The Impact of Culture on Group Decision Making Processes

§Introduction

Earlier in this volume I have argued that cultural elements must be considered in participatory environmental decision-making. Here I tie this work together with the previous chapter which focused on support for public involvement in environmental decision making by exploring some of the literature on cultural impacts on the decision making process. I provide specific examples of some of the ways that cultural processes can impact cognitive and decision making processes to demonstrate why these factors must be considered for effective policy.

As I am specifically interested in three cultures, Latvia, Fiji, and Belize, I initially attempted to locate literature specifically pertaining to these regions but found little that is directly related to group decision making with potential for extrapolation to situations involving environmental policy. As a result, I primarily examine literature involving several other cultures demonstrating factors of interest that are important in consideration in group decision making processes. Prior to the conclusion of this chapter I provide a brief summary of the work I was able to find specifically related to cultural differences in psychology for my case studies and explore their implications for participatory decision making.

As the focus of this paper is primarily on group decision-making in environmental policy, I only briefly discuss the individual cognitive processes with more of a focus on cross-cultural differences in-group dynamics. First I present two examples pertaining to specific micro-cognitive processes that provide support in demonstrating that cultural differences are found in some cognitive processes while other processes may be identical across cultures. Following a brief examination of the impact of culture on individual cognitive processes, I explore work that delves into larger scale dynamic processes found in groups.

Along with each highlighted study, I point to concerns which demonstrate the difficulty found in this research and further support the central thesis of this work that policy must be fundamentally driven by a representative grouping of the local populace and that many of these problems could be limited if the research and policy is locally guided. The primary aim of this chapter is to point to the fact that these factors must be considered for policy to be effective. If policy were to be enacted from a non-local level, I argue that to adequately inform this policy one would have to conduct individual targeted studies to investigate cross-cultural psychological differences on any potential factors involved in the policy and that these studies could still suffer from various complex problems prone to this field of research.

As there are a wide variety of potential problems in conducting such studies, I believe that the simplest means would be, in key with the rest of this work, to have policy enacted by a representative grouping of the local populace, which would implicitly result in inclusion of potentially important implications for cultural factors in designing policy. For a much more complete review of the literature on cross-cultural psychological differences, please see Smith and Bond (1999) and Matsumoto and Juang (2004).

§Cultural Differences in Individual Cognitive Processes

There is considerable evidence that cultural differences can have a dramatic impact on individual cognitive sub processes. For example in a cross cultural examination of individual cognitive processes, Oishi and Diener (2003) use an anagram puzzle and basketball task to examine how Asian Americans and European Americans differ in their memory of performance and decisions regarding future task choice and level of enjoyment on future tasks. The authors found that, in contrast to Asian Americans, European Americans remembered their task one performance significantly better and that their choice and enjoyment in task two was significantly predicted by their performance in task one. The researchers point to how these findings support previous research that indicate lower levels of satisfaction, more self criticism, and a greater interest in self improvement for East Asians, whereas North Americans display higher levels of life satisfaction, see themselves in a positive light, and tend to focus on aspects at which they already do well.

Though Oishi and Diener (2003) provide an interesting study with support for previous research, a very small sample size (18 European Americans and 25 Asian Americans) and the fact that samples were taken from undergraduate classes reflecting limited age and education ranges make strong conclusions problematic. Many of the Asian students had English as a second language, which could also have biased the results; especially considering the task was anagrams, which are language intensive. Alleviating some of this concern is the fact that Asian students solved more anagrams than the European students (6.6 compared to 6.4). In the basketball task, again we might have some bias because possibly one culture spends more time in this activity and is thus feels more comfortable engaging in it.

Despite evidence such as the Oishi and Diener study which point to cultural differences for some tasks, other studies also indicate that some cognitive processes truly do function similarly cross culturally. In this vein, Granberg (1999) presents a cross-cultural comparison of the Monty Hall dilemma, a two-stage decision problem of conditional probability. Previous studies have shown that when subjects are given a choice of three doors, one of which contains a prize and then one of the not selected doors is revealed as not containing the prize, when allowed the option of switching to the other door not opened, subjects generally incorrectly choose to stay on their initial choice even though the best choice would be to switch. The option of switching provides a 2/3 probability of winning the prize whereas it is incorrectly perceived as 50/50. Granberg (1999) found that subjects in all cultural groups chose the poor choice most often.

