approximately two thirds of the publications dealing with racial differences in mental traits have been conducted with measurements of intelligence by means of tests designed for use with American and European whites . . . . Now the fact cannot be too strongly emphasized that neither intelligence tests or the so-called tests of personality and character are measuring devices properly speaking, . . . They are sampling devices . . . . The tests items from which the total trait is to be judged are representative and valid samples of the ability (or trait-author) in question as it is displayed within the particular culture with which we are concerned (pp. 5-6).

In psychology, emotions are regarded as hypothetical constructs and have to

be operationally defined. The operational definitions of an emotion include observations o f behavior, measurements o f physiological concomitants and

subjective reports o f subjective feelings o f emotion (Plutchik 1980). Self-report rating scales such as the Self-rating Depression Scale (Zung 1965), Beck's De-

pression Inventory (Beck et al. 1961) and the Depression Scale in Minnesota Multiphasic Personality Inventory (MMPI-D Scale) Dalstrom et al. 1960) are

commonly used to measure the depression state of an individual. These scales

are used widely for research and clinical purposes and thus have played an

important role in the practice of psychology and psychiatry. Most of these depression measurements were constructed on the basis of

clinical observations of patients diagnosed as depressed by means other than the test in question. The items included in such depression measurements are thus empirically derived. However, the observations and validations of these tests

were carried out in Western hospitals with patients o f Western culture. The items or symptoms included in the tests are thus not unlike Goodenough's conclusion: samples of the trait in question, in this case depression, as it is displayed in a particular culture context.

This sampling problem was evident even when the tests were applied to people o f the same culture. Lack of consistency of diagnosis of depression has been reported by researchers using different depression measuring instruments on the same subjects (e.g., Akiskal et al. 1979; Depue and Monroe 1978; and Hessel- brock et al. 1983). For instance, the D scale of the MMPI tends to "overdiagnose" subjects for depression (Depue and Monroe 1978; Hesselbrock et al. 1983). The

A CROSS-CULTURAL STUDY OF DEPRESSIVE SYMPTOMOLOGY 297

problem is amplified when the subjects are of different culture and ethnic

backgrounds (Dohrenwend and Dohrenwend 1965, 1967, and 1974). Faced with the sampling problem, Zung (1977) has emphasized the concept

of core symptoms contained in his Self-rating Depression Scale (SDS) (Zung 1965) which taps a group of symptoms commonly shared by people of different cultures. The list of twenty items in the SDS does not include religion-related items (see MMPI-D Scale for a comparison) and guilt related items (see BDI for a comparison). Cross-cultural studies have been conducted by Zung (1969, 1972) with the SDS; it was found that the twenty items on the list are symptoms experienced by people of a variety of cultures. This group of symptoms is perhaps a better sampling of depressive symptoms in a cross-cultural sense. This makes the SDS a useful tool with which to explore relations between symptoms and cultural variation in the configurations of symptoms.

It is agreed by most researchers that depression is a phenomenon of a multi- dimensional constellation of symptoms. The description of depression given by the Diagnostic and Statistical Manual-Third Edition (DSM-III) (American Psy- chiatric Association 1980) states the following:

The essential feature is either a disphoric mood, usually depression, or loss of interest or pleasure in all or almost all usual activities and past times. This disturbance is prominent, relatively persistent and associated with other symptoms, including appetite disturbance, change in weight, sleep disturbance psychomotor agitation or retardation, decreased energy, feeling of worthlessness or guilt, difficulty in concentration or thinking and thoughts of death or suicidal attempts.

This description not only fists the symptoms of depression but also the rela- tionships between the symptoms, that is, the affective complaints are major symptoms; somatic symptoms and feeling of loss and self-worth concerns are secondary concomitants which accompany the primary affective complaints. This description of depression with its emphasis on affective complaints repre- sents a consensus of the operational definitions of depression as it is viewed in

the United States (Eysenck, Wakefield, and Friedman 1983). When this cluster of symptoms was used as criteria for depression in cross-cultural studies, incidences of depression seem to be rare in non-Western cultures. Noticeably absent in the case reports were affective complaints (Marsella 1980). Cross- cultural surveys, however, indicated strong presences of somatic complaints, physiological symptoms, that usually accompany affective symptoms. Some of the researchers noted the presence of affective symptoms; but the symptoms were often de-emphasized during reporting or interviewing (Kleinman 1977, 1982a, 1982b). Since there are many theories and studies that explain the bio- logical mechanisms of depression, which seem to be universally applied, it seems that a more productive approach is not to argue for the presence or absence of certain symptoms but to search for the configurations and the underlying


dimensions of phenomenological manifestations of depression and to identify dimensions of symptoms that are better indicators of depression for people of different cultures. Aimed at that goal, recent cross-cultural studies have em- ployed multivariate techniques. For instance, Marsella, Kinzi, and Gordon (1973) using a sixty item symptom check list composed of a spectrum of symptoms on clinically depressed college students of white, Japanese and Chinese American backgrounds, found that subjects of different ethnic groups demon- strated qualitatively different symptomology of depression.

Irvine and Carroll (1980) suggested that a combination of multivariate psychometric methods with ethnographic studies can provide an effective tool to attack the complicated problem of testing subjects of different cultures.

Analyses of patterns of correlations and factor profiles can generate information about the nature of the process being studied; comparing the correlation matrices and factor profiles obtained from the test results of different cultural groups can provide an answer to the question whether the tests are measuring the same construct or underlying processes across different cultural groups.

Information gathered with ethnographic studies can provide background contextual information, especially the "knowledge system" of a culture which "shapes" the experience and the culturally conditioned ways of expressing underlying psychological processes.

Marsella (1980) suggested that the construct of depression: "is perhaps conceptualized differently in different cultures." A theoretical framework is proposed to explain culture variations in surveys of psychological symptoms (see Marsella and White 1982) that the cultural milieu in which the individual is brought up and has to interact with others provides an "epistemological framework" with which the individual perceives, interprets and categorizes his personal conditions. This epistemological framework also gives rise to culturally conditioned ways in which the individual communicates with others about his personal conditions including psychological symptoms. This appeal to cognitive intervention seems particularly applicable to self-report measures of depressive symptoms.

A self-report measure of psychological symptoms does not directly tap the symptoms; it fneasures the symptoms through a subjective report of the perception of symptoms. Cognitive processes involved in perceiving, categorizing and interpreting the symptoms and the culturally conditioned ways of communication are all potential factors that might influence the outcome of a self-report measure of symptoms.

Studies in individual cognitive function and perceptual development have suggested perception is a much more subjective process than it was once envi- sioned. Gibson (1969), based on empirical observations of child development, proclaimed that perception is a process of "filtering and selection" of relevant


information out of an array of information afforded by the environment. Neisser (1976) went a step further and proposed that perception is a process of"recon- struction of reality" based on a set of "schemata" formed, validated, and modi- fied by the individual's interaction with the environment.

The new ethnoscience which adopted linguistic models for the study of culture systems and focused largely on the semantic organization of terminological domains is concerned specifically with the task of discovering and representing cultural knowledge as a cognitive system (Ricoeur 1981). Researchers with backgrounds in ethnoscience have been engaged in the search for epistemological frameworks in different cultures (e.g., D'Adrade 1976, 1981).

These separate but overlapping lines of research all point to the process in which an individual utilizes an internalized cognitive model to select, categorize and to interpret information from the environment in which he is embedded.

