RESEARCH AND PRACTICE
IN HUMAN RESOURCE MANAGEMENT

Highlight, copy & paste to cite:

Nibler, R., & Harris, K. L., (1994). A Comparison of Group Consensus Decision Making: Chinese and American Cultures, Research and Practice in Human Resource Management, 2(1), 35-45.

A Comparison of Group Consensus Decision Making: Chinese and American Cultures

Roger Nibler & Karen L. Harris

Abstract

Chinese and American student groups were examined for differences in their performance on the “Lost at Sea” group exercise. In this exercise, groups which feel free to express their ideas openly, welcome opposing viewpoints and are not inhibited in the decision making process typically perform better than do individual members of the group. The exercise was performed individually, and then in groups of five using consensus decision making. Results showed that American groups outperformed their individual members more often than did Chinese groups. These findings support the prediction that, because American subjects belong to a more individualistic culture, they would have less tendency to conform and therefore would be more inclined to provide controversial information and ideas to their group. Although the Chinese subjects were better individual performers, there was no difference in group scores between the Chinese and American subjects.

Introduction

The outcomes of studies pertaining to the superiority of group versus individual decision making have led to checkered results. Some researchers (Davis, 1982; See Davis, 1992 for an historical summary) have concluded that task-oriented groups are better able to catch errors, creatively stimulate their members, and offset biases as opposed to individuals working alone. On the other hand, there are numerous studies, especially those stemming from the classic research by Lorge and Solomon (1955), that cast significant doubt as to the general superiority of group decision making (Davis, 1969; Hastie & Sakes, 1986).

The mixed results in this research area suggest that perhaps a multitude of intervening variables are affecting the outcomes of these studies, many of which were conducted in carefully controlled laboratory settings. Analysts have suggested that various interacting variables, including group norms, group size, individual tendencies toward social loafing, and social culture, can influence effective decision making (Levine & Moreland, 1990; McGrath, 1984). The effects of these variables, many of which may operate simultaneously, most likely contributed to the confusing and controversial outcomes of these studies.

A factor of particular interest in group decision making is conformity, which is defined as some behavioural or attitudinal change that occurs as the result of a real or imagined group pressure (Keisler & Kielser, 1969). Classical studies by such authors as Asch (1955) and Festinger (1954) have provided longstanding testimony to the forceful effect of group pressure and its influence on an individual(s) conforming to the will of the group.

The concept of conformity is a complex area as well as a confused one, and there has been some question as to what it means to conform. The literature in this area divides conformity into categories of (1) an enduring personality characteristic; (2) attitudinal change as a result of real or imagined group pressure; and (3) compliance, or simply as going along with the will of the group (Mills, 1969). Although numerous studies have been conducted on conformity (compliance), the outcomes are generally focused on group pressure and its effect on individuals supplanting their own judgements for the will of the group (Mills, 1969; Krech, Crutchfield & Ballachey, 1962). In a similar manner, the current study utilised only behavioural data to illustrate the power that groups may possess over the individual. Thus, in this study, conformity will be the generic term used in the context of compliance, or outward behavioural change.

Conformity, however, operates in the broader context of the cultural environment. The rather extensive studies by Triandis, Bontempo & Villareal (1988) ranked several countries along a dimension of individualism and collectivism. In individualistic cultures personal values, self-reliance, and achievement goals strongly influence people’s behaviour. In collectivist cultures people’s behaviour is more affected by attitudes, goals and values that are s1ured collectively with particular groups which the person either belongs to or identifies with (Triandis, 1988). The Triandis et a!, (2988) study cited northern and western Europe and North American countries as those whose people tended to he individualistic. Collectivistic tendencies, on the other hand, have been found in Italy, Greece and Hong Kong. The ranking of these countries basically supported similar findings by Hofstede (1980).

The concept of collectivism is in many ways similar to conformity arid may help to account for behavioural differences among cultures. Basically, the greater the degree to which a culture or individual is collectivist, the greater will be the tendency for people in that culture to be conformists (Hui & Triandis, 1986). Conformity in any culture will depend upon a person’s vulnerability to a particular in-group with which they identify. People in individualistic cultures have a wider range of in-groups and tend to drift more easily from one in-group to another. In collectivist cultures, on the other hand, people tend to have a limited number of in-groups, often consisting of family, co-workers, and friends, and their conformity in these groups can be quite intense (Triandis et al, 1988, 1990). However, individuals in a collectivist culture may be rather nonconforming to groups which are not considered as their in-group (Hui, 2988).