Though this study provides evidence that the Monty Hall dilemma may be a universal cognitive illusion Granberg (1999) also caution against strong inferences without replication in other cultures. The author also points out that, though the question achieved adequate back translation, possibly some of the underlying semantic meanings of the things being asked might be different in these cultures and thus more work needs to be done. For example, this dilemma focuses on winning a car and was asked in China where there are few cars and also people aren’t used to the format of a game show which is the basis of the dilemma.

From my review and examination of the literature, including the aforementioned manuscripts of Smith and Bond Smith (1999) and Matsumoto and Juang (2004), it appears that though there is considerable research indicating cross cultural difference or similarity on various psychological traits, including personality, emotional expression, and time perspective, more research needs to be done to better understand how individual cognitive biases might differ in different cultures.

§Cultural Differences in Group Processes

§Participatory Decision Making

There is a large amount of support for the contention that culture significantly impacts group interaction and specifically group decision making. Sagie and Aycan (2003) argue that culture is a key cause of difference in participative decision making (PDM) processes and approaches throughout the world. They further provide support for their opinion that sub-cultural and organizational cultural differences might explain a significant amount of the variance in PDM within nations. To examine these theories, they focus on differences in individualism-collectivism and power distance in relation to four common PDM approaches: collective PDM, pseudo PDM, paternalistic PDM, and face-to-face PDM. The authors believe that culture is a major determinant of the style of PDM at each level of organization.

Sagie and Aycan (2003) provide a useful elaboration of PDM processes in general and specifically in relation to cultural dimensions such as individualism and collectivism. The authors provide detail on specific cultural factors that might cause variance in decision-making and also demonstrate ideas for how PDM might be modified within different cultural settings. Though their study is relatively general and does not provide much in the way of empirical hypothesis testing it provides a useful connecting piece between experimental decision-making and policy decision-making.

Lam et al. (2002) study the effect of participatory decision making on employee performance for subjects in branches of the same organization in Hong Kong and the U.S. The researchers found that self-efficacy in relation to decision-making and ideocentrism moderated the relationship between the perception of opportunity for participatory decision-making and individual performance. This paper examines culture in relation to two fundamental concepts: the construct of allocentrism which is the concept of how much one views collective goals as more important and the concept of ideocentrism which is the concept of how much one prioritizes individual goals over collective goals. This paper concludes by demonstrating that U.S. participants scored higher on ideocentrism, and reported higher self-efficacy, whereas Hong Kong participants scored higher on allocentrism and reported higher collective efficacy. The authors further argue for the implications of these differences for management styles and propose that management should take individual cultural and other differences into account when considering possible methods of participatory decision-making.

Lam et al. (2001) used a robust methodological framework with a large subject pool, back translation for questionnaires, and seemingly representative samples. This paper provides important points regarding participatory decision making with demonstration that certain underlying cognitive factors that might vary within cultures might be more important than generalized cultural influence even though the shape of underlying cognitive factors might have greater similarities in certain cultures than others.

§Expert and Lay Opinion

Of particular interest for policy decisions in relation to public participation are differences found between experts and the lay public. Finucane (2002) examines the influence of cultural factors on perceptions of risk and the implications of such perceptions for decision-making that incorporates both the lay public and experts in different cultural settings. Finucane (2002) focuses on a contrast of the contextual factors in the U.S., Europe, and developing countries in relation to perceptions of issues relating to genetically modified (GM) foods. The fundamental thesis is that cultural values in different areas profoundly impact perceptions of risks of new technologies and that these cultural differences must be considered for effective communication and decision-making.

The analysis of Finucane (2002) provides an overview on the differences in reaction to GM goods in different cultural settings and their potential implications for policy. Finucane (2002) provides a particularly interesting examination of risk in relation to cultural factors with the argument for the need for greater public involvement in all three cultural settings examined. This paper provides a bridge between those focusing on cognitive processes and others that give greater details about how public participation might occur in different settings and what specific implications might exist for public policy as a result of socio-cultural factors.