Objective validation of subjective feelings about emotions is very difficult if not impossible. When faced with such illusive targets for perception, the perceiver of the subjective experiences of emotions has to rely on a set of cognitive references to "make sense of" , to label and to categorize the experiences. In other words, it is proposed that emotional experiences are perceived differently in different cultures because they are selected, labeled and categorized through different cognitive filters.

Self-reporting of psychological symptoms is essentially a communicative behavior. In a broad sense, all psychological testings are communicative behaviors. During the process of communication, to achieve the purpose of communication, the communicator has to use symbols and gestures that have been proven effective in the past. These symbols, gestures and display rules constitute a "language." In addition to studying cultural differences in tools used for communication, scholars of sociolinguistics (Bernstein 1970) are particularly concerned with the social contexts in which communications take place (Henderson 1973). Since communication styles vary with the contexts and the contexts are perceived differently by people of different cultures, Cole and Scfibner (1974) contended that the use of standard testing procedures is not really "standard" for people of different cultures and that field studies of the culture have to be conducted prior to the administration of tests or psychological experiments in order to identify situational settings that are culturally equivalent subjects of different cultures.

Psychologists are thus faced with a situation in which the psychological test results are end products of at least the following kinds of variables: (1) perceiving, categorizing and interpreting of the psychological traits, (2) communicative tools and styles used by the subjects and (3) situational and contextual variables involved during the administration of the test in addition to the trait being measured. The situation is quite complicated and can be attacked from many


different angles. Analyses of the final product, the test scores, represent one of the approaches to the problem. By analyzing differences in configurations and underlying dimensions of the scores, information can be obtained to serve as basis to chart directions for future approaches to help formulate better questions and hypotheses which will lead to a more comprehensive undersanding of the

problem. The present study used Zung's SDS (Zung 1965) as a tool to explore different

dimensions and configurations of depressive symptomology in Black, White and Overseas Chinese college students. Through factor analyses, the present study sought to (1) identify the basic categories or dimensions of depressive symptoms for each culture group, (2) to compare the factor profiles and study the pattern of culture variations in depressive symptomology and (3) to identify culturally salient dimensions of symptoms for diagnosis of depression in different cultures.

METHOD

Subjects

Students from two neighboring state universities in Houston, Texas were recruited for the study. Black students were enrolled in an introductory psychology class at Texas Southern University, a predominantly Black university; White and Overseas Chinese students were enrolled in introductory classes at the University of Houston/University Park. There were 26 Black students, 26 White students and 31 Overseas Chinese students. The gender distribution between male and female was 13:13 in the White group; 16:10 in Black group and 22:9 in the Overseas Chinese group. The average age for all three groups was 21.

Background information was obtained from the Overseas Chinese subjects. Their average length of study in the United States was two years. A majority of the students were from Hong Kong, where they had several years of high school education in English as a Second Language (ESL). In accordance with the admission policy of the University of Houston, no foreign students can be admitted into the University without having achieved a score on the Test of English as a Foreign Language (TOFEL) above 550. Therefore it is confident to assume that the Overseas Chinese subjects had no difficulty reading the items in the measurement of depressive sympt~3ms. Answers to questions added to the measurement confirmed this assumption.

Instruments

The Self-rating Depression Scales (SDS) developed by Zung (1965) was used. The SDS is a short list of twenty items which describe four categories of depressive


symptoms: pervasive psychic disturbance, physiological disturbance, psychomotor disturbance and psychological disturbance. Each item in the scale contains a short statement reflecting one of the symptoms. The subject is asked to respond to the statement by checking one of the following ratings: (a) a little of the time, (b) some of the time, (c) good part of the time and (d) most of the time. Items are rated on a four point scale, half positively beyed and half negatively keyed. In scoring the result, a response of (a) was given a 1 ; (b) was given a 2; (c) was given a 3; and (d) was given a 4, for the positively keyed items. For the negatively keyed items, a response of (a) was given a 4; (b), a 3; (c) a 2; and (d), a l .

Two additional questions were added to the instrument for the Overseas Chinese subjects: (1) Did you have any difficulty reading the questions? and (2) What do you think this questionnaire measures?

Procedure

The instrument was administered by the instructor of each of the classes. The subjects were instructed to do the rating as part of the course requirements. Anonymity was assured by instructing the subjects to put down only their age, gender and ethnic identity on the scale. The Overseas Chinese students were asked to give their length of stay in the United States. The scale was given without a title to avoid the subjects' interpretation of the term: "depression."

Data Analyses

One way analyses of variance were performed across the ethnic groups for each

of the items in the scale. Inter-item correlation coefficients were calculated for each of the ethnic groups separately.

Factor analyses via principal component analyses were conducted for each of the groups separately. The resulting factor matrices were rotated against each other. Item cosines were calculated for each of the items after the rotation; means of cosines were obtained as indicators of the extent to which the matrics were similar to each other.

RESULTS

The results of analyses of variances are presented in Table I. With the exception of two items, item 8 and item 13, there were no significant differences of depressive symptoms across these three ethnic groups.

302 WEINNING C. CHANG

TABLE I Ethnic differences in item responses

White Black Chinese df (N = 26) (N = 26) (N = 31)

a b a b a b

1. I feel downhearted and blue. 2. Morning is when I feel the best. 3. I have crying spells or feellike it. 4. I have trouble sleeping at night. 5. I eat as much as I used to. 6. I still enjoy sex. 7. I notice that I am loosing

weight. 8. I have trouble with constipation. 9. My heart beats faster than usual.

10. I got tired for no reason. 11. My mind is as clear as it used

to be. 12. I find it easy to do the things

I used to. 13. I am restless and can't keep still. 14. I feel hopeful about the future. 15. I am more irritable than usual. 16. I find it easy to make decisions. 17. I feel that I am useful and

needed. 18. My life is pretty full. 19. I feel that others would be

better off if I were dead. 20. I still enjoy the things I used to.

1.30-0.63 1.16-0.57 1.77-0.88 2/80 2.08 2.76-1.14 2.34-1.27 2.35-1.05 2/80 1.30 1.10-0.40 1.27-0.53 1.71-1.03 2/80 1.82 1.42-0.75 1.35-0.29 1.68-0.98 2/80 0.60 2.19-1.20 2.69-1.08 2.16-1.00 2/80 1.62 1.46-0.90 1.38-0.64 1.84-1.00 2/80 2.34

1.35-0.63 1.42-0.75 1.64-0.91 2/80 0.50 1.12-0.32 1.07-0.27 1.52-0.77 2/80 4.34* 1.50-0.86 1.38-0.69 1.32-0.47 2/80 0.57 1.58-0.58 1.50-0.70 1.93-0.85 2/80 1.77

1.57-0.54 1.96-1.11 1.67-0.74 2/80 1.40

1.81-0.98 2.11-1.17 1.87-0.92 2/80 1.06 2.08-0.74 1.50-0.90 2.26-1.15 2/80 4.14" 1.57-0.81 2.04-1.15 1.97-1.14 2/80 0.82 1.62-0.75 1.69-0.74 1.94-0.89 2/80 0.43 2.11-0.10 1.84-1.08 2.42-1.11 2/80 1.69

1.88-0.86 1.65-0.93 1.90-0.91 2/80 0.57 1.73-0.78 1.54-0.86 2.23-0.99 2/80 2.66

1.12-0.43 1.27-0.72 1.35-0.84 2/80 0.50 1.42-0.58 1.88-1.14 2.03-1.05 2/80 1.98

* p < 0.05 a. mean b. standard deviation

Table II, III, and IV p re sen t the in te r - i t em co r re l a t ion coef f ic ien ts for each o f

the e t h n i c groups. A close i n s p e c t i o n o f these co r re l a t ion coef f ic ien ts reveals

th ree d i f f e ren t p a t t e r n s o f c lus ter ing o f the i tems. F o r the Whi te group, a c lus ter

o f s ignif icant cor re la t ions is f o u n d b e t w e e n the i t ems t h a t c o n t a i n ex i s ten t ia l

conce rns and conce rns for cogni t ive f u n c t i o n i n g ; for the Black group , a c lus ter

o f s ignif icant co r re l a t ion coef f ic ien t s is f o u n d b e t w e e n the i t ems c o n t a i n i n g

somat i c c o m p l a i n t s and affect ive compla in t s , a n o t h e r c lus ter is f o u n d b e t w e e n

the cogni t ive and ex is ten t ia l concerns . Fo r the Overseas Chinese group, i t is

observed t h a t the self-esteem i t ems and cogni t ive c o n c e r n i t ems are related.

