D. Douglas Caulkins

Using Consensus Analysis to Measure Cultural Diversity in Organizations and Social Movements

Consensus analysis, a technique developed in cognitive anthropology for analyzing structured interview data, produces three useful results: (1) a measure of the degree of agreement among informants about a domain of knowledge, belief, or practice; (2) the “culturally correct” information about that domain according to the pooled answers of the informants; and (3) a score for each informant representing that person’s knowledge of the domain. Consensus analysis is not just for high-agreement domains, however. This article explores a typology for conceptualizing diversity in low-consensus domains, including (1) weak agreement, (2) turbulent, (3) subcultural, and (4) contested domains, using case study examples from an English social movement, a Scottish high-technology firm, and a Scottish business support and training organization. The typology helps measure and interpret diversity and change within organizations and social movements.

Evaluating consistency in typology and classification

Typological systems are essential for communication between anthropologists as well as for interpretive purposes. For both communication and interpretation, it is important to know that different individuals using the same typology classify artifacts in similar ways, but the consistency with which typologies are used is rarely evaluated or explicitly tested. There are theoretical, practical, and cultural reasons for this failure. Disagreements among archaeologists using the same typology may originate in the typology itself (i.e., imprecise type definitions, confusing structure) or in the classification process, because of observer errors, differences in perception and interpretation, and biases. We review previous attempts to evaluate consistency in typology and classification, and use consensus analysis to examine one well-established typology. Both consensus and disparity are apparent among the typologists in our case study, and this allows us to explore the kinds of forces that shape agreement and diversity in the use of all typological systems. We argue that issues of typological consistency are theoretically and methodologically important. Typological consistency can be explicitly tested, and must be if we hope to use typologies confidently.

Is Mary Douglas's grid/group analysis useful for cross-cultural research ?

During the 1970s, anthropologist Mary Douglas developed a twodimensional framework for cultural comparisons: (a) grid or constraint by rules, and (b) group or incorporation into a bounded social unit. According to Douglas and her colleagues, the four grid/ group types constitute stable social configurations that are associated with distinctive values or ideologies: individualism, fatalism, hierarchy, and egalitarianism. The theory has inspired a great deal of elaboration, numerous intrasocietal studies, some controlled comparisons between social units in different societies, and some limited cross-national studies. Although plausible, the framework has not been subjected to extensive cross-cultural testing. This article addresses a preliminary question in the more rigorous crosscultural testing of the framework: Are grid, group, and ideology empirically separate dimensions as Douglas contends? A pilot study employing factor analysis of precoded variables from the 60-culture Human Relations Area Files (HRAF) probability sample file suggests that they may be. Further steps in testing this theoretical framework are proposed.

Identifying Culture as a Threshold of Shared Knowledge A Consensus Analysis Method

This methodological article takes Fredrik Barth’s anthropology of knowledge as a point of departure and identifies culture as knowledge shared above a specified threshold. The method samples domains of knowledge in organizational settings and asks diverse members of the organization to list the elements of the domains. After compiling the elements, the informants are re-interviewed, along with a wider sample, to prioritize the elements according to a criterion of importance. A consensus analysis of the informants’ data matrix reveals the degree to which the knowledge is shared and constitutes a culture, or is less shared and constitutes a proto-culture, subculture, counter-culture, or a fragmented and idiosyncratic domain. Three case studies are used as illustrations. Widely adaptable in international management research for exploring organizational cultures and subcultures, inter-organizational fields, and international ventures, the consensus analysis method articulates with the three major theoretical perspectives on culture, the integration, differentiation, and fragmentation perspectives.

Consensus, Clines, and Edges in Celtic Cultures

Using data from four localities in peripheral areas of the British Isles, this article explores a conception of culture that emphasizes continuous variation, or clines, rather than boundaries. These localities are sites for the performance of cultural practices, many of which may be shared across socially constructed boundaries such as those of nation, ethnic group, and class. Consensus analysis provides a tool for exploring areas of greater or lesser sharing of cultural models indicated by responses to 21 brief narratives of everyday cultural practices. Informants were asked to judge on a 5-point scale whether the practices were characteristic of their location. With a high consensus in each site about typical practices, it was possible to compare the culturally correct profiles of responses between each of the sites, revealing incremental changes, or clines, from site to site. Edges, or the conjunction of clines, might be mapped using consensus analysis.