Harris and Nibler (1998) compare the decision-making ability of individuals, in-groups, and minimal groups in Hong Kong and the U.S., to that of experts. They used individuals, groups of five people that know each other, and groups of five people without connection, as comparison groups. Harris and Nibler (1998) defined decisions similar to those of people classified as experts as the most effective. Harris and Nibler (1998) found that Chinese in-groups were more effective than Chinese minimal groups and that U.S. minimal groups outperformed U.S. in-groups. Furthermore, they found that Chinese groups under performed US groups, that they took 3 times longer in deliberation, and that they perceived their interactions as more conflict ridden and intense. They hypothesize that the differences in minimal versus in-groups result from cultural differences in bonding structures. They postulate that individuals in intimately bonded groups in collectivist cultures, like China, are more aware of their hierarchical status and thus not concerned with engaging in conflict where the rule structure is already dictated. U.S. in-groups, on the other hand, might suppress their full abilities because they want to repress such possible hierarchies.

Though examination of Harris & Nibler (1998)’s study points to the possibility that hierarchy might perform different functions in China and the US more work needs to be done to fully draw such conclusions. Social bonds in the US might be primarily for non-performance related functions whereas in China these groupings (or at least the ones the researchers chose) might be more practically aimed for problem solving. Several factors from this article should be considered in future studies: 1) The researchers administered the task only in English arguing that the Chinese students didn’t have a problem with understanding but did not provide adequate proof or examination of potential problems arising from this issue, 2) The researchers compared two fairly different groups (fraternities/sororities in the US with a not well elaborated choice of the networks of ‘key students” in China, and 3) the researchers had no elaboration on the process used for determining validity of expert opinion beyond saying that subject groups were compared to experts in the U.S. Merchant Marine. Perhaps experts in the Hong Kong merchant Marine would have differing opinions? Furthermore, future studies should address how to select who is considered an expert (Harris & Nibler 1998).

Nibler and Harris (2003) aim to replicate and expand on the previously cited Harris and Nibler (1998) study to investigate the influence of culture on effectiveness of intra-group conflict resolution and group decision making processes. They used five member groups from Hong Kong and the U.S. to determine the influence of friendship bonding on the effectiveness of decision-making and task conflict levels in these two different cultures. They found that U.S. participant performance and task conflict levels varied together with U.S. poor performance and low levels of task conflict in strangers and high levels of both conflict and performance in U.S. friendship groups. Chinese groups were found to have high levels of conflict and poor performance levels regardless of friendship status. They suggest that their study gives support for culture specific task conflict advantage for individualistic cultures in which members might feel more comfortable expressing and exchanging differing ideas and opinions.

Nibler and Harris (2003) along with Harris and Nibler (1998), provide interesting, yet seemingly very contradictory results. In Nibler and Harris (2003), it appears that individualistic cultures such as the U.S. strongly benefit from friendship bonds whereas in the Harris and Nibler (1998) study, the conclusions implied that collectivist cultures were those benefiting of such bonds. Future work would benefit from understanding the reasons for these seemingly contradictory findings.

In the Harris and Nibler (1998) study they refer to “in-groups” whereas in the 2003 study they talk about friendship groups. Future work would benefit from elaboration on the possible differences between such groupings; especially since it appears that the methodology for recruiting participants was the same in both of the articles (fraternities and sororities in the US and key students in Hong Kong). As in the previous study, this study only uses English without adequate justification for possible differences (especially considering both studies showed poorer time performance in the Hong Kong sample which may well have resulted from attempting to be clear on the correct definitions in what may have been a second language for the Hong Kong participants). Task specific biases were also not adequately addressed. Both studies use an abstract exercise focusing on skills in being lost at sea; previous knowledge of such skills might well vary widely in different cultures. The researchers point to a variety of interesting possible cultural difference such as Chinese participants being more algorithmic and U.S. participants being more reliant on interpersonal interaction in decision-making but adequate support for these theories is not offered. The researchers themselves point to several possible weaknesses in their study, including the possibility of use of response sets, sample bias, and task bias (Harris and Nibler, 1998, Nibler and Harris, 2003).

§Escalation for Project Commitment

Greer and Stephens (2001) who examine the tendency towards escalation of commitment in projects for Mexican and U.S. decision makers demonstrated further support for cultural differences. The authors found that Mexican subjects were more likely to proceed to escalation and have more confidence in escalatory decisions. Decision makers in both countries escalated commitment decisions upon hearing bad news from subordinates. The context for this study is attempting to understand the common poor decision of continuing investments in projects that are not providing returns and seeing if such practices are similar cross-culturally.