However , h igher cor re la t ions are f o u n d b e t w e e n the i t ems c o n t a i n i n g somat i c

compla in t s . This la t te r p a t t e r n seems to be charac te r i s t i c o f the Overseas Chinese

group.

TA

BL

E I

I In

ter-

item

cor

rela

tion

mat

rix

of t

he b

lack

dat

a > C

~ It

em N

o.1

2 3

4 5

6 7

$ 9

10

11

12

13

14

15

16

17

18

19

20

~0

O

~n

o'2

1.

1.00

2.

0.434

_.___

_* 1.

00

3.

0.09

1 0.

255

1.00

4.

0.19

9 0.

087

0.20

4 1.

00

5. -0

.263

-0

.201

-0

.127

0.

257

1.00

6.

0.09

3 0.

156

0.27

1 -0

.i0

6

-0.1

11

1.00

7.

-0.0

71

-0.2

60

0.20

2 0.

048

-0.2

24

-0.2

68

1.00

8.

-0.0

59

0.14

3 -0

.149

-0

.051

0.

219

0,51

5"*

-0.1

64

1.00

9.

0.68

8"*

0.32

4 -0

.074

0.

416

0.00

4 0.

104

-0,2

45

0.04

9 1.

00

10.

0.09

9 0.

223

0.15

9 -0

.222

-0

.260

0.

266

-0.0

37

0.00

0 0.

000

1.00

11.

0.03

9 0.

380

0.22

0 -0

.054

0.

122

0.19

1 0.

020

0.14

2 -0

.032

0.

076

1.00

12.

-0.I

t0

0.26

4 0.

076

0.07

8 02

73

0.05

4 -0

.057

0.

096

0139

0.

168

0.37

0 1.

00

13.

0.07

7 0.

105

0.12

4 0322

0.28

4 -0

.069

-0

.087

0.163

-0.0

63

0.156

0.298

-0.2

06

1.00

14.

0.08

4 0.

154

0.11

3 0.

014

0.17

0 -0

.130

-0

.065

-0

.138

-0

.119

-0

.271

0.

001

0.02

6 0.

019

1.00

15.

0.46

9*

0.38

9 0.

118

0.56

3**-

0.02

3 -0

.079

-0

.331

-0

;077

0.

474*

0.

231

0.03

4 0.

043

0.48

0"*-

0.I2

7 1.

00

16.

-0.4

17"

0.21

9 0.

005

0.23

2 0.

264

0.20

5 -0

.356

-0

.094

0.

505*

* 0.

104

0.02

8 0.

422*

-0.0

81

0.13

3 0.

289

1.00

17.

0.11

5 0.

285

-0.0

46

0278

0.

206

-0.3

04

0.10

2 -0

.408

* 0.

396"

0.

091

0.02

5 0.

219

0.02

4 0.

050

0.18

8 0.

221

1.00

18.

0.60

2**

0.64

7**

0.28

2 0.

305

0:09

9 -0

.101

-0

.116

-0

.013

0.

575"

* 0.

132

0.19

0 0.

331

0.15

4 0.

019

0.58

9**0

.264

0.

390*

1.0

0

19.

0.45

4*

0.31

8 0.

322

0.46

0"

0.26

2 -0

.067

0.

076

-0.1

09

.0,:

,579

" 0.

039

0.11

3 0.

431"

0.

030

0.17

9 0.

312

0.56

0 0.

379

0.65

7**

1.00

20.

0.35

8 0.

529"

* 0.

512"

* 0.

312

0.16

3 0.

283

-0.2

65

-0.0

09

0.40

9 0.

223

0.37

4 0.

516"

* 0.

097

0.21

7 0r

479"

0.

631"

* 0.

186

0.51

4"*

0£22

**

LO

0

C

C

> 0 r~

< 0 0 0

* p

< 0.05

=* p

< 0

.01

TA

BL

E I

II

0

inte

r-ite

m c

orre

latio

n m

atri

x o

f th

e w

hite

dat

a "~

lttm

No.l

2

3 4

5 6

7 8

9 10

11

12

13

14

15

16

17

18

19

20

i.

1.00

0

2.

0.237

1.000

3.

0.16

8 0.275

1.000

4.

0.147

-0.1

60

-0.0

15

1.000

5.

0.056

0.004

0.086

0.215

1.000

6.

0.235

0.185

0.186

-0.1

96

0.283

1.00

0

7.

0.154

-0.1

07

-0.1

16

-0.1

52

0.120

-0.0

01

1.00

0

8. -0

.222

0,182

0.129

-0.2

06

-0.3

36

-0.1

88

-0.0

08

1.00

0

9.

0.000

-0.1

63

0.174

0.092

-0.2

13

-0.2

57

0.03

7 -0

.071

1.

000

10.

0.025

-0.0

93

0.020

-0.1

23

-0.2

24

0,00

6 0,

199

0,05

7 0.

121

1,00

0

ii.

0.281

0,276

0.117

-0.0

75

-0,1

02

0.23

7 0.

256

-0.0

95

0.02

5 -0

.048

1,

000

12.

0.12

3 0.173

-0.0

04

0,114

-0.1

71

0,01

4 -0

.082

-0

.178

0,

166

-0,1

49

0.72

9"*

1.00

0

13.

0.35

7 -0

.214

-0.051

0.649"*

-0.0

17

-0.2

33

-0.0

59

-0.0

38

0.00

0 0.

079

-0.0

09

0.07

6 1.

000

14.

0.48

4"*

0119

3 0

.01

4

0.43

5*

-0.0

36

0.2

23

-0

.172

-0

.111

-0

.144

0.

030

0.12

3 0.

347

0.32

2 1.

000

15.

0.57

1"*

0.17

2 -0

.010

-0

.194

-0

.048

0.

330

-0.1

30

0.02

5 0.

185

0.07

1 0.

212

0.27

5 0

.12

6

.0..4

4.~*

1.

000

16.

0.331

0.054

--0.

052

-0,0

13

0.045

0.30

4 0.

457*

--

0.03

8 0.

189

0.39

2* 0.

087

--0,

126

0,18

3 0.

235

0.34

3 1.

000

17.

0,52

0*

-0,1

50

--0,

279

-0.0

45

0.215

0.22

4 0.

518"

--

0.23

5 0.

081

0,29

9 0,

134

0.02

0 0.

263

0,15

6 0.

421"

0.

