Simonas Audickas
Vytautas Magnus University, Lithuania
Charles Davis
Edgewood College, USA
Magda Szczepańska
University of Lodz, Poland
Abstract
Drawing on Hofstede’s theory and research, this study examined the hypothesis that differences in cultural dimensions of individualism/collectivism and homogeneity/heterogeneity will influence the level of task performance and socialization.
Swedes (individualists) and Greeks (collectivists) were assigned to two separate groups. In the first phase of the experiment the two groups were homogenous – consisting of 3 participants from each of the cultures (3 Swedes, 3 Greeks). Their task was to put together a 54 piece puzzle. Both homogenous groups were timed and observed separately. The next step was to bring in two heterogeneous (ethnically mixed) groups. For this purpose the size of each homogeneous group was increased to 6 participants by assigning one American, Canadian and German representative. Their task was to put together a larger, 300 piece puzzle within a certain time frame of 25 minutes. Individual and group responses were measured using a fine-grained checklist. The hypotheses were: 1) Homogenous individualistic group would finish the task faster than the homogenous collectivistic group. 2) Heterogeneous individualistic group will finish the task faster than heterogeneous collectivistic group. 3) Collectivists will be more socially interactive than individualists. In this research hypotheses 1 and 3 were supported, whereas hypothesis 2 was not. The implications for future research are that individualism/collectivism and homogeneity/heterogeneity of a group may have an influence on group dynamics as it relates to task performance.
Introduction
According to Hofstede’s (1980) monumental study of more than 60,000 IBM workers, consisting of various ethnicities, he extracted four cultural dimensions: power distance, uncertainty avoidance, individualism/collectivism and masculinity. Many people have done extensive research and written several articles about these four cultural dimensions, which have paved the way for culturally sensitive approaches in various fields including business, education, and health care. Moreover, through understanding the implications of these cultural dimensions, many organizations (domestic and international) have become more effective in communication, productivity, and competitiveness.
The primary cultural dimension that we directed our attention towards was that of individualism/collectivism. We chose this cultural dimension in light of their differences of task performance and socialization behaviours. In a general sense, individualism refers to a culture’s tendency to be competitive and focused more on “individual success” versus group success. As it relates to task performance and socialization behaviours, individualists are considered to be task-oriented, thus primarily focused on completing the task with a competitive edge. Socialization behaviours for individualists are exemplified in their emphasis on identifying distinctions and complementary skills, to maintain autonomy and independence in group-based working relationships that they view as temporary and self-serving (Lincoln, Hanada, and McBride, 1986). In other words, it can be deductively reasoned that individualists are less socially interactive in a group setting as compared to collectivists.
Conversely, collectivism is generally group orientated, particularly in-group orientated. Moreover, collectivists are known to be self-sacrificing toward in–group members and thus more focused on group success versus individual success. Another fundamental aspect of collectivism is the extent to which they value interdependence and affiliation in long-term relationships (Breer and Locke, 1965). According to Triandis, “People in collectivist cultures pay more attention to context (emotional expressions, touching, distance between bodies, body orientation, level of voice, and eye-contact). We used this cultural profile to speculate that collectivists are more prone to socialization behaviours (e.g. more socializing, looking at each other, and laughing), which we hypothesized to render collectivists to be less task oriented. To put this in perspective, the research undertaken should be focused primarily on three theoretical dimensions: Homogeneity vs. Heterogeneity; High-Low Contact; and High- Low Context.
The role that homogeneity versus heterogeneity plays in group socialization and task performance are of key importance to our research. According to John J. Sosik and Dong I. Jung, their conception of “functional heterogeneity” is as follows: “…represents group members’ perceptions about the diversity of various functional experiences and skills of other group members.” Those perceptions are important given the increased diversity (Cox et al., 1991) and delegation of decision-making tasks (Campion, Medsker, and Higgs, 1993) in organizations, because a wide range of skills, competencies, and knowledge are required for such initiatives.