Greer and Stephens (2001) also provide a nice contextual literature review that relates the concepts presented to basic decision-making theory such as the prospect theory work of Kahneman and Tversky. Methodologically their study is interesting because it examines culture as a moderator variable instead of as a main effect. This paper also points to how cultural dimensions have changed over time and must regularly be recalculated and updated to ensure validity in comparison studies. Though these authors provide an interesting context and multiple hypotheses for the apparent cultural difference (such as machismo in Mexican culture causing decision makers to be more wary of changing decisions), I’m interested to find out if the impact of escalation is equally detrimental in both cultural settings. Possibly in Mexico escalation is actually a more valid decision than in the U.S. during circumstances of apparent financial loss due to some, as yet unknown factors. This study is interesting in terms of participatory decision making because Mexico is viewed as a more hierarchical culture than the U.S. where decisions are solely the responsibility of decision makers, whereas the management culture in the U.S. is viewed as more accepting of participatory and egalitarian ideologies.

§Ethics and Policy

Jackson (2001) presents a 10 nation empirical study used to examine cultural differences in how business managers deal with ethical gray areas, such as bribery, insider dealing, and pollution in management decisions. Countries were chosen to allow for examination of variation on the cultural dimensions of collectivism-individualism and uncertainty avoidance. Jackson (2001) states that his study confirms that elaborations of these dimensions explain much of the difference in ethical attitudes among differing cultural groups and that national differences are often due to differences in cultural values. He furthermore believes that his study provides support for the idea that, though ethical attitudes differ among different cultural groups, the perception of other managers as more unethical than themselves is a universal for those cultures under study.

Jackson (2001) incorporated a fairly large sample size of 425 participants from10 different nations. A relatively extensive literature review and discussion is incorporated that provide a detailed context of the specific cultural dimensions under study and their relationship to managerial ethics and decision-making. The nations examined are also relatively diverse, including representatives of each common combination of cultural dimensions. These include the U.S., France, the People’s Republic of China, and India. The author also points to possible future improvements in measuring cultural dimensions, especially in relation to the subject of managerial ethics. Though effort at back translation was made, only Spanish and Chinese samples received non-English versions of the instruments, thus possibly resulting in second language bias. In general this study was based on a strong methodological framework with instruments validated for reliability in all cultures studied. Furthermore, Jackson (2001) provides support for his hypothesis of a cultural universal and also cultural differences.

Rahim et al. (2001) attempt to determine whether justice relationships differ between U.S. employees and Bangladesh employees. To accomplish this they use hierarchical regression analyses to examine the interaction of distributive, procedural, and interactional justice with organizational commitment and turnover intension in the U.S. and Bangladesh. Four samples were examined: U.S. Faculty, U.S. Managers, Bangladesh Faculty, and Bangladesh managers. Rahim et al. (2001) state that there is more variance within the four sub-samples than between the two cultural groups examined.

A potential flaw in the study conducted by Rahim et al. (2001), to which the authors point, is that Hofstede’s dimensions, which measure factors such as individualism versus collectivism and degree of inequality in power, are not directly measured in these samples. The researchers go by the assumption that Bangladesh would be similar to India and Pakistan because of their historical proximity and the claim that these countries are similarly different from the U.S. on cultural dimensions. Closer examination of the numbers they provide indicate though that both Pakistan and India independently vary widely on the dimension scales and thus the assumption of homogeneity of these three countries and further whether Bangladesh’s has an equivalent place on these dimensions is in question (India was 10th/21st, Pakistan 32nd/47th and U.S. 38th/1st on power distance and individualism respectively). Other concerns relate to sample selection. While Bangladesh had reasonable response, the U.S. sample was comprised of a relatively small percentage of respondents (154/750 for U.S. Faculty and 148/427 for U.S. managers). Also, age ranges and work experience appear significantly different cross-culturally. Furthermore, as in many studies, the questionnaire was asked only in English, thus possibly resulting in second language bias. Though the authors claim that their study supports the view that the same normative measurements can be used cross culturally, concerns about sample differences and inadequate analyses of other cross cultural variables put into question the advisability of drawing such conclusions without further research.