684 =*

1,000

18.

0.459"*

0.062

0.044

--0.

003

-0.0

71

0.35

4 0.

280

--0.

303

-0.0

30

0.27

1 0.

(~2

-0.0

18

0.38

3 0.

257

0.29

5 0.

688*

* 0.667 ~*

1.000

19.

0.56

0**

0.056

0.328

-0.1

55

0.033

0.47

3*

0,43

7"--

0.09

8 --

0.05

4 0.

043

0,32

1 -0

.040

0.

096

0.26

0 0.

265

0.473*

0.466"

0,454*

1.000

20.

0,62

7"*

0.336

0.152

0.032

--0.

122

0.22

4 0.

131

--0,

057

0.04

0 --

0.04

1 0.

268

0,29

1 0.

107

(k57

0"*

0,57

3"*

0,421"

0,422"

0.53

1**

0.438*

1,000

Z C3

¢3

> Z C3

" p

< 0.05

** p

< 0.01

TA

BL

E I

V

Inte

r-it

em c

orre

lati

on m

atri

x of

the

Chi

nese

dat

a >

Item

No.

1 2

3 4

5 6

7 8

9 10

11

12

13

14

15

16

17

18

19

20

;0

©

1.

1.00

0

2.

0.16

1 1.

000

3.

0.253

-0.0

55

1.00

0

4.

0.14

4 -0

.015

0.

200

1.0

00

5.

0.26

8 -0

.119

0.

110

0.53

0"*

1.00

0

6.

-0.0

80

0.18

3 0.

018

0.21

7 0.

192

1.00

0

7.

0.3

10

-0

.246

0.4

84"*

0.

091

0.1

73 -

0.0

64

1.0

00

8.

0A23

" 0.

013

0.27

7 -0

.126

0.

061

-0.1

05

0.17

4 1.

000

9.

-0.0

59

-0.2

37

0.12

8 0.

303

0.30

7 0.

043

0.27

2 0.

077

1.00

0

10.

0.20

1 -0

.048

0.

204

0.41

3"

0.32

4 0

.14

3

0,4

39

"-0

.15

1

0.38

2*

1.00

0

11.

0.18

9 -0

.274

0.

047

0.1

26

0.

161

-0.3

38

0.02

2 0.

531

0.20

9 0.

071

1.00

0

12.

0.20

9 0.

118

0.06

4 0.

506"

* 0.

456*

0.

049

0.22

1 0.

003

0.3

27

0.

455"

0

.17

9

1.00

0

13.

-0.1

04

0.19

7 0.

259

0.01

7 -0

.095

--

0.07

8 0.

058

0.18

3 -0

.096

-0

.186

0.

061

-0.0

30

1.00

0

14.

0.39

0 0.

233

-0.1

21

0.05

0 0.

209

0.40

3*

0.21

2 0.

362*

0.

266

0.16

9 0.

222

0.3

13

-0.

044

1.00

0

15.

0.27

7 0.

025

0.19

5 -0

.063

-0

.025

-0

.049

0.

461"

* 0.

147

0.12

9 0.

213

0.11

8 -0

.092

0.

081

0.16

2 1.

000

16.

0.13

3 0.

096

0.19

4 0.

158

0.41

3"

0.24

0 0.

313

0.28

2 0.

301

0.48

3"*-

0.03

2 0.

248

-0.3

97*

0.22

0 0.

162

1,00

0

17.

0.47

1"*

0.10

7 0.

040

0.18

9 0.

31i

0.05

6 0.

238

0.07

4 0.

229

0.50

8"*

0.10

0 0.

024

-0.5

17"*

0.3

84"

0.11

6 0.

567"

* 1.

000

18.

0.44

1"

0.4

97

"*-0

.12

9

-0.0

26

0.13

0 0.

004

-0.0

93

0.19

2 -0

.018

0.

215

-0.0

33

0.36

2"-0

.053

0.

420*

0.

055

0.21

3 0.

322

1.00

0

19.

-0.0

68

-0.0

34

0.39

0*

-0.1

40

-0.2

29

0.07

0 0.

213

0.27

5 0

.28

9-0

.06

0

0.13

5 -0

.241

0.

178

0.15

2 0.

244

0.08

5 -0

.08

5-0

.14

0

1.00

0

20.

0.00

8 -0

.292

0.

293

-0.0

23

0.35

4*

0.26

7 0.

443

-0.0

65

0.32

3 0.

310

-0.2

05

0.1

11

-0.1

21

0.

116

0.00

2 0.

428*

0

.22

0-0

.07

4

0.10

3 1.

000

(b

C

,..-]

C

>

t'-'

C

0 r~

< ~n

,<

0 o 0 0

* p

<

0.0

5

*'*

p <

0.0

1 O

G

el


Principal component analyses with Varimax rotation (Harmon 1960)were separately performed for each of the ethnic groups. Orthogonal factors which reflect the underlying dimensions of the construct measured are obtained through these analyses. Seven factors were obtained from the Overseas Chinese sample. The items with relatively high loadings (> 0.40) on each of the factor do not overlap with items with high loadings on other factors. Seven factors were obtained from the White sample, with some overlapping of high loading items. Eight factors were obtained from the Black sample, also with some overlapping of items. Table V presents the results of factor analyses and item loadings.

The factor which contributes the most variance in the Black sample is characterized by a mixture of affective and somatic complaints; the factor which contributes the most variance in the White group is characterized by cognitive and existential concerns; the factor which contributes the most variance in the Overseas Chinese group is characterized by somatic complaints. Tables VI, VII, and VIII present the three major factors for each of the ethnic groups. Table IX, presents the item consines after rotating the factor matrices against each other. The means of cosines were calculated for each pair of the matrices and are used as indicators for the extent of similarity between these matrices. The mean of cosines between Black and White samples is 0.6489; between Chinese and Black sample is 0.5921; between White and Chinese samples is 0.6070. None of these values reached the value of 0.85 suggested by Kaiser (1958). These values indicated that the factor matrices were sufficiently different to warrant the conclusion that there are different dimensions underlying the self-reported symptoms for the ethnic groups.

CONCLUSION AND DISCUSSION

The present study used Zung's SDS as a tool to explore cultural differences in configurations of depressive symptomology. The results indicated that the three college student groups, Black, White, and Overseas Chinese, evidenced different configurations as well as different underlying dimensions of the symptoms of depression. Item analyses revealed that two of the items showed marginally significant differences across the cultural groups. The two items, constipation and psychomotor agitation, perhaps reflect cultural differences in reporting such symptoms; it is socially inappropriate to report problems of constipation for American subjects; however, it is not socially inappropriate to report such a symptom for the Chinese, because constipation is a part of the somatic function- hag of a person and the Chinese have less inhibitions about reporting bodily symptoms than affective symptoms (Lin 1953; Tseng and Hsu 1969). Psycho- motor agitation or restlessness is considered more pathological than psychomotor retardation by the Chinese (Lin 1953). The higher mean score and larger variance

TA

BL

E V

R

esul

t of

fact

or a

naly

ses

per

form

ed s

epar

atel

y fo

r ea

ch e

thni

c gr

oup

>

©

Fac

tor

I F

acto

r II

F

acto

r II

I /

Item

s W

hite

B

lack

C

hine

se

Whi

te

Bla

ck

Chi

nese

W

hite

B

lack

C

hine

se

1.

I fe

el d

ownh

eart

ed a

nd b

lue.

0.

785

0.74

1 0.