Collectivist societies differ on how the collectivism is manifested. For example, in Far East cultures the emphasis is on in-group harmony. One must present oneself to others as modestly as possible and avoid conflict with others at all costs because confrontation is highly undesirable. When collectivists are presented with a task, it is not unusual that the successful completion of that task is subordinate to the maintenance of harmonious relations within the in-group (Triandis et a!, 1988).

Purpose

The purpose of the current study was to test the effects of consensus decision making on Chinese and American college students. Although it has not been documented in studies, it is a rather common notion that Chinese are more conforming than Americans (Hsu, 1953; Chu, 1979; Domino and Hannah, 1987). With respect to previous research, however, only one empirical study was uncovered to substantiate this belief (Meade & Barnard, 1973). In fact, another study conducted at approximately the same time (Whittaker & Meade, 1967) reported evidence which did not favour this popular notion; they found no differences between Chinese and American subjects in their conformity levels.

Recognising that groups are culturally embedded, some of the variability among groups may be the result of cultural differences (Levine & Moreland, 1990). The current study implemented a decision making task requiring groups of Chinese and American subjects to reach a consensus. The general tenet being tested was that the Chinese subjects would inhibit their contribution to the group to a higher degree than the American subjects. This inhibition would stem from the collectivist cultural environment of the Giinese students and their resulting tendency to be more inclined toward conformity. Implicit in this hypothesis, however, was the assumption that the fellow students comprising the Chinese groups would be considered by most of the members as being in-group colleagues. In the case of the American students, who belong to a highly individualistic culture, the assumption made was that the task group was but one of numerous groups to which they belong, and therefore was likely to have a lower influence on students’ vulnerability to group pressure. As a result, the American students were expected to be more likely to express their ideas openly, to feel free to express opposing viewpoints, and to be less inhibited in the group decision making process than the Chinese subjects.

The result of the higher inhibition on the part of the Chinese students would be an impairment on their group’s effectiveness. The extent of this impairment would be manifested in the numbers of group members who individually perform their task in a manner superior to that of their group (Nemiroff & Ffeiffer, 1975). The basic hypothesis, then, was that the number of individuals in the Chinese groups who outperformed their group would be greater than the number of American individuals who outperformed their groups.

Method

Subjects

The Chinese subjects (N = 210) were students attending a government supported university in Hong Kong and were enrolled in a course consisting of a two hour mass lecture. The American students (N = 180) were students attending a similar sized state university in Illinois. The exercise was administered to all the subjects under nearly identical conditions. Subjects who had participated in a similar exercise or were not reared in the culture in which the exercise was administered were excluded.

Task

The exercise was the well-known “Lost At Sea” group consensus exercise as presented in The 1975 Annual Handbook for Group Facilitators (Nemiroff & Pfeiffer, 1975). In this exercise the subjects were presented with a situation in which they were part of a group of survivors on a sinking yacht. There were 15 survival items such as a shaving mirror, maps, radio, etc. which had to be ranked according to their highest priority as a survival or rescue item to be taken onto the lifeboat.

Procedure

In the first phase of the experiment, all subjects ranked the items on an individual basis, and were not allowed to seek help from anyone or examine their results. This phase lasted for about 20 minutes.

In the second phase of the experiment, groups consisting of five members were formed on a mutual attraction basis. That is, subjects were allowed to form their own five-member groups voluntarily, with minimal guidance by the experimenter. The groups were instructed to arrive at a group ranking using only group consensus. The groups were instructed not to use voting or any other structured methodology. The group consensus method was used because it represents the purest form of group interaction, and if systematic differences were to take place as a result of conformity and group pressure, this decision making method would be most sensitive in detecting it. In addition, group consensus is typical of many group decision making practices in the business world.

Although the length of the scheduled experimental period did place a potential upper time limit, all groups were finished within the allotted period. A time constraint was not imposed in order to discourage groups from using structured techniques such as voting.

