Toward a Theory-Based Measurement of Culture - Semantic Scholar

Vol. 10, No. 1

Toward a Theory-Based Measurement of Culture DETMAR STRAUB, Georgia State University, USA KAREN LOCH, Georgia State University, USA ROBERTO EVARISTO, University of Illinois, USA ELENA KARAHANNA, University of Georgia, USA MARK STRITE, University of Wisconsin, Milwaukee, USA

In reviewing the history of the conceptualization and measurement of “culture,” one quickly realizes that there is wide-ranging and contradictory scholarly opinion about which values, norms, and beliefs should be measured to represent the concept of “culture.” We explore an alternate theory-based view of culture via social identity theory (SIT), which suggests that each individual is influenced by plethora of cultures and sub-cultures–some ethnic, some national, and some organizational. In IS research, the culture of subjects and respondents is problematic because it is typically an overly simplistic categorization. IS research nearly always assumes that an individual living in a particular place and time belongs to a single “culture,” e.g., someone living in Egypt is automatically classified as being a member of the Egyptian culture, or, more broadly, the Arab culture. This dearth of clear concepts and measures for “culture” may explain why cross-cultural research has been so exceedingly difficult to conduct. It may also explain why it has been hard to develop and refine theories. Moreover, it may give insight into why reasonable explained variance in predictive models has not been higher. Finally, it is very possible that much cross-cultural business research could be rightly accused of advancing an “ecological fallacy” by not recognizing the individual makeup of persons with respect to culture. Using SIT (or other theory bases) as grounding for cultural research programs implies the use of certain methodological approaches. Each study would have to establish the salient “cultures” in each individual’s background and include these different “cultures” as independent variables in positivist research. In qualitative research, there would need to be an equally rigorous assessment of the cultural identifiers of each individual.

INTRODUCTION Globalization of business highlights the need to understand the management of organizations that span different nations and cultures. In modern multinational, transnational organizations, information technology (IT) must be utilized to achieve efficiencies, coordination, and communication. Clearly, though, cultural differences between countries impact the effectiveness and efficiency of this IT deployment. A study of cultural conflicts, therefore, is of paramount importance for modern organizations and for IT scholars. Despite its universally recognized importance, the effect of cultural factors on IT outcomes has received limited attention from information systems (IS) researchers. As a result cross-cultural information systems research, in general, remains in a state of infancy. Although several important research endeavors have been recently published in the topranked, established IS journals, the overall number of cross-

cultural articles is fairly low, considering the number of practical and theoretical critical questions that remain unanswered (Gallupe & Tan, 1999). This disparity can be partly explained by methodological and resource difficulties inherent in cross-cultural research as well as the long time horizon required to complete/conduct these types of studies. It may also be explained by the lack of unanimity about the underlying meaning and definition of the underlying construct “culture.” In this essay, therefore, we explore the meaning of “culture” and consider new ways of conceptualizing and measuring it for global information management research. In reviewing the history of definitions of “culture,” one quickly realizes that there is wide-ranging and contradictory scholarly opinion about what constitutes “the” set or even a reasonable set of values, norms, and beliefs for “culture.” We explore an alternate theory-based view of culture via social identity theory (SIT), which suggests that each individual is

Journal of Global Information Management

Jan - Mar 2002 Copyright © 2002, Idea Group Publishing.