116

0.18

9 -0

.42

7

0.31

4 0.

129

0.21

6 0.

518

2.

Mor

ning

is w

hen

I fe

el t

he b

est.

0.

199

0.46

3 -0

.04

4

0.48

8 -0

.13

9

-0.1

62

-0

.41

6

0.17

8 0.

832

3.

Ihav

e cr

ying

spe

lls o

r fe

elli

kei

t.

0.10

1 0.

071

0.09

4 0.

285

-0.0

58

, -0

.16

9

-0,2

87

-0

.17

5

-0.0

74

4.

I

have

tor

uble

sle

epin

g at

nig

ht.

0.02

6 0.

546

0.82

0 0.

185

0.25

0 -0

.04

5

0.85

8.

-0.0

96

-0

.01

7

5.

I ea

t as

mu

chas

Iu

sed

to.

0.

047

-0.0

27

0.

653

-0.1

28

0.

794

0.27

5 0.

060

0.16

1 -0

.06

2

6.

I st

ille

njoy

sex

. 0.

465

-0.0

64

0.

108

0.05

9 -0

.11

9

0.03

5 -0

.47

6

0.30

5 0.

140

7.

Ino

tice

tha

t I

am l

osin

g w

eigh

t.

0.35

3 -0

.01

9

0.17

5 -0

.54

3

-0.0

14

0.

222

-0.1

92

-0

.83

8

-0.0

84

8.

I

have

tro

uble

wit

h c

onst

ipat

ion.

-0

.18

4

-0.0

15

-0

.13

2

-0.0

08

0.

113

-0.0

17

-0

.25

5

-0.0

24

0.

147

9.

My

hea

rt b

eats

fas

ter

than

usu

al.

0.04

3 0.

885

0.48

4 -0

.04

5

-0.0

10

0.

204

0.14

1 0.

237

-0.3

15

10

. Ig

et t

ired

for

no r

easo

n.

0.20

9 -0

.05

2

0.58

1 -0

.44

7

-0.0

92

0.

454

-0.0

16

0.

186

0.08

4 11

, M

y m

ind

is a

s cl

ear

as i

t us

ed t

o be

. 0.

380

-0.0

55

0.

244

0.45

2 0.

257

-0.0

09

-0

.25

2

-0.0

78

-0

.18

8

12.

I fi

nd i

t ea

sy to

do

the

thin

gs I

use

d to

do.

0.

230

0.17

8 0.

824

0.66

4 0.

757

0.02

5 0.

080

0.13

4 0,

215

13.

I am

res

tles

s an

d c

an't

kee

p st

ill.

0.30

9 -0

.04

7

0.05

7 -0

.01

4

0.01

8 -0

.81

4

0.77

6 0.

011

0.13

5 14

. I

feel

ho

pef

ula

bo

ut t

he f

utur

e.

0.54

9 -0

.04

7

0.18

5 0.

456

-0.0

01

0.

235

0.35

5 0.

216

0,43

0 15

. I

am m

ore

irri

tabl

e th

an u

sual

. 0.

637

0.54

3 -0

.09

1

0.26

9 -0

.08

9

0.09

6 -0

.06

4

0.32

0 0.

191

16.

I fi

red

it e

asy

to m

ake

deci

sion

s.

0.72

2 0.

433

0.18

3 -0

.45

0

0.32

2 0.

634

-0.0

78

0.

657

0,07

2 17

. I

feel

tha

t I

am u

sefu

l an

d ne

eded

. 0.

755

0.57

4 0.

129

-0.4

67

0.

362

0.82

6 0.

111

-0.2

33

0.

237

18.

My

life

is p

rett

y f

ull.

0.74

9 0.

755

0.15

3 -0

.30

7

0.06

3 0.

232

0.06

6 0.

039

0.77

3 19

. I

feel

that

oth

ers

wou

ld b

e b

ette

r o

ff if

I w

ere

dead

. 0.

706

0.71

5 -0

.26

6

-0.1

12

0.

328

-0.1

92

-0

,27

2

-0.0

57

-0

.13

7

20.

I st

ill e

njoy

th

e th

ings

I u

sed

to d

o.

0.76

6 0.

415

0.09

7 0.

300

0.27

9 0.

368

-0.0

27

0.

442

-0.3

19

>

0 <

O~

0 0 0

Loa

ding

s la

rger

tha

n 0.

40 a

re u

nder

line

d.

~o

o


TABLE VI The first three prominent factors of the black data

Rank order of factor loading self-rating depression scale - the black group

Facrot I Eigenvatue 5.1480, % of variance 19.8%

Rank order of loading

1. My heart bea~s faster than usual. 0.885 2. My life is pret~y full. 0.755 3. 1 feel down~earted and blue. 0.741 4. I feel that others would be better off if I were dead. 0.715 5. I feel that I am useful and needed. 0.575 6. I have trouble sleeping at night, 0.594 7. I am more irritable than usual. 0.534 8. Morning is when I feet the best. 0.463 9. I find it easy to make decisions. 0.433

10. I still enjoy the things I used to do. 0.415

Factor II Eigenvahie 2.2635, % of variance 10.0%

Rank order of loading

1. I eat as much as I used to. 0.794 2. I find it easy to do the things I used to. 0.757 3. I feel down hearted and blue. -0 .427 4. I feel that I am losing weight. 0.362 5. I feel that others would be better off if I were dead. 0.382 6. I find it easy to make decisions. 0.322

Factor III Eigenvalue 1.9885, % of variance 9.5%

Rank order of factor loading

1. I noticed that I am losing weight. -0 .838 2. I find it easy to make decisions. 0.657 3. I still enjoy the things that I used to do. 0.442 4. I am more irritable than usual. 0.320 5. I feel that others would be better off if I were dead. 0.305

There are eight factors; five of the factors have eigenvalue less than 2.

o f p s y c h o m o t o r ag i t a t ion r e p o r t e d b y the Overseas Chinese subjec t s in this s tudy

pe rhaps ind ica tes salience o f this s y m p t o m as an i nd i ca to r o f psycho log ica l well-

be ing o f a pe rson .

In t e r - i t em cor re la t ions and s u b s e q u e n t fac to r analyses revealed cu l tu ra l varia-

t ions in p a t t e r n s o f se l f - repor t s y m p t o m s . F a c t o r analyses o f resul ts y ie lded

th ree d i f f e ren t f ac to r profiles. I t is pe r haps no t surpr is ing to f ind d i f fe ren t f ac to r

p a t t e r n s b e t w e e n the Chinese subjec ts and the A m e r i c a n subjec ts ; i t is surpris ing

to f ind d i f fe ren t f ac to r p a t t e r n s b e t w e e n the A m e r i c a n groups. The da ta ob-

t a ined f r o m the Black sample co r r e sponds w i t h an earl ier s t u d y b y S imon et al.

( 1 9 7 3 ) in t h a t qual i ta t ive d i f fe rences in s y m p t o m o l o g y were f o u n d b e t w e e n

A C R O S S - C U L T U R A L STUDY OF D E P R E S S I V E S Y M P T O M O L O G Y

TABLE VII The first three prominent factors in the white data

309

Rank order of factor loading on SDS - The white group

Factor I, Eigenvalue 4.829, % of total variance 17.9%

Rank order of item loading

1. I feel that I am useful and needed. 2. I find it easy to make decisions. 3. I have crying spells or feel like it. 4. My life is pretty full. 5. I feel that others would be better off if I were dead. 6. I get tired for no reason. 7. I feel down hearted and blue. 8. I still enjoy the things I used to do.