The exercise was scored by use of the model rankings of the items reported by experts (officers in the U.S. Merchant Marine) and by comparing these rankings against each individual’s ranking as well as the group rankings. The absolute difference between the model ranking for each item and the individual/group ranking was computed for each survival item. For example, the shaving mirror was ranked as number I priority among all the items by the experts’ reports. If an individual or group ranked this item as a number 7, their score for this item would be 6. The score for the individual or group was then calculated by adding these absolute differences for each of the 15 items. The lower the score, the closer the individual or group approached the experts’ model answer. Thus, a low score was deemed to be more accurate, or correct, than a high score.

Results

The data are summarised in Table 1. The individual averages for each of the Chinese and American groups are reported arid arranged according to the average of the most correct person in the group (lowest score, or #1 in the table), the average of the second most correct person in the group (second lowest score, or #2 in the table), ranging down to the least correct person in the group (highest score, or #5 in the table). For example, the average score of the top American performers across all American groups was M 56.2. One of the far right-hand columns in Table 1 includes the “average group” score, which is the average score over all American groups (or all Chinese groups), recorded during the group consensus phase of the task.

Table 1
Chinese and American Individual and Group Averages
American Individual Averages Ave Group Net Gain
#1 #2 #3 #4 #5
Mean 56.2 68.1 74.6 79.8 85.9 62.5 10.6
Std Dev 13.5 7.9 5.6 5.1 6.3 11.9 9.9
Chinese Individual Averages
#1 #2 #3 #4 #5
Mean 48.6 57.7 67.5 72.7 80.2 60.7 4.2
Std Dev 12.3 11.1 9.5 9.0 9.1 14.4 10.6

Table 1 also includes a net gain column. A gain score was used as a rough measure of group effectiveness, and was calculated as follows: An average was computed over all individual scores within a given group, and the group score was then subtracted from that average. Gain scores were accumulated over all groups and reported as a net gain score separately for the American and Chinese samples. A negative gain score indicated that the group effort was inferior to the composite of the individual efforts, while a positive score indicated that the group effort was superior to the composite individual scores.

First, the two cultures were compared on their group scores (ie, the score recorded during the group consensus phase of the task), M 62.5 for American groups versus M = 60.7 for Chinese groups. Results of a t-test showed no significant difference in average group scores between the American and Chinese subjects. Indeed, the difference of only 1.8 is indicative of nearly identical outcomes.

In examining the individual scores listed in Table 1, however, a different picture emerged. Using one way analysis of variance to test for differences among all ten individual scores (American and Chinese), significant differences among means were indicated, F (9,350) = 59.2, p = .001. Post Hoc analysis using Scheffe’s test to compare the best American individual score against the best Chinese individual score, followed by the second best American individual score against the second best Chinese individual score, and so forth, were all significant at the p .05 level or less for all five categories. Scheffe’s test conducted on all Chinese individuals against all American individuals was also significant at the same level. Overall, it appears that the Chinese subjects were better individual performers.

This result suggests that although the group outputs were the same, on an individual basis the American subjects did not appear to be as knowledgeable about survival at sea as were the Chinese subjects. This reason is based on the fact that the American students mostly came from areas located over 1,000 miles from the nearest ocean. The Chinese subjects, on the other hand, live in a thriving international sea port and accounts of sea disasters in the South China Sea occur quite frequently as a result of typhoons, pirates and other catastrophies.

Examining the individual scores (from Phase #1) against the group scores (from Phase #2) provides some possible insights into differences taking place in the group processes between the two sets of subjects.

For example, a comparison of the American net gain score and the Chinese net gain score was performed. The gain score, it may be recalled, is a measure of the difference between the (average) individual scores and the group score. The higher this value, the more a group is benefiting from the collective wisdom of its group members. The higher American net gain score of 10.6 against the 4.2 net gain score of the Chinese subjects is indicative of a higher overall level of synergism on the part of the American groups, with the difference in gain scores being significant, t(76)2.73,p= .01.

The data was then examined by observing the number of group members whose individual scores were superior to their group scores. Three such indices were formed, and each of them revealed that, compared to the Americans, many Chinese individuals outperformed their groups. Yet those superior individual performances did not appear to benefit the groups, in many cases.