13

influenced by a plethora of cultures and sub-cultures æ some ethnic, some national, and some organizational. In IS research, the culture of subjects and respondents is problematic because it is typically an overly simplistic categorization. IS research nearly always assumes that an individual living in a particular place and time belongs to a single “culture,” e.g., someone living in Egypt is automatically classified as being a member of the national Egyptian culture, or, more broadly, the ethnic Arab culture. This dearth of clear concepts and measures for “culture” may explain why cross-cultural research has been so exceedingly difficult to conduct. Rather, we suggest that an individual’s social identity represents that amalgamation of cultures across boundaries (national, organizational, professional, etc.), which fuse together to create one’s overall culture. The combination is unique to each individual. It may also give insight into why it has been hard to find reasonable explained variance in predictive models. Finally, it is very possible that much of the work could be rightly accused of advancing an “ecological fallacy” by not recognizing the individual makeup of individuals with respect to culture. Using SIT (or other theory bases) as grounding for cultural research programs implies the use of certain methodological approaches. Each study would have to establish the salient “cultures” in each individual’s background, the composition of these “cultures,” and then include these different “cultures” as independent variables in positivist research. In qualitative research, there would need to be an equally rigorous assessment of the cultural identifiers of each individual. This approach also has the advantage of explicitly recognizing that these different layers of culture can intertwine in complex ways. Therefore, a particular behavior may be more influenced by a given layer of culture than others, implying that the layers do not have a fixed sequential position (Evaristo, Karahanna et al. 2000). If we consider an “onion” metaphor to describe these layers where layers closest to the core of the onion are more relevant, then the picture is of a “virtual onion” with inner layers occasionally exchanging places with outer layers. The layers are permeable. The thickness of the layers signals the strength of the value held by the individual. Moreover, the interrelationship between a layer and a specific behavior may vary depending on external circumstances – creating a virtual onion where each layer may move in or away from the core.

WHAT IS CULTURE? “Culture” has always been a thorny concept and an even thornier research construct. It has been studied for over a hundred years in disciplines such as cultural anthropology and in numerous other academic areas ranging from psychology to cross-cultural business management. The wide variety of scholars working in these areas have produced numerous definitions of culture. These definitions range from the simple to the complex, incorporate and extend previous

14

definitions, and even contradict prior definitions. Many researchers have used more than one definition of culture depending upon the time the definition was formulated and the subject matter to which it referred. In this section, we first examine the term culture from a historical perspective by assembling some of its earliest and most general definitions. We next group these historical definitions into three classes. “Culture” in Early Work According to Kroeber (1949), the word “culture” came into English usage (as distinct from cultivation and refinement) from nurture, from agriculture and pearl culture, and from test tube cultures in 1871. Tylor’s Primitive Culture (1871) defined culture as “that complex whole which includes knowledge, beliefs, art, morals, laws, customs and any other capabilities and habits acquired by man as a member of society” (p. 1). Prior to this, the term culture was used with its modern meaning in the German word “Cultur” as early as 1843 (Kroeber, 1949). Kroeber and Kluckhohn’s early review (1952) of cultural definitions found over 160 different instantiations. Definitions in the 1950’s were instrumental in establishing distinctions and etiological perspectives. Kroeber (1952) defined culture as “the historically differentiated and variable mass of customary ways of functioning of human societies” (p. 157). Parsons and Shils (1951) intimate that culture is composed of a set of values, norms, and symbols that guide individual behavior. Herskovits (1955) later argued that there “is a general agreement that culture is learned; that it allows man to adapt himself to his natural and social setting; that it is greatly variable; that it is manifested in institutions, thought patterns, and material objects” (p. 305). Subsequently, there has been a multiplicity of definitions of culture, classifiable into three main groups. The first group represents the most common view on culture and is labeled Definitions Based on Shared Values. The second group is Definitions Based on Problem Solving while the third group details a number of General All-Encompassing Definitions. Below we discuss key definitions in each of these categories. Definitions Based on Shared Values A number of scholars have focused on shared values as the central feature and distinguishing characteristic of a culture. Values refer to relationships among abstract categories that are characterized by strong affective components and imply a preference for a certain type of action. According to Rokeach (1973) a value is an “enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence. A value system is an enduring organization of beliefs concerning preferable modes of conduct or end-states of existence along a continuum of relative

Jan - Mar 2002

Journal of Global Information Management Copyright © 2002, Idea Group Publishing.