Factor II, Eigenvalue 2.4489, % of total variance 14.8%


1, I am more irritable than usual. 2, I feel hopeful about the future. 3, I still enjoy the things I used to do. 4, I feel down hearted and blue. 5. I still enjoy sex. 6. I notice that I am loosing weight. 7. My life is pretty full. 8. Morning is when I feel the best.

Factor III, Eigenvalue 2,2543, % of total variance 10.6%


1. I have trouble sleeping at night. 2. I am restless and can't keep still. 3. I still enjoy sex. 4. I feel hopeful about the future.

0.820 0.801 0.774 0.722 0.657 0.435 0.423 0.325

0.833 0.716 0.695 0.589 0.415

-0 .406 0.396 0.336

0.858 0.776 0.476 0.407

There were seven factors. Because of low eigenvalue only three of the factors are presented here.

Black and Whi te subjects . In a n a t i o n a l survey c o n d u c t e d b y Levi t t and Lub in

( 1 9 7 5 ) us ing the Depressive Adjec t ive Check List (DACL) , i t was f o u n d t h a t

Blacks in t h a t sample scored s igni f icant ly h igher t h a n the Whi tes o n t he D A C L

w h i c h cons is t s o f a series o f af fec t ive descr ipt ive t e rms (i.e., sad, h a p p y , b lue ,

etc.). The first f ac to r o b t a i n e d f r o m t he Black sample in the p re sen t s t udy

s h o w e d m o r e o f the p r i m a r y affect ive s y m p t o m s desc r ibed in DSM-III. In o t h e r

words , the s y m p t o m c o n f i g u r a t i o n o b t a i n e d f r o m the Black subjec t s s h o w e d

more r e semb lance to a "c lass ica l " s y m p t o m o l o g y o f depress ion t h a n t h a t ob-

t a i n e d f r o m the Whi te or the Overseas Chinese samples.

Because t he subjec ts were all college s tuden t s , t he re is r o o m to argue t ha t the


TABLE VIII The first three prominent factor in the Chinese data

Rank order of factor loading on SDS - The overseas Chinese group

Factor I Eigenvalue 4.3172, % of total variance 13.5%


1. I t'md it easy to do the things I used to. 2. I have trouble sleeping at night. 3. I eat as much as I used to. 4. I get tired for no reason. 5. My heart beats faster than usual.

Factor II Eigenvalue 2.490, % of total variance 12.8%


1. I feel that I am useful and needed. 2. I am restless and can't keep still. 3. I find it easy to make decisions. 4. I get tired for no reason. 5. I still enjoy the things I used to. 6. I feel downhearted and blue.

Factor III Eigenvalue 2.3259, % of variance 11.1%


1. Morning is when I feel the best. 2. My life is pretty full. 3. I feel downhearted and blue. 4. I feel hopeful about the future. 5. I still enjoy the things I used to. 6. My heart beats faster than usual.

0.824 0.820 0.653 0.58 0.48

0.826 -0.814

0.63 0.45 0.36 0.31

0.83 0.77 0.51 0.43

-0.319 -0.315

There were seven factors. Because of low eigenvalues, only three of the factors are presented here.

Black/White differences observed in this s tudy reflects more o f a Black/White

college s tudent difference than a Black/White dif ference in the popula t ion at

large. Since a large number o f the s tudents at Texas Southern Univers i ty are on

financial aid and m a n y of the s tudents are f rom working class backgrounds, i t

is unclear whe ther the Black/White difference observed here reflects a class

difference or an ethnic difference. For this reason, the present s tudy is no t

immune f rom the p rob lem that plagues m a n y o f the o ther studies in Black/

White e thnic differences in the Uni ted States, that is, e thnic differences are o f ten

masked or con founded by significant class differences caused by the large dis-

pari ty in economic condi t ions be tween Black and White Americans. Studies

by Handerson (1973) comparing language differences be tween working class

mother-chi ld diads and middle class mother-chi ld diads in England no ted that

A C R O S S - C U L T U R A L STUDY OF DEPRESSIVE SYMPTOMOLOGY 311

TABLE IX Results of rotating factor matrices

Item cosines and mean item cosines of factor matrices after rotated against comparison factor matrices

White/Black White/Chinese Black/Chinese Mean = 0.6489 Mean = 0.6070 Mean = 0.5921

1. I feel down hearted and blue. 0.6645 0.6068 0.3347 2. Morning is when I feel the best. 0.6664 0.4354 0.7455 3. I have crying spells or feel like it. 0.9014 0.7392 0.9585 4. I have trouble sleeping at night. 0.6549 0.8173 0.3913 5. I eat as much as I used to. 0.5259 0.6394 0.7097 6. I stillenjoy sex. 0.4805 0.8263 0.6532 7. I notice that I am losing wight. 0.6655 0.5813 0.5063 8. I have trouble with constipation. 0.4448 0.3127 0.5063 9. My heart beats faster than usual. 0.0465 0.7518 0.7020

10. I got tired for no reason. 0.4925 0.5594 0.4062 11. My mind is as clear as it used to be. 0.8003 0.8677 0.2617 12. I find it easy to do the things I used

to. 0.6419 0.4912 0.7138 13. I am restless and can't keep still. 0.7766 0.3248 0.8640 14. I feel hopeful about the future. 0.5832 0.7439 0.3273 15. I am more irritable than usual. 0.5877 0.4596 0.6517 16. I find it easy to make decisions. 0.4858 0.8575 0.7923 17. I feel that I am useful and needed. 0.8953 0.9602 0.6931 18. My life is pretty full. 0.9044 0.6213 0.7153 19. I feel that others would be better

off if I were dead. 0.9040 0.4018 0.3281 20. I still enjoy the things I used to. 0.8565 0.1431 0.5810

the work ing class mo the r s used m a n y affect ive terms while the middle class

mothers used more cognit ive terms. This in fo rma t ion reinforces the possibi l i ty

that the current Black/White differences in depressive s y m p t o m o l o g y is perhaps

a class difference in perceiving and in terpret ing depressive symptoms. In light

o f the fact that the Black s y m p t o m o l o g y resembles more o f a "classical" syn-

d rome o f depression, it is logical to assume tha t the s y m p t o m o l o g y of the Black

college s tudents is more like depression in the general popula t ion and it is the

White depressive s y m p t o m o l o g y tha t ref lects a " subcu l t u r e " o f middle class

intellectuals.

The in ter- i tem corre la t ion mat r ix ob ta ined f rom the Chinese sample indicated

tha t the somat ic complain ts only have low correlat ions wi th existent ial concerns

and the affect ive symptoms . Fac to r analysis o f the Chinese data also showed

that somat ic compla in ts cons t i tu te a relatively or thogonal dimension. The same

d imens ion was no t observed in the Black or the White sample. It is interest ing

to consider this f inding wi th long standing observat ions that for the Chinese

psychiatr ic patients , somat ic complain ts are the chief presenting symptoms


(Cheung 1982; Kleinman 1977, 1982a, 1982b; Tseng and Hsu 1969). Results in the present study did not indicate that the Chinese had quantitatively less affective complaints, only that the affective complaints were not "grouped" with the somatic complaints. The dimension of somatic complaints contributed to the largest proportion of variance indicating that the somatic complaints are more "indicative" of depression for the Chinese. Finally a convergence is reached, that is both psychiatric reports and psychometric studies (Marsella et al. 1973) indicate a phenomenon of "somatization" by the Chinese (Kleinman 1977, 1982b).