It has been argued that functional heterogeneity among individualists is more exaggerated than collectivists. Individualists see such perceived diversity in group members as a way of bringing to the group unique qualities and multiple perspectives on problem solving (Eisenhardt and Tabrizi, 1995). Collectivists, on the other hand, emphasize shared values, similarities, and commonness among group members; hence, it can be deduced that they prefer homogeneous groups than heterogeneous groups. The results of John J. Sosik and Dong I. Jung’s research confirmed that there were higher levels of functional heterogeneity among individualists than collectivists. It was also concluded that individualists encouraged group members to focus on the task rather than on social and interpersonal relations. We used this research to hypothesize that collectivists will perform to a lesser degree than the individualists, especially in a heterogeneous group, consisting of 3 Greeks defined as collectivists (Hofstede, 1980) and 3 persons from cultures defined as individualistic (German, American, Canadian).
Furthermore, we hypothesized that because the collectivists are more socially oriented it will take them longer to finish the task. High-low contact is another cultural variable that is a distinguishing factor among individualists and collectivists, especially as it relates to geographic location. High contact cultures are usually located in countries that are closer to the equator and have warmer climates (Triandis, 1994). As was aluded to earlier, high contact cultures, i.e. collectivists, have a general tendency to be more expressive of emotions, touching, and close social distance (Hall, 1996). These aspects of high contact cultures coincide with collectivist’s tendency to prefer homogenous (in-group) work and high levels of socialization. Conversely, low-contact cultures, i.e. individualists, are generally located in cooler climates and exemplify several opposing behaviours of collectivists: less emotionally expressive, more social distance, and generally more socially interactive. As a result of these cultural variables for low-contact cultures, we concluded that individualists would be more task oriented and less social; thus rendering them more effective in the respective task.
High-low context, in many ways, is closely related to high-low contact. The only difference between them is that high-low context relates to the message that is being conveyed through verbalization or non-verbalization (paralinguistics and body language). High-context cultures are ones in which the meaning of a message is understood implicitly (e.g. generally misinterpreted by people outside of the in-group). Low-context cultures differ in that the message is conveyed explicitly; hence the meaning is much clearer.
In conclusion, we derived three main hypotheses that we believe can be directly related to the three cultural variables (homogeneity/heterogeneity, high-low contact, and high-low context) to a significant degree: 1) the homogenous Swedish group will complete the task faster because they are an individualistic culture (i.e., task oriented); whereas the homogenous Greek group will take more time to finish the task in that they are collectivistic (i.e., less task oriented and more social). 2) the heterogeneous individualistic group will finish the task faster than heterogeneous collectivistic group 3) the heterogeneous Greek group will be more socially interactive versus the Swedish group, which will be less socially interactive.
Method
Participants
Our experiment consisted of 12 students of various ethnicities at Växjö University.
There were 3 Swedes, 3 Greeks, 2 Americans, 2 Canadians and 2 Germans. The mean age was 23.7 (SD=5.99). The mean gender demographic was 67% male and 33% female. There were 9 persons from individualistic cultures and 3 persons from collectivistic cultures (Hofstede, 1980). Each participant volunteered and was randomly assigned to two groups. Each group consisted of 6 participants.
Materials/Apparatus
The two main instruments for our experiment were two jigsaw puzzles (54 and 300 pieces, respectively) and a video recorder, which we used to record each session. The puzzles were located in the middle of the table and were not accessible until we began timing. The video recorder was situated in the corner of the room to make it possible to view each participant. We also made observations from an adjacent room with a two-way mirror to ensure that we would interpret the recordings accurately.