Vol. 10, No. 1

importance” (p. 5). of traditional (i.e., historically derived and selected) ideas and Values are acquired early in life, mainly through the especially their attached values” (p. 86). These patterned family and neighborhood and later through school. They values are said to differentiate cultures. provide us with fundamental values and assumptions about 1. Is innate human nature basically evil, good and evil, or how things are. Once a value is learned, it becomes integrated good? Is innate human nature mutable or immutable? into an organized system of values where each value has a 2. Is the relation of people to nature one of subjugation, relative priority. This value system is relatively stable in harmony, or mastery? nature but can change over time reflecting changes in culture 3. What is the temporal focus? Is human nature oriented as well as personal experience. Therefore, individuals based toward the past, present, or future? on their unique experiences not only differ in their value 4. What is the action orientation of the people? Is it toward systems but also in the relative stability of these value sysintensive appreciation of experience (being), toward total tems. development of the person (being-in-becoming), or toThere are a host of culture definitions based on values. ward accomplishment (action)? Identifying and describing culture as a set of value patterns that are shared across individuals and within groups is comThe presence of shared values that recur with regularity mon in this work. As early as the 1950s, both Parsons and was so evident that Murdock (1965) was able to list universal Shils (1951) and Kroeber (1952) include values as a core cultural traits, or values (e.g. athletic sports, bodily adornconcept in their definitions of culture. As mentioned above, ment, division of labor, government, hospitality, luck superParsons and Shils (1951) view culture as comprised of three stitions, population poicy, status differentiation, etc.), that distinct aspects: (1) types of cultural symbol systems, (2) exist in some form in all known cultures. The list is more types of standards of value-orientation (values), and (3) types extensive than the above examples and useful for identifying of orientation action (norms) (p. 166). universal points of similarities and differences across culTaking a similar point of view, Kroeber (1952) asserts tures, but provides no insight regarding the enactment of a that culture has the following qualities: “(1) it is transmitted particular value in a specific context. and continued not by the genetic mechanism of heredity but Another shared values commentator, Geertz (1973), by the inter-conditioning of zygotes; (2) whatever its origins in or through individuals, Table 1. Murdock’s (1965) Universal Cultural Values culture quickly tends to beUniversal Values come supra-personal and anonymous; (3) it falls into Age-Grading Food Taboos Music patterns, or regularities of form Athletic Sports Funeral Rites Mythology and style and significance; (4) Bodily Adornment Games Numerals it embodies values, which may Calendar Law Gestures Obstetrics be formulated (overtly, as Cleanliness Training Gift Giving Penal Sanctions mores) or felt (implicitly, as in Cooking Government Personal Names folkways) by the society carCooperative Labor Greetings Population Policy rying the culture, and which it Cosmology Hair Styles Postnatal Care Courtship Hospitality Pregnancy Usages is part of the business of the Dancing Housing Property Rights anthropologist to characterize Decorative Art Hygiene Propitiation of Supernatural Beings and define” (p. 104). Divination Incest Taboos Puberty Customs Kluckholn’s (1951) defiDivision of Labor Inheritance Rules Religious Ritual nition of culture draws attenDream Interpretation Joking Residence Rules tion to shared patterns of thinkEducation Kin-Groups Sexual Restrictions ing based on values. He deEschatology Kinship Nomenclature Soul Concepts fines culture as “patterned Ethics Language Status Differentiation ways of thinking, feeling and Ethnobotany Law Surgery Etiquette Luck Superstitions Tool Making reacting, acquired and transFaith Healing Marriage Trade mitted mainly by symbols, conFamily Mealtimes Visiting stituting the distinctive Feasting Medicine Weaning achievements of human Fire Making Modesty about Weather Control groups, including their emFolklore Natural Functions bodiments in artifacts; the esMourning sential core of culture consists