The first set of analyses was conducted by counting the number of groups in which three or more individuals outperformed the group. We reasoned that a group in which the majority, three out of five individuals, did relatively well on the task yet were unable or unwilling to convince the group as a whole may be appropriately labelled a highly conforming group. A contingency table was formed. This showed 13 groups (31.0%) of Chinese subjects where three or more individuals were able to outperform the group and another 29 groups where this did not occur. Among the American subjects, only 3 groups (8.3%) had three or more individuals outperforming the group, and 33 groups in which this did not occur. A chi-square test indicated significance for this set of frequencies, χ2 (1, N = 78) = 6.09, p < .05.

A second set of similar analyses compared the total number of Chinese subjects who outperformed their group (77 out of a total of 210 Chinese subjects, or 37%) against the total number of American subjects who outperformed their group (40 out a total of 180 American subjects, or 22%). That comparison was also found to be significant, X2 (1, N = 390) 9.63. p = .01. A third index examined those groups in which the group score was superior to every member of the group. With an index of this nature, highly individual groups are likely to benefit from the uninhibited contributions among members. The American subjects contained 12 such groups (36.0%) and the Chinese subjects had 8 groups (19%). Although these numbers are in the predicted direction, the results were not significant using chi-square analysis.

Discussion

The results of this study help to confirm the hypothesis that there would be a greater number of individual Chinese subjects outperforming their groups. To this extent it is apparent that the Chinese subjects were more inclined to either withhold information and/or not present their ideas in a mariner which was sufficient to sway the group’s opinion and to reach an optimal task solution.

Although the groups were nearly identical with respect to age and educational background, there was a significantly higher percentage of females among the American subjects (65% vs 53%). However, studies pertaining to differences in conformity between male and female subjects tend to be non-conclusive, and those studies which have found such differences (Asch, 1956; Krech, et al, 1962) found females to be more conforming than males. If females are more conforming, then most likely it would have worked against the hypothesis.

This study may perhaps open similar avenues for future studies in cross-cultural research. One avenue is the use of the “Lost at Sea” task (or exercises of a similar type) as an instrument to be used in studying cross-cultural group decision-making processes. For example, numerous discussions with Chinese students after the study indicated that the exercise was not as important to them as the acceptance of the group members. The reason frequently stated was their concern that, although the exercise would be soon forgotten by the rest of the group members, being congenial or trying 4’too” hard to get one’s way during the group consensus phase might be remembered by the other group members for a long time. American subjects, on the other hand, gave indications which were opposite and felt they would be remembered primarily for their contribution toward solving the problem, whereas the method they used to convey their ideas might be quickly forgotten by the other group members. Thus, future research efforts might focus on the particular concerns held by the group members regarding perceived future consequences resulting from their behaviour in the group.

The current study departs from past work of a similar nature with regard to its methodology. Perusal of the past literature uncovered only two cross-cultural experimental studies comparing American subjects to Hong Kong Chinese in conformity levels (Whittaker & Meade, 1967; Meade & Barnard, 1973). Both of these past studies made use of the classic Asch (1956) paradigm or some variation of it, in which subjects were asked to announce publicly their personal judgements in the presence of a majority holding a counter-opinion. The results of these past studies were inconclusive; one study found virtually no difference between the cultures in conformity levels (Whittaker & Meade, 1967), while the other found that Chinese subjects’ opinions were more closely guided by the group than were those of American subjects (Meade & Barnard, 1973). The current study utilised the “Lost at Sea” task, which is different from the Asch paradigm in that it allows subjects to interact during the group consensus stage. This opportunity to interact enabled the current study to capture the effects of “informational social influence” better (Deutsch & Gerard, 1955), a form of conformity in which attitudes are changed via the exchange of convincing information among group members. This form of social influence goes beyond mere normative influence and group pressure to win over a minority, and represents a more fundamental and long-lasting form of conformity.

That Chinese tend to be more conforming than Americans should certainly not come as a surprise to many readers. After all, this assumption has formed the foundation for much cross-cultural research being conducted today. Although one might argue that Hong Kong Chinese are considered to be “Westernised,” the current study provides no evidence of this notion. On the contrary, this study has provided empirical documentation of the long-held assumption that the Chinese culture is, in general, more conforming than the American culture. The basic Chinese culture appears to be alive and well, even in Hong Kong. Access to relatively large and comparable groups of Chinese and American students has added to confidence in the results of this study. Perhaps the use of this type of instrument, the “Lost at Sea” task, coupled with studies focusing on the differences in group processes, can facilitate a better understanding between these two cultures.