Anthropological studies of the belief systems in Asian Cultures have shown that many Asian cultures believe in a close interaction between the mind and the body, and see the mind and the body as an integral whole. Thus, the same problem, be it physical or psychological in origin, will find its expression in both the physical and the psychological domains of a person's life (Kleinman and Lin 1981 ; Sue and Sue 1974; Sue et al. 1976, and Tseng 1975).

It should be emphasized that "somatization" in symptom presentation does not mean that the Chinese language does not have terms or concepts that make fine distinctions between different kinds of affective feelings or that the affective symptoms are suppressed or even repressed by the Chinese. The affective symptoms of depression or any other kind of emotion, for that matter, are experienced and communicated differently by the Chinese. A full investigation of the nature of the epistemological framework concerning emotions in the Chinese culture is beyond the scope of the present paper. However, even a cursory reading into Chinese literature would reveal a wealth of terms which describe a large variety of different human emotions. There is certainly no shortage of terms that convey sadness and despair in Chinese language. To fully test the conceptual equivalency of the term depression cross-culturally, a nomological net (Cronbach and Meehl 1955) has to be established for the term. At the present time, such a nomological net has not yet been established. A search into the Chinese language for terms of depression has yielded a term which seems to reflect the affective aspect of depression; YOU-YU (melancholia)(Zhihai 1957, pp. 1171). This term is basically descriptive of mood, and has a connotation of subjective feelings of a prevailing mood of sadness and despair and listlessness combined with a tendency toward crying. This description fits the description of the affective item is SDS: "I feel downhearted and blue." It is therefore argued that the affective aspect of depression does have its counterpart in the Chinese culture. Only this affective aspect of depression is treated and reported differently by the Chinese. For future studies on depression in Chinese, semantic differential techniques (Osgood 1964) shall be used to map out the domain of meaning of the term YOU-YU to test the concept equivalency between YOU-YU

and "depression".


Wu (1982) argued that there are different display rules for emotion in Chinese culture and that the difference in display rules is responsible for the absence of affective complaints by the Chinese. The present author would like to propose that more fundamental to the display rules are a set of symbols and gestures which constitute the "language of emotion" for the Chinese. Chinese frequently use metaphor in communication. Somatic concerns are often used by the Chinese to convey affective meanings. Among the Chinese, giving and withdrawal of affection are customarily carried out via giving and withdrawal of goods and services intended for physical comfort. Between parents and children, husbands and wives, material goods and services are exchanged for each other's physical well-being while no words of affection have been uttered. It is understood by the provider and the receiver that the exchange means affection. Thus, a language of emotion is spoken eloquently between the Chinese without having to invoke emotional words or terms. Those terms are usually reserved for literature but not for everyday interaction. Similarly, individual happiness and unhappiness are also expressed in a set of somatic terms. Heartache means sadness; fatigue, tired- ness means hurt and despair. The speaker uses this set of somatic metaphors to convey his (her) emotional conditions and the listener understands that these are metaphors which actually convey meanings of emotion. What is needed is longitudinal systematic studies in interpersonal communications between the Chinese to observe the nature and the development of this language of emotion, and the situations or social context in which the language is used.

Situational variables involved in psychological research have been systemati- cally studied in a series of studies. Orne (1962) used the term "demand characteristics" to describe the configuration of stimuli present in a psychological study that involves human subjects. The demand characteristics oftentimes confound the experimental results. In a broader sense this concept of demand characteristics can be applied to psychological testing and psychiatric interviewing where human performance is required. Research in cross-cultural studies have identified situation variables which radically change the performances of subjects (Labov 1970; LeVine 1970). Furthermore, Labov (1970) and Cole et al. (1971) suggested that people of different cultures are likely to respond differently to any given situation. The study conducted by Cole et al. (1971) in KpeU farmers illustrated the point clearly. Each culture has its own conception as to the right time and the right place to do the right behavior. Under the broad epistemological framework to which an individual subscribes in orde? to interpret social events, situational variables involved in a testing situation are likely to be perceived differently. The demand characteristics would thus differ for people of different cultures even with standard testing procedures (Cole and Bruner 1971). It is noted that evidence of "somatization" has been mostly obtained in medical settings (Cheung 1982; Kleinman 1977, 1982a, 1982b) where the demand


characteristics were such that it was more conducive to voice medical (somatic) complaints. The present study was conducted in psychology classes, a relatively

formal and impersonal setting. The SDS was given without a title. Answers to the question added to the end of the SDS: "What do you think this questionnaire measures?" indicated that a majority of the Chinese students thought it

was a measure of "general health conditions o f the subject". None of the subjects

thought it was a measure of depression. If such answers can be used to infer the perceived demand characteristics by the Chinese, it can then be deduced that the

Chinese subjects were more oriented to answer the somatic items of the scale. The study was conducted at the wake of the "human potential movement"

where the zeitgeist among the college students in the United States was charac-

terized by the "search for the self" or the "search for the meaning of life" (Lasch 1978). Could it be that the White samples in the present study were

using an epistemological framework and a set of languages that are based on

existentialist ideas embedded in the culture of the 1960's and 1970's in the

United States, much in the same way as the Chinese students used a somato-

psychic cognitive system to perceive and to convey emotional conditions?

Generalization of conclusions drawn from the results are limited because (1) the subjects were college students attending American universities and (2) the

sample sizes were small for multivariate analyses. Given such limitations, the results are encouraging in that they provide information complementary to data obtained in ethnoscience. The factor profile obtained from each ethnic group

can provide additional information to help validate the construct validity of existing standard measuring instruments of depression and to help construct new measurements of depression for subjects o f different cultures. The different

configurations of symptomology can also be used as a reference for diagnosing

depression for subjects of different cultures.

Department of Psychology Texas Southern University Houston, TX 77004, U.S.A.

WEINING C. CHANG

REFERENCES

Akiskal, H. S. et al. 1979 Differentiation of Primary Affective Illness from Situational Symptomatic and

Secondary Depression. Archives of General Psychiatry 36: 635-643. Al-Issa, I.

1970 Culture and Symptoms. In C. Costello (ed.), Symptoms of Psychopathology: A Handbook. New York: Wiley.

American Psychiatric Association 1980 Diagnostic and Statistical Manual, Third Edition. Washington, D.C.: American

Psychiatric Association.

A C R O S S - C U L T U R A L STUDY OF D E P R E S S I V E S Y M P T O M O L O G Y 315

Beck, A. et al 1961 An Inventory for Measuring Depression. Archives of General Psychiatry 4: 53-63.

Benedict, R. and Jack, I. 1954 Mental Illness in Primitive Societies. Psychiatry 17: 377-389.

Bernstein, B. 1970 In A. Ahronson et al. (eds.), Social Class, Language and Socialization, Current

Trends in Linguistics, 12. The Hague: Mouton. Cole, M. et al.

1971 The Culture Context of Learning and Thinking. New York: Basic Books. Cole, M. and Brunner, J. S.

1971 Culture Difference and Inference about Psychological Processes. American Psy- chologist 26: 867-876 .

Cole, M. and Scribner, S. 1974) Culture and Thought: A Psychological Introduction. New York: John Wiley and

Sons. Cronback, L. J. and Meehl, P. E.