§Cultural Differences Specific to Latvia, Fiji, and Belize

Though I had difficulty finding sources specifically directly related to cultural differences in participatory group decision making for my case study areas, I did find some references indicating possible cultural psychological differences which might indirectly influence the decision making process. For example, Lelkes (2002) provide results of a World Values Survey from 1996 in which only three percent of Latvians rated themselves as very happy. This is the lowest number of any group presented, with values typically in the range of fifteen to thirty percent and as high as forty two percent for Iceland and Ireland. Diener & Tov (in press) provide further support for low levels of comparative happiness in Latvia with a mean life satisfaction level at 5.27 out of ten, indicating Latvia was in fifty fifth place out of sixty eight countries, topped by Puerto Rico at 8.49 (for reference, the United States was thirteenth from the top at 7.66).

Particularly interesting for the present report is the work of Au (1999). Au found that Latvians had the highest mean level of pride in work out of forty two cultures shown but had one of the lowest mean levels of freedom in decision making. In this study, Latvians had approximately average levels of job satisfaction and appreciation for change and new ideas.

Though one potential avenue that might initially appear promising in situations with a lack of adequate data on specific cultural differences would be to examine studies with apparently similar cultures and assume similarities to the culture in which one is planning policy, there are inherent problems with making such assumptions. Embarking in such practice may appear to save time and money but the long-term policy might well suffer. For example, Smith and Bond (1999) provide summary results indicating that Estonians are very high on conservatism and thus one might assume that Latvians are also high in this regard because both are Baltic states with various historical factors in common. As the following study shows, this could be a dangerous assumption and individual cultural studies should ideally be conducted for any given study area with specific focus on the factors which might be of interest if an outside source is hoping to inform policy.

Despite possible similarities on some levels between the Baltic States, work, such as that of Liubiniene (2003) demonstrates that these countries can also differentiate dramatically. Liubiniene (2003) showed that Latvia, Lithuania, and Estonia all differentiated on attribution of importance to such factors as family health, world peace, honoring parents, social justice, social life, wealth, self fulfillment, self-discipline, variance in life, being influential, euthanasia, divorce, abortion, avoidance of fairs, homosexuality, prostitution, stolen goods, suicide, accepting bribes, and social benefits. Furthermore, Liubiene (2003) demonstrated that contrasts are found between various age groups even within these cultures, with demonstrated differentiation between three groups: ages18-29, ages 30-49, and ages 50-75.

Even within countries there can be dramatic cultural variation, which must be considered. For example one might decide to enact a policy that assumes residents of Fiji are high on levels of conformity following perusal of work such as that presented by Smith and Bond (1999: 16) who provide summary results of studies indicating that Fiji is very high in conformity. The fallacy in this assumption is apparent in further examination; later on in the same manuscript Smith and Bond (1999: 165) indicate that these conformity studies might have been done on the Indian population alone. Possibly Indians and Fijians (both large groups within Fiji) vary dramatically in their levels of conformity.

The issue of cultural variation within a country is of specific important for consideration in all three of my case studies. Along with the original inhabitants, Latvia has a large populace of Russians, Fiji has a large populace of Indians, and Belize has seven major ethnic groups with potentially important cross-cultural differences.

This issue is even further compounded by the strong component of intra-cultural variation (ICV), which is often ignored within studies of cultural psychology. ICV is a measurement of the way a characteristic is distributed within a culture (Au, 1999). As Au (1999) demonstrates in a study of 42 cultures, there is dramatic variance on some characteristics within cultures themselves, whereas other characteristics are relatively homogonous on an individual cultural basis.

§Summary and Conclusions

This chapter provides a cursory literature review with the primary point being to demonstrate that cultural factors can have significant impact on cognitive and, more importantly for the present document, group decision making processes. Cultures do not always differ on any given aspect but before adopting any policy based on a different culture or creating any new decision making process, it is important to consider these factors.

The literature demonstrates that cultural differences do exist in cognitive biases and group decision-making processes. Furthermore, though there is a relatively large amount of literature in this direction, many of the studies have possibly serious flaws, such as being conducted in participants’ second language and inadequate attention to biases of the researchers conducting the study.