Design
The independent variables were the culture (individualistic or collectivistic) and the type of group (homogenous vs. heterogeneous), whereas the dependent variables were the behaviours (social referencing, touching, looking, silence, smiling, and content of verbalization)*. Social referencing= initiating, responding/initiating, and response; touching= touching the puzzle and passing the pieces among the group; silence= absence of verbal socialization; smiling= smiling or laughing; content of verbalization= task verbalization or social verbalization. Our checklist measured the frequency and duration of these various social behaviours. For instance, touching (as defined below) is measured through how many a time a participant comes into contact with a piece of puzzle. Conversely, silence is measured in regards to duration (e.g. how long they were silent). There were two sets of groups: homogenous and heterogeneous. Homogeneity refers to person(s) from the same culture compared to heterogeneity, referring to person(s) from different cultures. For the two homogenous groups there were 3 Swedes and 3 Greeks. Each were separately given the task of completing a 54 piece puzzle in 5 minutes, specifically used as a baseline to measure the difference in task performance. The two heterogeneous groups were comprised of 3 Swedes and 3 non-Swedes (1 German, 1 Canadian, and 1 American). The other heterogeneous group consisted of 3 Greeks and 3 non-Greeks (1German, 1 Canadian, and 1 American). It is important to note that the Swedish heterogeneous group, at least theoretically, is heterogeneous than Greek. According to Hofstede, Swedes, Canadians, Germans, and Americans are all individualistic cultures, however, it’s important to mention, that individualism and collectivism is a continuous dimension and its level in every country varies. Consequently, we felt it necessary to define homogeneity and heterogeneity in these terms. Each heterogeneous group was given 25 minutes to complete as much of the puzzle as possible (neither group finished). We measured the difference by counting how many of the pieces of the puzzle were connected. The checklist was created to measure specific task and social behaviours. Using this checklist, we began by taking 2-minute segments as a reliability test to measure whether we were observing a corresponding phenomenon. The agreement among our group was over 90%, at which point we decided to observe three six minute segments of recording for each heterogeneous group. For another reliability test, we observed one six-minute segment that had already been observed by one of our group members.
Procedure
The instructions given to each participant was generally the same for both the homogenous and heterogeneous group. The only instruction that differed between the groups were as follows: the homogenous group were allowed to speak in their native language; the heterogeneous group were asked to speak only in English. After entering the room each participant was asked to not begin touching the puzzle until timing begins and to stay behind the table, so that we could record and observe them. We also told them how much time they were being given to accomplish the task: 5 minutes for homogenous group and 25 minutes for heterogeneous group. The time given depended on the complexity of the task. When the time was complete we entered the room and told them to leave all the pieces of the puzzle in their exact location.
Results
According to our first hypothesis, we conjectured that the homogenous Swedish group would finish the task faster than the homogenous Greek group. The results show that there was a difference in task performance: Swedes (54 puzzle in 4,58 min), Greeks (54 puzzle in 5,36 min), which results in a 38 second difference. We used this comparison as the baseline for our research in regards to individualism- collectivism task performance.
To test for the differences and similarities between the heterogeneous groups we performed Pearson’s correlation (table 1) to see whether there is relationship between the variables that are significant for our study, and to test significant cultural differences between Swedes and Greeks, we performed the Independent Sample T-test (table 2).
Table 1: Pearson's correlation
* Correlation is significant at the 0.05 level (2-tailed).
** Correlation is significant at the 0.01 level (2-tailed).
We have chosen to state the correlation between two types of verbalization and other variables. According to our hypotheses, we speculated that social verbalization would detract from participants’ task performance. As our results have shown there was a general phenomenon of social verbalization among all participants (N=12).
In the table of correlations we concluded that people look at each other when they are verbalizing about the task r=.645, p<0,05. That means that when they talked about the puzzle they were looking at each other. Social verbalization is positively correlated to: conversation initiation r=.747, p<0,01; responding/initiating the conversation r=.694, p<0,05; looking at people r=.587, p<0,05; smiling/laughing r=.847, p<0,01; and negatively correlated to silence duration r=-.906, p<0,05. These results lend themselves to the possibility that depending on the task and period of time appropriated, that there will not be a significant difference between collectivists and individualists.