Grid-Group Analysis, Social Capital, and Entrepreneurship Among North American Ethnic Groups

A prevailing hypothesis among social scientists is that a high degree of social capital within immigrant groups promotes the development of self-employment, particularly in the first generation of immigrants. The authors identify social capital with high group placement in grid-group analysis for which they have produced new codes suitable for use with secondary data. Using the Electronic Human Relations Area Files sample of North American Immigrant Groups, each culture was coded on six measures each for grid and for group. Testing the social capital hypothesis using income and self-employment data from the 1990 census, the authors found a negative correlation between high group and high self-employment, particularly for the first generation. This finding suggests that high levels of general social capital on a group level may not be as supportive of entrepreneurship as previously asserted. The studys results are compatible with the “family social capital” model of entrepreneurship.

Using Scenarios to Construct Models of Identity in Multiethnic Settings

The task of inferring cultural models from lengthy texts or transcripts of intensive interviews is highly dependent on the representativeness of the sample of consultants, the researchers command of the language, and his or her ability to detect themes through iterative analysis. We recommend using scenarios and systematic interviewing to verify and refine those inferred themes, schema, or cultural models. Scenarios are brief narratives derived from observed or reported events. A set of scenarios can be used to elicit judgments concerning key themes of inferred cultural models. This technique is illustrated with a battery of twenty-one scenarios used, along with consensus analysis, to describe identity profiles for samples from Wales, Scotland, and Northeast England. Scenarios can be used to document claimed identity (what we are like) and attributed identity (what group X is like) and to discover similarities and differences across ethnic or national group boundaries.

Community, Subculture, and Organizational Networks in Western Norway

During the past five years there has been a resurgence of interest in the study of community social structure (Fischer, 1975, 1976; Wellman, 1979; Wellman and Leighton, 1979). Perhaps the greatest stimulus to the renewal of interests in the community as a unit has been the rapid development of techniques for the analysis of social networks, including interorganizational networks formed by overlapping leadership or some other form of exchange relationship among organizations (Galaskiewicz, 1979; Laumann, 1979; Levine and Roy, 1979).

Discovery-Mode Teaching Using the Electronic Human Relations Area Files for Cross-Cultural Comparison

To promote active, discovery-mode learning, the authors have experimented with the SilverPlatter Cross-Cultural Data Base and the electronic Human Relations Area Files in a variety of courses, from introductory to advanced. These include classes in cultural anthropology, archaeology, and ecological anthropology. The Probability Sample Files of 60 cultures has been used both as the central focus of the class and as a supplementary focus. After describing several of these classes, the authors review some of the decisions that must be made in designing a course using one or both of the databases. Both the benefits and the problems of this teaching approach are discussed.

A Note on the Prevalence of Voluntary Associations in Two Norwegian Provinces

The study of Norwegian voluntary associations on the national level is well developed (Hallenstvedt 1974, Hallenstvedt, et al 1972) but information about the distribution of voluntary associations on the local level is scattered and less accessible. Ropeid (1974, p.78) cites two local studies, one of which is unpublished, and Hallenstvedt (1974, p.216) calls attention to a third. The purpose of this note is to draw together some additional information concerning the prevalence of voluntary associations in local communities in two geographically separated Norwegian provinces. Park (1972, p. 6) notes that Norway comprises five regions: East, South, West, Midland and North. The local surveys discussed here were carried out in the provinces of More og Romsdal in the West and Nordland in the North. Three municipalities in More og Romsdal have been surveyed. In 1949 a team of Norwegian social scientistsl conducted a detailed study of Stranda, a town of approximately 2,200 inhabitants, which had shifted from economic dependence on farming to a diversified light industrial base. During 1967-1969 I studied voluntary associations in two municipalities, Volda, an economically diversified educational and cultural center with a population of 7,262, and Sunndal, where a new aluminum plant and a new town were located in the post-World War II period. In 1945 the population of the municipality was only 1,643; by 1968 the number had grown to 7,872. The survey of voluntary associations in the North (Arner 1959) was part of