Response Styles in a Cross-Cultural Managerial Study

This article was downloaded by: [UNSW Library] On: 23 February 2015, At: 16:27 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Journal of Social Psychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/vsoc20

Response Styles in a CrossCultural Managerial Study a

B. W. Stening & J. E. Everett

a

a

Department of Management , University of Western Australia Published online: 01 Jul 2010.

To cite this article: B. W. Stening & J. E. Everett (1984) Response Styles in a CrossCultural Managerial Study, The Journal of Social Psychology, 122:2, 151-156, DOI: 10.1080/00224545.1984.9713475 To link to this article: http://dx.doi.org/10.1080/00224545.1984.9713475

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

Downloaded by [UNSW Library] at 16:27 23 February 2015

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

The Journal of Social Psychology, 1984, 122, 151-156.

RESPONSE STYLES IN A CROSS-CULTURAL MANAGERIAL STUDY* Department of Management, University of Western Austratia

B. W.

STENING AND

J. E.

EVERETT

Downloaded by [UNSW Library] at 16:27 23 February 2015

SUMMARY

An examination was made of the response sets of a total of 1647 expatriate and local managers from nine countries who were respondents to a study of stereotyping undertaken in American, British, and Japanese firms in Singapore, and in Japanese firms in Britain, Hong Kong, Indonesia, Malaysia, the Philippines, and Thailand. The data consisted of responses to semantic differential items. There were marked differences between various nationalities in terms of their tendency to use extreme responses and midpoint responses. Moreover, age and education were shown to have an effect in certain instances. A.

INTRODUCTION

Cross-cultural research is beset by a number of methodological problems in addition to those confronted in carrying out a research project within one society (2). In large part, they constitute what might be called the equivalence problem: ensuring the selection of comparable samples; the development of research instruments that convey the same meaning to respondents of different cultural (and, probably, linguistic) background; and ensuring that there is no bias in the way that respondents from different cultures use particular scales (for example, Likert or semantic differential scales). With respect to the problem of bias, Kerlinger (3, pp. 496-497) has pointed out that individuals have different tendencies to use certain types of responses: extreme, neutral, agree, or disagree. Of particular concern to the cross-cultural researcher is the possibility that one or more of the national groups being studied is especially prone to a certain response-set

* Received in the Editorial Office, Provincetown, Massachusetts, on March 21, 1983, and given special consideration in accordance with our policy for cross-cultural research. Copyright, 1984, by The Journal Press. 151

152

JOURNAL OF SOCIAL PSYCHOLOGY

bias. For example, studies of American and Japanese respondents show that there is a significantly greater tendency for the Japanese to use the "undecided" category on a Likert scale ( e. g., 5). The present study examines the so-called response-set-bias issue by using the responses of a large sample of expatriate managers and their local counterparts in Britain and each of six South-East Asian nations.

Downloaded by [UNSW Library] at 16:27 23 February 2015

B.

METHOD

The total sample consisted of 1647 male managers who were respondents to a study of stereotyping undertaken in Japanese, British, and American firms in Singapore, and in Japanese firms in Britain, Hong Kong, Indonesia, Malaysia, the Philippines, and Thailand. The numbers of respondents for each national group are presented in Table 1, broken down by age and education. Each respondent received a questionnaire in his own language, asking him to provide an autostereotype (that is, to indicate his impression of the typical manager of his own nationality with whom he worked); a heterostereotype (indicating his impression of the typical manager of the other nationality with whom he worked); and a metastereotype (an assessment of how the other nationality would perceive a typical manager of the respondent's own nationality). For each stereotype, a set of 18 seven-point semantic differential items were used, with the following pairs of bipolar adjectives: unmethodical-methodical; cooperative-uncooperative; modest-immodest; shy-assertive; frank-secretive; ambitious-unambitious; serious-humorous; cautious-rash; intolerant-tolerant; logical-illogical; patient-impatient;

TABLE 1 PERCENTAGES OF RESPONDENTS. BY AGE AND EDl'CATION

Under 40

Over 40

Nationality

N

Graduate

Nongrad

Graduate

Nongrad

Japanese British American Hong Kong Singaporean

769 128 34 95 190 113 112 111 95 1647

37 23 44 13 55 42 43 52 48 39

9 2i 12 42 18 33 9 5 12 15

48 25 41 6 18 11 32 39 29 35

6 26 3 39 8 14 16 4 11 11

Malavsian

Indo~esian Filipino Thai Total

Downloaded by [UNSW Library] at 16:27 23 February 2015

B. W. STENING AND

J.