15

defined culture as “an historically transmitted pattern of meanings embodied in symbols, a system of inherited conceptions expressed in symbolic forms by means of which men communicate, perpetuate, and develop their knowledge about and attitudes toward life” (p. 89). Also focusing on patterns of thinking and values, Triandis (1972) examined subjective culture, which he defined as a “group’s characteristic way of perceiving the manmade part of its environment. The perception of rules and the group’s norms, roles, and values are aspects of subjective culture.” (p. 4). Triandis further states that “people who live next to one another, speak the same dialect, and engage in similar activities (e.g., have similar occupations) are likely to share the same subjective culture” (p. 4). Triandis’ definition is unique in that it embeds patterns of action and values within an extensive nomological network of proximal and distal antecedents that include, among others, historical events, economic activities, social and political organizations, language, religion, occupation, roles, tasks, and ideals. This worldview is intuitively attractive in that clanning or “staying with your own,” whether determined by historical events, religious belief, or occupation, is a readily observable behavior. It is also closely related to the social identification theories that we explore later as a new possible basis for cultural research. Patterns of thinking based on values underline the definition of culture proposed by one of the most prominent scholars in this school of thought. Hofstede (1980) defined culture as “the collective programming of the mind which distinguishes the members of one human group from another” (p. 260). In his study of 150,000 individuals around the world, he developed the following four heavily-used cultural patterns or dimensions: (1) individualism/collectivism, (2) power distance, (3) uncertainty avoidance and (4) masculinity/femininity. Individualism reflects the way members emphasize their own needs over the group’s needs. Power distance is the extent that large differentials of power and, therefore, inequality are accepted in a given culture. Uncertainty avoidance is the level of risk accepted by a culture, which can be gleaned from the emphasis on rule obedience, ritual behavior, and labor mobility. Finally, masculinity/femininity refers to culture differentiation on the basis of activity. This dimension does not related to attitude toward gender, although it is frequently misinterpreted to mean that. Masculine cultures tend to emphasize work goals such as earnings, advancement, and assertiveness. Feminine cultures tend to emphasize personal goals such as a friendly atmosphere, getting along with the boss and others, and a comfortable work environment. In collaboration with a group of Chinese researchers, Hofstede and Bond (1988) subsequently added a fifth dimension of culture: long-term orientation. In this cultural pattern, particularly observable in East Asia, groups have a much longer time horizon in their decisions, and,

16

therefore, the immediacy of any given situation is not as pressing as in the western cultures. Hofstede makes a distinction between values1 and practices. He suggests that while values are acquired in early life and are enduring, practices, which can be altered, are learned later, through socialization at the workplace, after an individual’s values are firmly in place. Among others, symbols, heroes, and rituals are all learned practices (Hofstede, 1991). Hofstede suggests that national cultural differences are composed primarily of differences in values and, to a lesser extent, of differences in practices. In a similar vein, Lachman (1983) argues that “there are values and attitudes that are more central, important, or dominant to the individual than others and hence will be called ‘core values’; second the core values resist change more than those which are peripheral” (p. 566-567). He also states that these core values are formed during childhood and are reinforced throughout life. Like Hofstede, Lachman also considers culture to be composed primarily of the core values and beliefs of people in a society. Lachman opens a line of thinking about core and noncore values that dovetails nicely with our virtual onion model, discussed later. In the 1990s, the “shared values” perspective is advanced by numerous researchers. Trompenaars (1993) believes that culture is composed of shared values. In particular Hampden-Turner and Trompenaars (1993) say, “members of a culture are likely to share common attitudes because they share a common history” (p. 13). Differing in many ways from Hofstede, he proposes seven dimensions of culture: (1) universalism/particularism, (2) individualism/collectivism, (3) neutral/affective relationships, (4) specific/diffuse relationships, (5) achievement/ascription, (6) internal/external control, and (7) perspectives on time2. The shared patterns view was still being advocated near the turn of the millennium. Erez and Earley (1997) for example, defined culture as the “shared way a group of people view the world” (p. 23). Hofstede’s Universal and Geographic Sense of “Culture” Hofstede’s (1980) work is unique because it offers a mechanism whereby a culture-value can be assigned to a particular group of people. This group is determined by a geographical boundary. The shortcoming in this approach is that there are recognized subcultures that: (1) span national geographical boundaries (Arab and Latin America cultures come to mind) and (2) nations that have strong internal cultural differences (English and French-speaking Canada; India with over 14 official languages & hundreds of dialects) or recognized intra-regional differences (United States and Germany). Organizational and professional cultures have also been ascribed identifiable value-sets (Laurent, 1991; Martin, 1992; Schein, 1985) such as management styles,