1955 Construct Validity in Psychological Tests. Psychological Bulletin 5 2 : 2 8 1 - 302. Cheung, F. M.

1982 Psychological Symptoms Among Chinese hi Urban Hong Kong. Social Science and Medicine 16: 1339-1344.

D'Andrade, R. G. 1976 A Propositional Analysis of U.S. American Beliefs about Disease. In K. Bosso and

H. Selby (eds.), Meanings hi Anthropology. Albuquerque: University of New Mexico Press.

1981 The Culture Part of Cognition. Cognitive Science 5: 179-195. Dalstrom, W., Welsh, G., and Dalstrom, L

1960 A MMPI Handbook. Minneapolis: University of Minnesota Press. Depue, R. A. and Monroe, S. M.

1978 Learned Helplessness in the Perception of the Depressed Disorders: Conceptual and Definition Issues. Journal of Abnormal Psychology 87: 3-20.

Dohrenwend, B. P. and Dhrenwend, B. S. 1965 The Problems of Validity in Field Studies of Psychological Disorder. Journal of

Abnormal Psychology 70: 52-66 . 1969 Social Status and Psychological Disorder: A Casual Inquiry. New York: Wiley-

Intersciences. 1974 Social and Cultural Influence on Psychopathology. Annual Review of Psychology

25: 417-452 . Eysenck, H. J., Wakefield, J. A., and Friedman, A. F.

1983 Diagnosis and Clinical Assessment: The DMS-III, Third Edition. Annual Review of Psychology 34: 167-194.

Fabrega, H. 1974 Problems Implicit in the Culture and Social Study of Depression. Psychosomatic

Medicine 36(5): 377-398. Gibson, E. J.

1969 Principles ° f Perceptual Learning and Devel°pment" New Y°rk: Applet°n'Century- Croft.

Goodenough, F. 1936 The Measurement of Mental Functions in Primitive Groups. American Anthro-

pologist 3 8 : 1 - 1 1 . Harmon, H. H.

1960 Modern Factor Analysis. Chicago: University of Chicago Press. Hesselbrock, M. N. et al.

1983 Methodological Considerations in the Assessment of Depression in Alcoholics. Journal of Consulting and Clinical Psychology 51 (3): 399- 405.


Henderson, D. 1973 Contextual Specificity, Discretion and Cognitive Socialization; with Special

Reference to Language. In B. Bernstein (ed.), Class, Codes and Control: Applied Studies Towards a Sociology of Language, VoL II: 48-78 .

Irvine, S. H. and Carroll, W. K. 1980 Testing and Assessment across Cultures. In H. C. Triandis and J. W. Berry (eds.),

Handbook of Cross-cultural Psychology, Vol. II, Methodology, pp. 181-243. Boston: Allyn and Bacon.

Kaiser, H. F. 1958 The Varimax Criterion for Analytical Rotation in Factor Analysis. Psychometrika

23: 187-200. Kleinman, A.

1977 Depression, Somatization and the "New Cross-Cultural Psychiatry." Social Science and Medicine 11 : 3-10.

1982a Patients and Healers in the Context of Culture. Berkeley: University of California Press.

1982bNeurathenia and Depression: A Study of Somatization and Culture in China. Culture, Medicine and Psychiatry 6(2): 117-190.

Kleinman, A. and Lin, T. Y. 1981 Normal and Abnormal Behavioi in Chinese Culture. Boston: D. Reidel Publishing

Co. Labov, W.

1970 The Logic of Non-Standard English. In F. Williams (ed.), Language and Poverty. Chicago: Markam Publishing Co.

Lasch, C. 1978 The Culture of Narcissism: American Life in an Age of Missing Expectations. New

York: Norton. Le Vine, R. A.

1970 Cross-Cultural Study in Child Psychology. In Mussen (ed.), ~armichael's Manual of Child Psychology, Vol. II: 559-614.

Levitt, F. and Lubin, B. 1975 Depression, Concepts, Controversies and Some New Facts. New York: Springer.

Lin, T. Y. 1953 Mental Disorders in Chinese and other Cultures. Psychiatry 16: 313-336.

MarseUa, A. 1980 Depressive Affect and Disorder across Cultures. In H. Triandis and J. Draguns

(eds.), Handbook of Cross-Cultural Psychology, Vol. V, Psychopathology. Boston: Allyn and Bacon.

MarseUa, A., Kinzie, J., and Gordon, P. 1973 Ethnocultural Variations in the Expression of Depression. Journal of Cross-

Cultural Psychology 4: 453-458. Marsella, A. J. and White, G. M. (eds.)

1982 Cultural Conceptions of Mental Health and Therapy. Boston: D. Reidel Publishing Co.

Neisser, U. 1976 Cognition and Reality, Principles and Implications of Cognitive Psychology. San

Francisco: W. H. Freeman and Co. Orne, M. T.

1962 On the Social Psychology of the Psychological Experiment: With Particular Reference to Demand Characteristics and Theory Implications. American Psy- chologist 17: 776-783.

Osgood, C. E. 1964 Semantic Differential Technique in the Comparative Studies of Cultures. American

Anthropologists 6 6 : 1 7 1 - 2 0 0 .

A C R O S S - C U L T U R A L STUDY OF D E P R E S S I V E S Y M P T O M O L O G Y 317

Pfefffer, W. 1968 The Symptomology of Depression Viewed TransculturaUy. Transcultural Psy-

chiatric Research Review 15 : 121-123. Plutchik, R.

1980 Emotion, a Psychoevolutionary Synthesis. New York: Harper and Row. Ricoeur, P.

1981 In J. B. Thomas (ed.), Hermaneutics and the Human Sciences. Cambridge, Eng- land: Cambridge University Press.

Sue, S. and Sue, D. W. 1974 MMPI Comparisons between Asian American and Non-Asian Students Utilizing a

Student Health Psychiatric Clinic. Journal of Counseling Psychology 21(5): 4 2 3 - 427.

Sue, S. et al. 1976 Conceptions of Mental Illness amoung Asian and Caucasion American Students.

Psychological Reports 38: 703-738. Tseng, W. S.

1975 The Nature of Somatic Complaints Among Psychiatric Patients: The Chinese Case. Comprehensive Psychiatry 16: 237-245.

Tseng, W. S. and Hsu, J. 1969 Chinese Culture, Personality Formation and Mental Illness. International Journal

of Social Psychiatry 6: 5 -14 . Simon, R. et al.

1973 Depression and Schizophrenia in Hospitalized Black and White Mental Patients. Archives of General Psychiatry 28 : 509 - 512 .

Wu, D. H. Y. 1982 Psychotherapy and Emotion in Traditional Chinese Medicine. In A. J. Marsella

G. M. White (eds.), Culture Conceptions of Mental Health and Therapy. Boston: D. Reidel Publishing Co.

Zhihai 1957 Zhihai (Ocean of Words and Terms). Taipei: Shangwu (Commercial) Printing

House. Zung, W. W. K.

1965 A Self-Rating Depression Scale. Archives of General Psychiatry 1 2 : 5 3 - 60. 1969 A Cross-Cultural Survey of Symptoms in Depression. American Journal of Psy-

chiatry 126: 116-121. 1972 A Cross-Cultural Survey of Depressive Symptoms in Normal Adults. Journal of

Cross-cultural Psychology 3 : 177-183. 1977 Operational Diagnosis and Diagnostic Categories of Depressive Disorder. In

Phenomenology and Treatment of Depression. New York: Spectrum Publishing, Inc.

Documents

A cross-cultural study of depressive symptomology