For this reason, I argue that this further supports my primary thesis that representative groupings of the local population are the most ideal people to enact policy. Policy makers from a different culture may well not reflect the needs of local cultures and thus create a situation in which the local residents refuse to participate in any created policies. If a truly representative local grouping is driving the policy, the needs created by potential cultural differences of the involved groups are much more likely to be incorporated. An outside party that is fully unaware of how factors differentiate in the target population’s culture from their own, or even one who has knowledge of findings of some cross cultural studies could fail because of an inadequate knowledge of the local culture.

Decision-making conducted by a representative local grouping is implicitly going to be informed by relevant cultural factors. Furthermore, this chapter demonstrates that as these factors can differ across cultures, if one is adopting theory from another culture, ideally local researchers would examine the work and, if necessary, conduct studies to examine whether relevant cognitive and group decision making biases function in the same ways for their own situation.

The potential impacts of cultural differences are of primary import for the three cultures presented as case studies. All three are multi-cultural societies and all three have policy that is largely impacted by other countries. Latvia has recently acceded to the European Union. As was earlier demonstrated in this volume, oftentimes environmental policy in smaller or less wealthy countries can be driven by larger and wealthier states and oftentimes also with inadequate results.

The British Common Wealth and more recently the United States have historically impacted Belize and Fiji. Latvia is in an interesting position as a new member state in the European Union because other member states of the European Union, particularly those with more money, might be interested in enacting policies without adequately considering differences found in the Latvian populace and thus these policies could well fail if cultural and other local factors are not adequately taken into account. I argue that the only way for effective policy to be enacted in these countries is to create a situation in which a representative local grouping is the primary architect of the decision making process.

§Literature Cited

• Au, K. (1999). Intra-cultural variation: Evidence and implications for international business. Journal of International Business Studies, 30 (4), 799-812.
• Diener, E., & Tov, W. (in press). Culture and Subjective Well-being. In S. Kitayama & D. Cohen (Eds.), Handbook of cultural psychology.
• Finucane, M. L. (2002). Mad cows, mad corn, and mad communities: The role of socio-cultural factors in the perceived risk of genetically-modified food. Proceedings of the Nutrition Society, 61, 31-37.
• Granberg, D. (1999). Cross-cultural comparison of response to the Monty Hall dilemma. Social Behavior and Personality, 27(4), 431-438.
• Greer, C. R., & Stephens, G. K. (2001). Escalation of commitment: A comparison of differences between Mexican and U.S. decision-makers. Journal of Management, 27(1), 51-79.
• Harris, K. L., & Nibler, R. (1998). Decision making by Chinese and U.S. students. The Journal of Social Psychology, 138, 102-114.
• Jackson, T. (2001). Cultural values and management ethics: A 10-nation study. Human Relations, 54(10), 1267-1302.
• Lam, S. K., Chen, X., & Schaubroeck, J. (2002). Participative decision making and employee performance in different cultures: The moderating effects of allocentrism/ideocentrism and efficacy. Academy of Management Journal, 45(5), 905-915.
• Lelkes, O. (2002). Life satisfaction and transition from socialism. Paper prepared for ISA Research Committee on Social Stratification and Mobility (RC28) Oxford Meeting. (http://www.nuff.ox.ac.uk/rc28/Papers/Lelkes.pdf).
• Liubiniene, V. (2003). Process of Value Change in the Baltic Countries. Paper prepared for the European Consotrium for Political Research. (http://www.essex.ac.uk/ecpr/events/generalconference/marburg/papers/16/6/Liubiniene.pdf).
• Matsumoto, D., & Juang, L. (2004). Culture and Psychology (3rd ed.). Belmont, CA: Wadsworth Publishing.
• Nibler, R., & Harris, K. L. (2003). The effects of culture and cohesiveness on intragroup conflict and effectiveness. The Journal of Social Psychology, 143(5), 613-619.
• Oishi, S., & Diener, E. (2003). The cycle of action, evaluation, and decision. Personality and Social Psychology Bulletin, 29(8), 939-949.
• Rahim, M., Afzalur, M. Nace, R., Antonioni, D., & Rahman, S. (2001). Do justice relationships with organization-directed reactions differ across U.S. and Bangladesh employees? International Journal of Conflict Management, 12(4), 4044-4068.
• Sagie, A. & Aycan, Z. (2003). A cross-cultural analysis of participative decision making in organizations. Human Relations, 56(4), 453-473.
• Smith, P. & Bond, M. (1999). Social psychology across cultures. Allyn & Bacon, Needham Heights, Massachusettes.