Jan - Mar 2002


Vol. 10, No. 1

appraisals, reward systems, communication styles, and manner of taking decisions that vary across context.3 Fundamentally, these instantiations of culture are all value-based; the key distinction is the boundary – e.g., the nation-state/geographic borders, organization, or profession. In summary, this categorization of definitions of culture suggests that culture consists of patterned ways of thinking that are shared across people in a society; these patterns are based on values. These values influence individuals’ cognitions, attitudes, and behaviors. Furthermore, there is a distinction between core and peripheral values. Culture is primarily a manifestation of core values. Finally, many researchers have attempted to capture cultural patterns in a set of dimensions. Table 2 contains a summary of the most commonly cited cultural patterns identified in the literature. However, researchers differ not only in which cultural patterns underlie culture, but also in their views on how these cultural patterns define culture. For example, Parsons and Shils (1951) suggest that culture is defined by the unique combination of cultural patterns.4 Viewing each pattern variable in isolation would likely lead to misleading conclusions. Kluckhohn and Strodtbeck (1961), on the other hand, suggest that each of the pattern variables can be meaningfully viewed separately. Furthermore, they argue that rank ordering of these cultural patterns distinguishes one culture from another. Thus, even though there appears to be a convergence in these conceptual definitions of culture, there is still considerable divergence in the specific nature of values that underlie culture and in the exact nature of the relationship between these values and culture.

Definitions Based on Problem Solving Rather than trying to define culture from the perspective of its composition (as in the previous section), another group of scholars have looked at the outcomes of culture and what it can accomplish. These researchers focus on problem solving and how this defines a particular group of people. This view persisted from the 1940s through the 1980s and into the early 1990s. Ford (1942) sees problem solving as the outcome when he states that culture is “traditional problem solving through accepted responses that have met with success. It consists of learned problem-solutions” (p. 546). Part of Kluckhohn and Leighton’s (1946) reasoning also is responsive to this theme. They argue that culture consists of “habitual and traditional ways of thinking, feeling, and [reacting] to that [is] characteristic of the ways a particular society meets its problems at a particular point in time” (p. 28). The group’s need to deal with its environment is also central to Schein (1999), who views culture as “the sum total of all the shared, taken-for-granted assumptions that a group has learned throughout its history” (p. 29). Schein (1985) also believes that culture involves learning within a group as that group solves its problems of survival in an external environment and its problem of internal integration. Finally, a goalsoriented definition also characterizes Moran and Stripp’s (1991) definition when they define culture as “a group problem-solving tool that enables individuals to survive in a particular environment” (p. 43).

Table 2. Cultural Patterns Value Orientation/Pattern Variable Associative versus Abstractive Ideologists versus Pragmatists Individualism versus Collectivism Masculinity versus Femininity Power Distance Uncertainty Avoidance Long Term Orientation, Temporal Focus Human Nature Man-Nature Activity Orientation Relational Orientation Dionysian, Promethean, Budhistic Self-Orientation versus Collectivity-Orientation Universalism versus Particularism Achievement versus Ascription Specificity versus Diffuseness Affective versus Affective-Neutral Internal versus External Control Time Perspective