E. EVERETT

153

extroverted-introverted; inflexible-flexible; indecisive-decisive; predictableunpredictable; impolite-polite; honest-dishonest; and lazy-industrious. Thus each respondent answered 54 items (three sets of 18 items), each on a seven-point scale. The study in Singapore was, however, slightly different. The questionnaires for the 112 Japanese, 30 British, 34 American, and 190 Singaporean managers there did not ask for the metastereotype, but instead requested heterostereotypes of the two nationalities with whom the respondent did not work. Accordingly, each respondent in Singapore answered 72 items (four sets of 18 items). The present study is not concerned with the stereotypes themselves, but with the response-set bias that may be associated with nationality, age, or education. Accordingly, the responses from each respondent were combined to give a frequency distribution of responses "I" to "7" for that respondent, and the mean distributions for each category of respondents were compared. C.

RESULTS

The overall response distributions for each of the nine nationalities are given in Table 2. The nationalities have been rank ordered in terms of their tendency to use the neutral response alternative ("4"). Over 18 percent of the Japanese responses fell in this category while, in contrast, just under 10.5 percent of the Thai responses did. There are also considerable differences to be seen between the different national tendencies to use the extreme responses ("I" and "7"). The mean response for each nationality is also reported in Table 2, with 99 percent confidence limits. Assuming each group of respondents to be a sample from a much larger population, the confidence limits refer to the population mean: if the confidence-limit range for two groups do not overlap, we can conclude that their means are significantly different. For each nationality of respondent, except Hong Kong, the mean score is, at a one-percent significance level, slightly less than the midpoint value of "4," suggesting a systematic but small bias towards the left-hand side of the response scale. There are also small but significant differences in the magnitude of the bias, with the Thais having the greatest downward bias. Since the median age for each nationality of respondent is close to 40, age categories of under 40 and over 40 are used in Table 3, which gives the mean proportion of midpoint responses used by the respondents in each age and educational group. The "graduate" category represents holders of a

154

JOURNAL OF SOCIAL PSYCHOLOGY TABLE 2 PERCDITAGI-: RESPO:-;SE FOR SI-:VEN-POI:-;T SEMANTIC DIFFERENTIAL ITEMS

Responses

Downloaded by [UNSW Library] at 16:27 23 February 2015

Nationality Japanese Hong Kong British American Singaporean Malaysian Filipino Indonesian Thai Total

6.6 10.3 9.8 6.9 9.0 13.4 14.4 I i.5 1i.1 9. i

20.2 18.4 18.6 23. i 20. i 1i.4 21.8 18.2 24.2 20.1

3

4

5

6

I i.O

18.1 1i.O 16.0 15.9 14.5 14.5 12.4 11.9 10.5 15.9

14. t 12.4 15. i 14.9 13.4 13.0 10.3 9. I 8.3 13.3

1i.s 1i. i 16.9 16.6 1i. i 15.i 19.1 20.2 15.6 1i.5

13.4 16.0 16.8 16.5 15.3 12.1 11.6 1l.i 15.5

Mean score 6.1 10. i 6.9 5.2 8.3 10. i 10.0 11.5 12.6 8.0

3.91 3.99 3.88 3.i9 3.89 3.86 3.i9 3.83 3.66 3.8i

± ± ± ± ± ± ± ± ± ±

.02 .08 .05 .06 .04 .Oi .06 .08 .10 .02

No/I': Nationalities are ordered by neutral response tendency (response 4).

degree or diploma. The 99 percent confidence limits are again given for each group mean, and again provide a convenient means of indicating whether the results for any two groups are significantly different: if any two groups have nonoverlapping confidence limits, we can conclude that they have been sampled from populations having significantly different means. No results are reported for those groups where there were less than 10 respondents. The major effect is due to the nationality of the respondent, and there is no effect of age or education consistent across all nine nationalities. For the Japanese, all four age and educational groups use the midpoint response very similarly. However, for each nationality of the ASEAN group (The Association of South-East Asian Nations, comprising Malaysia, Singapore, the Philippines, Indonesia, and Thailand) there is a marked tendency for TABLE 3 PERCE:-;TAGE USAGI-: OF MIDPOII'T RESPONSI-:S

Under 40 Nationality

All groups

Japanese Hong Kong British American Singaporean Malaysian Filipino Indonesian Thai