We should mention that we did not find any significant differences between the non-groups (non-Swedes and non-Greeks). They were tested on the same variables but did not show statistically significant differences and in this part we are not going to mention their results.
Table 2: Show significant differences between Swedes and Greeks.
Significant differences between Swedes and Greeks are represented with a star (*).
Each respective variable differs at various levels of significance. We say that a small number of degrees of freedom led us to consider the significant level of 0,1. In respect to what we found about individuals working in heterogeneous groups, Greek people are more socially verbal in comparison to Swedes t(4)= -2,186, p<0,1; initiate the conversation more t(4)= -3,618, p<0,05; look at people more t(4)=-2,402, p<0,05; more frequent smile/laugh t(4)= -7,250, p<0,01; and as they socially verbalize more it is logical that they are less silent t(4)= 3,529, p<0,05.
As we further hypothesized, the collectivist group (Greek) would be more socially verbal than the individualist group (Swedes). We anticipated that the collectivists’ extent of social verbalization would hamper their task performance. The final results contradicted what we expected to find, i.e. less task performance for collectivist’s due to socialization. Instead we found that although the collectivist group was more socially interactive this it did not negatively affect their task performance. The heterogeneous Swedish group put together 105 pieces, while the Greek heterogeneous group put together 128 pieces within the 25 minutes.
Discussion
Our first hypothesis was merely a baseline, which was in agreement with Hofstede and Hall’s research about individualism/collectivism, and how it relates to task performance. Third hypothesis was confirmed, again in agreement with their conception of socialization between individualist and collectivist cultures. However, the second hypothesis: heterogeneous individualistic group will finish the task faster than heterogeneous collectivistic was not supported. The results were that the heterogeneous Greek group completed more pieces of the puzzle than the heterogeneous Swedish group.
As was substantiated in the baseline, we anticipated that the heterogeneous Greek group would have a lower level of task performance than the heterogeneous Swedish group.
Our statistics showed that the Greeks were more socially interactive than the Swedes (t-test), which we erroneously reasoned would cause them to complete less of the task. An interpretation of this could be that the Greeks fostered a more effective social climate, i.e. more laughing, smiling, initiating conversation, and other social verbalizations. It might be also a consequence of Greeks’ motivation to achieve, as a group, the best possible results within the time limit. Moreover, this social climate also may have been the catalyst for group cooperation, which was observed in the frequency of passing pieces of the puzzle among the other group members.
Conversely, the heterogeneous Swedish group was less socially interactive (t-test) and seemingly more focused on the task. The lower levels of social interaction were also confirmed in the duration of silence for the Swedes. In regards to social climate, the heterogeneous Swedish group seemed to be more stressed and less cooperative. Some of the behaviours observed were frequently touching the face (perhaps a sign of anxiety/nervousness), using profane language, and more individual work.
Another aspect that is important to consider, which was discussed in the introduction, is that of homogeneity/heterogeneity. According to the research that we cited, individualistic cultures prefer working with heterogeneous groups more so than collectivist cultures. The problem that we encountered was that the heterogeneous Swedish group was not necessarily heterogeneous, i.e. the three other cultures were individualistic, respectively. In this way, we were not able to accurately measure how Swede’s would potentially work within a true heterogeneous group as was possible with the Greeks.
It is important to mention that there is a wide range of individualists and collectivists. It is fallacious to think that the myriad of cultures can be categorized into two cultural dimensions, individualism and collectivism. The individualist’s that we observed were from Sweden, Germany, Canada, and America. Although this group shares the same cultural dimension of individualism, it does not account for their differences: language, culture, history, politics, etc.