Citation (Glenn, 1981) (Glenn, 1981) (Hofstede, 1980); (Trompenaars, 1993) (Hofstede, 1980) (Hofstede, 1980) (Hofstede, 1980) (Hofstede et al., 1988); (Kluckholn et al., 1961) (Kluckholn et al., 1961) (Kluckholn et al., 1961) (Kluckholn et al., 1961) (Kluckholn et al., 1961) (Morris, 1956) (Parsons et al., 1951) (Parsons et al., 1951); (Trompenaars, 1993) (Parsons et al., 1951); (Trompenaars, 1993) (Parsons et al., 1951); (Trompenaars, 1993) (Parsons et al., 1951); (Trompenaars, 1993) (Trompenaars, 1993) (Trompenaars, 1993)



17

General All-encompassing Definitions and Distinctions There are scholars who define culture in other ways than through shared values or outcomes. These tend to be more abstract, and in some cases esoteric or spiritual.5 Sheldon (1951) is a case in point. He splits culture into two categories — that of explicit and implicit culture: “Culture is a theoretical model, and the abstractions and principals from which it is made up are free creations of the mind. Some of these abstractions and principals deal with matters that are close to the minds of the individual culture bearers. This aspect of culture is usually called explicit culture or some similar term. Other aspects of culture, the implicit culture, are so generalized that in many cases the cultural bearers are unable to formulate them” (p. 39). Later, Hall (1976) extended the concept of unconscious culture which he defined as: those out-of-awareness cultural systems that have yet to be made explicit…Such systems have various features and dimensions which are governed by the order, selection, and congruence rules. These rules apply to the formative and active aspects of communications, discourses, perception (in all modalities), transactions between people, and the action chains by which humans achieve their goals in life. Significant portions of extension systems still function out of awareness. Much of the formation, development, use, and change processes of these extension systems, therefore, fall within the scope of out-of-awareness culture and not only exert a hidden influence on life but are subject to the same rules and laws…. Culture is therefore very closely related to if not synonymous with what has been defined as ‘mind’ (pp. 146147).

NEED FOR A THEORY-BASED INDIVIDUAL (VERSUS UNIVERSAL) CONCEPTUALIZATION AND MEASUREMENT OF CULTURE The prior section compared and contrasted different approaches to defining, and subsequently examining cultural differences. Of import is the lack of theory for the identification and examination of various proposed sets of cultural beliefs and values. Moreover, there is a link between what culture is held to be and the manner in which it is investigated. Two critical questions frame the theoretical conceptualization of culture, and subsequently, the manner in which it is investigated. First, is culture a structural phenomenon with properties irreducible to individuals, or is culture the sum of properties of individuals that constitute the cultural group? The answer to this question has significant implications for the level of analysis. Second, can culture be characterized by

18

universal dimensions or should it be understood in terms of the unique characteristics by a particular conceptualization of culture? Is there a way to define culture so as to avoid some of the complicating factors that have historically led to ambiguity and uncertainty in our research findings? Or are we destined to remain utterly vague about a concept that itself seems to resist precise definition? As discussed, culture may be a set of values, a set of outcomes, or a way of being. Whereas there is little doubt that these definitions vary widely, there is no doubt that, historically, the line of reasoning has tended to be atheoretical (conceptual rather than being derived from deductive processes). The epistemological basis for most definitions of culture has been formulated through argumentation or empirically derived and supported, as demonstrated by Hofstede’s work, one of the more frequently used frameworks for crosscultural research. Even though such empirically derived definitions of culture may adequately explain macro-level behavior, they often lack precision in explaining behavior at the individual level. Most such definitions rely on the assumption that an individual’s membership in a cultural group, such as their national culture, defines the nature of values they espouse. However, an individual’s values are influenced and modified by membership in other professional, organizational, ethnic, religious, and various other social groups, each of which has its own specialized culture and value set. Thus, individuals vary greatly in the degree in which they espouse, if at all, values dictated by a single cultural group, such as their national culture. Would a deductive, as opposed to inductive approach make sense in this context? There are socio-psychological theories that may have a direct or indirect bearing on what culture is as a scientific construct, and how it could be measured. In the current study, we offer one such theory as a candidate. There may be others that would work equally well or better. Our intention here is to stimulate thinking along these lines and not to offer the definitive solution.