References

Asch, S.E., (1955), Opinions and social pressure, Scientific American, 193, pp. 31-35.

Asch, S.E., (1956), Studies of independence and conformity: A minority of one against a unanimous majority, Psychological Monographs, 70, 9 (Whole No. 416).

Chu, L., (1979), The sensitivity of Chinese and American children to social influences, The Journal of Social Psychology, 109, pp. 175-186.

Davis, J.H., (1969), Group Performance, Addison-Wesley, Reading, Massachusetts.

Davis, J.H., (1982), Social interaction as a combinatorial process in group decision, in H. Branstatter, J.H. Davis, & C. Stocker-Kreichgauer (Eds.), Group decision making, Academic Press, London.

Davis, J.H., (1992), Some compelling intuitions about group consensus decisions, theoretical and empirical research, and interpersonal aggregation phenomena: Selected examples, 1950—1990, Organizational Behavior and Human Decision Processes, 52, pp. 3-38.

Deutsch, M. & Gerard, H., (1955), A study of normative and informational social influences upon individual judgment. Journal of Abnormal Social Psychology, 51, pp. 629-636.

Domino, G. & Hannah, M.T., (1987), A comparative analysis of social values of Chinese and American children. The Journal of Cross-Cultural Psychology, 18, pp. 58-77.

Festinger, L., (1954), A theory of social comparison processes. Human Relations, 7, pp. 117-140.

Hastie, R. & Sakes, M., (1986), Social Psychology in Court, Van Nostrand Reinhold, New York.

Hofstede, G., (1980), Culture’s Consequences: International Differences in Work-Related Values, Sage, Beverley Hills, California.

Hsu, F.L.K., (1953), Americans and Chinese: Two ways of life, Abelard Schuman, New York.

Hui, C.H., (1988), Measurement of individualism-collectivism, Journal of Research in Personality, 22, pp. 17-36.

Hui, C.H. & Triandis, H.C., (1986), Individualism-collectivism: A study of cross-cultural researchers, Journal of Cross-Cultural Psychology, 17, pp. 225-248.

Kiesler, C.A. & Kiesler, S.B., (1969), Conformity, Addison-Wesley, Reading, Massachusetts.

Krech, D., Crutchfield, R.S. & Ballachey, E.L., (1962), Individuals in Society, Macmillan, New York.

Levine, J.M. & Moreland, R.L., (1990), Progress in small group research, in M.R. Roserizweig & L.W. Porter (Eds.), Annual Review of Psychology, Annual Reviews, Palo Alto, California.

Lorge, I. & Solomon, H., (1955), Two models of group behavior in the solution of eureka-type problems. Psychometrika, 20, pp. 139-248.

McGrath, J.E., (1984), Groups: interaction and Performance, Prentice Hall, Englewood Cliffs, New Jersey.

Meade, R.D. & Barnard, W., (1973), Conformity and anticonformity among American and Chinese, The Journal of Social Psychology, 89, pp. 15-24.

Mills, J., (1969), Experimental Social Psychology, Macmillan, Toronto.

Nemiroff, P.M. & Pfeiffer, J.W., (1975), Lost at sea: A consensus-seeking task, in J.E. Jones & J.W. Pfeiffer (Eds.), The 1975 Annual Handbook for Group Facilitators, University Associates Publishers, La Jolla, California.

Triandis, H.C., (1988), Collectivism vs. individualism: A reconceptualization of a basic concept in cross-cultural social psychology, in G. Verma & C. Bagley (Eds.), Cross-cultural Studies of Personality, Attitudes and Cognition, Macmillan, London.

Triandis, H.C., Bontempo, R. & Villareal, M.J., (1988), Individualism and collectivism: Cross-cultural perspectives on self-in group relationships. Journal of Personality and Social Psychology, 54, pp. 323-338.

Triandis, H.C., McCusker, C. & Hui, H.C., (1990), Multimethod probes of individualism and collectivism, Journal of Personality and Social Psychology, 59, pp. 1006-1020.

Whittaker J. & Meade, R., (1967), Social pressure in the modification and distortion of judgment: A cross-cultural study. International Journal of Psychology, 2, pp. 109-113.