The collectivist group performance should also be explained in light of the variation of collectivist dimension. Most of the articles that we used for our research often compared the two extremes of the individualism/collectivism continuum: United States versus Japan or Korea. It is inconceivable to think that there are no differences between Japan or Korea and Greece. As was mentioned earlier, this cultural dimension does not account for each country’s language, culture, history, politics, etc. Individualism/collectivism is more appropriately understood in terms of a continuum than two separate categories. In other words, for both individualist and collectivist cultures one should not be surprised to find aspects of this cultural dimension in the differing culture, (e.g. Swedes have a tendency to collectivism; Greeks have a tendency to individualism). The impact that globalisation has had on the world and the ability to travel across the globe renders modern cultures to inevitably adopt values, norms, and such from different cultures. Hence, one will be able to witness aspects of individualism/collectivism either through mainstream or subculture.
The results of our experiment are in many ways task dependent. It was not until after the experiment that we realized that depending on the type of task, (e.g. individual oriented or group oriented) that it would favour one group over the other. This is one explanation why the Greek heterogeneous group was more effective than the Swedish heterogeneous group.
According to the theories that we used to formulate our hypotheses, individualists prefer individual oriented work (but do not mind working in groups, especially heterogeneous ones) compared to collectivists’ preference for group-oriented work (especially the more in collectivist cultures such as Japan, Korea, etc). However, the results may also suggest that cultural diversity in a group contributes to its’ more efficient work. Group size is another factor that inhibited our research from yielding significant results. That is not to say that the group size was not optimal (ideally, group size should be between 5-8 persons depending on the task). At the same time, it is hard to generalize the results for each group. Thus we used higher levels of significance to show similarities and/or differences. Furthermore, we had to use a fine-grained checklist to be as accurate as possible.
Along the same line of thought, we concluded that in order to have accurate results that the experiment should be replicated at least twice. We were under two restraints in regards to replication: time and participants. The problem was the coordination of these two variables, especially at the end of the semester.
The implications for our research relate to the way culture influences task performance among heterogeneous groups. It also opposes the inductive reasoning that talking inhibits task performance. We found that verbalizing and task performance can occur simultaneously and does not have a negative affect on the task performance. On the contrary, it is likely to create a more conducive work environment that generates more productivity. In a broad sense, this is a good example of how diversity can be a positive influence in the work environment, (e.g. globalisation and out-sourcing).
Further research could invert the experiment and observe how collectivist cultures task performance and socialization behaviours are demonstrated in an individualistic task. It will also be interesting to have a homogeneous collectivist group versus a homogeneous individualistic group (theoretically homogenous), which was the case for our experiment.
References
Albert, D. R., (2004). Latino/Anglo-American differences in attributions to situations involving touch and silence. International Journal of Intercultural Relations,28, 253 – 280.
Breer, P. E., & Locke, E. A. (1965). Task experience: A source of attitudes. Homewood, IL:Dorsey Press.
Campion, A. M., Medsker, J. G., & Higgs, A. C. (1993). Relations between work group characteristics and effectiveness: Implications for designing effective work groups. Personnel Psychology, 46, 823-850.
Cox, T.H., Lobel, S.A.& McLeod, P.L., (1991). Effects of ethnic group cultural differences on cooperative and competitive behaviour on group task. Academy of Management Journal, 34, 827 – 847.
Eisenhardt, K. M., & Tabrizi, B. N. (1995). Accelerating adaptive processes: Product innovation in the global computer industry. Administrative Science Quarterly, 40, 84-110.
Hall, E. T. (1988). The hidden dimensions of time and space in today’s world. In F.Poyatos (Ed), Cross-cultural perspectives in nonverbal communication (pp. 145-152).Goettingen, Germany: Hogrefe and Huber.
Hofstede, G. (1991). Cultures and organizations: Software of the mind. London: McGraw Hill.
Lincoln, J. R., Hanada, M., & McBride, K. (1986). Organizational structures in Japanese and U.S. manufacturing. Administrative Science Quarterly, 31, 338-364.
Triandis, H. C., (1994). Culture and social behavior. London: McGraw – Hill Inc.