SOCIAL IDENTITY THEORY AS A CANDIDATE THEORY BASE A theory that deals with the underlying mechanisms of group identification is Social Identity Theory (SIT) (Tajfel, 1970b, 1978). In SIT, individuals perceive that they belong to a group or not. If they belong to the group, they are part of the “ingroup”; if not, they are part of the “outgroup.” Other people are also classified in this same manner. The “ingroup” becomes a prominent way that individuals evaluate themselves and others, as to whether the group norms are adhered to, whether these norms are distinct from the norms of other groups, and so forth. The “ingroup” in this fashion becomes the “referent” for beliefs, attitudes, values, norms, and behavior (Deaux, 1984; Hogg, 1996; Tajfel, 1970a; Turner, 1982).

Jan - Mar 2002


Vol. 10, No. 1

The social identification process marks how similar individuals feel they are to others, as in organizational contexts such as the relationships between IT vendors and their clients (Gefen & Straub, 2001). The strength of the feeling of cultural similarity depends on how many of the characteristics individuals believe they have in common with others in the group. Thus, self-inclusive social grouping creates referents for country of origin (Hogg, 1996); support for a sports club (Hogg, 1996); and gender (Brown, 1996). The deeper level of this process is that perceptions of referent norms are internalized and incorporated into one’s sense of self, a process Turner (1982) calls Referent Informational Influence (RII). Internalization goes beyond what others do and say and how we respond to them, however. There is a proactive psychological process known as social categorization in which we attribute values to our own and others’ future actions. This attribution tells us that as members of the “ingroup,” we and others in the group act in certain ways and think in certain ways. It is what defines us, in a real sense (Brown, 1996; Hogg, 1996; Tajfel, 1978). Professors, for instance, are believed to feel, think, and act in certain ways, and identification with this group determines at least some of the attitudes and behaviors of members of their social group, or culture. SIT has been examined in many environments, including IT studies of the cultural similarity between vendors and clients (Gefen & Straub, 2001). Their argument is that client attitudes toward software are related to the “cultural similarity” they feel toward the vendor. The more alike clients find themselves to be to the vendor support personnel, the more positive they will be toward the software products associated with these vendors. Gefen and Straub (2001) found strong empirical evidence for SIT as a relevant theory base for cultural similarity. The use of SIT to theoretically enhance definitions of culture is consistent with the large body of work on shared values as a cultural foundation, since SIT is also value based. However, it departs and augments the explanatory power of earlier definitions for two reasons. First, it provides a theoretical framework that can explain the influence of several cultural layers (the “virtual onion”) on behaviors and outcomes; and second, it acknowledges that members of a cultural group vary in the extent to which they hold the values of the group (if at all).

IMPLICATIONS FOR THEORY-BASED MEASUREMENT OF CULTURE Assuming that SIT is a reasonable theory base from which to assess culture, how shall we proceed to measure it? Can culture even be measured? If it is unconscious, then it is embedded in values and behaviors and difficult to tease out. Use of the method of comparison and contrast is helpful in this aspect. This method is one of the primary scientific tools in robust positivist designs (Cook & Campbell, 1979).