18 ± Ii ± 16 ± 16 ± 14 ± 14± 12 ± 12 ± 10 ±

1 4 2 4 2 3

.J 3 .l

- - - - - - - - -

Graduate 18 23 19 15 15 1i 16 12 12

2

11 5 6 3 4 4

Oyer 40

---------

Nongrad

Graduate

Nongrad

1i ± 4 1i ± 5 18 ± 3

18 ± 2

19 ± 5 15 ± i l.l ± 3

14 ± 5 16 ± 5 13 ± 10 8 ± i

------

IS 15 16 10 9

± ± ± ± ± IS ± 11 ±

4 5 5 6 4 6 5

J.l ± 6

6 ± 6 5 ± 4 i ± 5

------------

155

Downloaded by [UNSW Library] at 16:27 23 February 2015

B. W. STENING AND J. E. EVERETT

the midpoint response to be avoided by older nongraduates. Indeed, this tendency accounts for much of the overall lesser use of the midpoint response by the ASEAN nationalities, although for Thais the younger nongraduates also appear to avoid the midpoint response. The highest usage of the midpoint response (23 percent) is shown by the younger Hong Kong graduates. Finally, Table 4 shows the percentage usage of the most extreme responses ("1" and "7"). For each group, the mean usage of the extreme responses, and its 99 percent confidence limits, are reported. Overall, as might be expected, it appears that high usage of the midpoint response tends to go with a low usage of the extreme responses, but this rule is not entirely valid. In particular, the Hong Kong respondents show a high usage of both the midpoint and the extreme responses. As before, there is very little difference between the Japanese age and educational groups, while the younger Hong Kong graduates stand out for their low use of the extreme responses. A very high use of the extreme responses is exhibited by older nongraduates from Malaysia and Indonesia. D.

DISCUSSION

The results indicate that the response sets of the groups were markedly different. At the most aggregate level there were a number of significant (and meaningful) differences between the nationalities in terms of their tendency to use both the midpoint and the extreme responses. Further, it has been shown that both education and age have an important influence upon the response styles of several nationalities, though, interestingly, not upon the Japanese. Excluding the Japanese and allowing for certain variaTABLE 4 PERCENTAGE USAGE OF EXTREME RESPONSES Under 40

Over 40

Nationality

All groups

Graduate

Nongrad

Graduate

Non grad

Japanese Hong Kong British American Singaporean Malaysian Filipino Indonesian Thai

13 ± 21 ± 17 ± 12 ± 17 ± 24 ± 24 ± 29 ± 30 ±

14 12 14 12 17 16 21 23 28

± ± ± ± ± ±

12 ± 4 19 ± 9 18 ± 8

12 ± 2

10 ± 4 26 ± 10 22 ± 9

1 6 4 5 4 6 6 7 6

3 8 7 7 4 6 ± 5 ± 11 ± 8

18 ± 9 23 ± 12 32 ± 23 28 ± 19

12 12 17 21 31 29 32

± 5 ± 7

± ± ± ± ±

8 11 12 12 11

18 ± 16 51 ± 25 43 ± 20 34 ± 19

Downloaded by [UNSW Library] at 16:27 23 February 2015

156

JOURNAL OF SOCIAL PSYCHOLOGY

tions for particular nationalities, the tendency of a respondent to use the midpoint (and, conversely, not to use the extremes) is greater if that person was either a graduate or young. This effect is particularly evident in cases where the respondent is from one of the ASEAN countries. It is not possible, nor indeed desirable, to lay down a set of decision rules for questionnaire design on the basis of the response sets revealed in this study. Ultimately each researcher must decide in the context of his or her own study whether biases revealed in the response sets are meaningful. Opinions will undoubtedly vary as to what action, if any, of the variety available (e.g., 4, pp. 53-54) should be taken to correct for such biases. Indeed, recent evidence (1) that response set is significantly related to the language in which questions are posed. a finding emanating from a study of bilingual managers, suggests that the response-set issue must be assessed in the context of a much wider range of considerations than has traditionally been the case. Nevertheless, the present study demonstrates the importance of maintaining a vigilance for response sets of such type and magnitude that they could provide a plausible rival hypothesis. REFERENCES 1.

2. 3. 4. 5.

BENNETT, M. Response characteristics of bilingual managers to organizational questionnaires. Person. Psychol., 1977, 30, 29-36. BERRY,]. W. Introduction to Methodology. In H. C. Triandis & J. W. Berry (Eds.), Handbook of Cross-Cultural Psychology, Vol. 1/: Methodology, Boston: Allyn & Bacon, 1980. Pp. 1-28. KERLINGER, F. W. Foundations of Behavioral Research (2nd cd.). New York: Holt, Rinehart & Winston, 1973. TRIANDIS, H. C. The Analysis of Subjective Culture. New York: Wiley, 1972. ZA.X, M., & TAKAHASHI, S. Cultural influences on response style: Comparisons of Japanese and American college students. J. Soc. Psychol., 1967, 71, 3-10.

Department of Management The University of Western Australia Nedlands, Western Australia 6009