In our view, instead of assuming that individuals espouse certain cultural values based on membership in a cultural group, individuals need to be queried as to the extent to which their values are similar to those of others in relevant cultural groups, and queried as to whether the outcomes of those values are perceived to be similar. This approach is also a culture-specific approach, as advocated by Hill et al. (1998,1994). A viable format is to elicit responses by embedding specific cultural values in scenarios along with other “distractor” variables (see Straub, Loch, and Hill, (2001), for more details of the approach). Thus, we advocate querying individuals about the layers of the onion that make up their overall “culture.” Clearly, the set of sub-cultures that can be investigated is limited, but IT researchers can diffuse this risk to a certain extent by exploiting their knowledge of the people being studied, and limiting the set of sub-cultures that are queried. In fact, culture must be measured at an individual level even though it is assumed that it is a group-level phenomenon. Thus, while Kroeber (1952) asserts that culture “quickly tends to become supra-personal and anonymous” (p. 102), it can only manifest itself through the individual. There is no way to query or probe into the collective unconscious of an entire culture. We argue, therefore, that the individual unit of analysis is both appropriate and meaningful. Once the individual level data is aggregated, it will also be possible to assert that certain cultural characteristics do or do not belong to certain cultures. A sample size as large as Hofstede’s would be desirable for such an assertion, but not necessary if stratified sampling is utilized. Our proposed approach explicitly recognizes that different layers of culture can interact and mesh in complex ways. We believe that such layers may influence particular behaviors differently for individuals or groups with different backgrounds (Evaristo, Karahanna et al. 2000). The layers are permeable, and their thickness signals the strength of the value held by the individual. We chose a “dynamic onion” metaphor to describe a situation where layers closest to the core are more salient, but understanding that inner layers occasionally exchange places with outer layers due to external circumstances. Measures of a person’s social identification with others who are presumed to be members of a given culture could serve as surrogates for an individual’s cultural ties. Questions related to values important to the study would be most critical, since, as Murdock’s (1965) values list suggests, it is highly unlikely that a researcher could very often query individuals about all of their cultural values. This implies that each study needs to determine the specific sub-cultures (e.g., national, organizational, professional) that are pertinent to a specific behavior and identify the values espoused by these cultures (Evaristo, Karahanna et al. 2000). For example, researchers who have focused on work- and task-related cultural values, such as Hofstede (1991), would test on values that are more



19

aligned with what takes place in the workplace. Organizational culture may be critical, especially when these values are distinct from ethnic and national values. It would be useful to create instrument items that exploit the learning that has accumulated in this arena (Hofstede, Neuijen, Ohayv, & Sanders, 1990). Our “virtual” onion metaphor assumes that the layering of the onion that makes up a person’s culture is not a permanent and immutable set of relationships. In certain circumstances, the sub-culture associated with religious values is most important. In other circumstances, the values that are most critical to an individual’s thinking or behaviors are his/her corporate sub-cultural values. While a person may be a cultural composite of only certain elements, he or she attributes importance to these sub-cultures according to changing conditions and stresses. The approach proposed has another pertinent practical advantage: improving the internal validity of studies that adopt it. Much current cross-cultural research (Evaristo, Karahanna et al. 2000) is performed by studying a given behavior, belief or attitude without employing corresponding and concurrent measurements of culture. Near the end of the scholarly exposition, conclusions are typically drawn between one’s measurement and cultural dimensions proposed by studies 10 to 30 years old. These studies also develop generalizations for a given group within a country, therefore opening themselves to several criticisms. First, the group being studied may have self-selected into a particular function (e.g., programmers) and such self-selection could override some of the earlier group characteristics. Second, national groups change over time. Therefore, when current studies try to compare two or more cultures which may have changed in sundry directions (potentially bringing them closer or farther apart on the studied cultural dimensions), researchers are introducing confounds, potentially rendering culturally-based conclusions meaningless. The development of new theoretically-based cultural measurement would allow future cross-cultural research endeavors to avoid these problems.

INSTANTIATING THE VIRTUAL ONION MODEL Offered tentatively, an instantiation of theory-based instrumentation of culture could help to flesh out the concepts, constructs, and metrics we are positing. The context is a study of Arab culture, specifically the influence of the sense of time in the Arab world and its influence on IT diffusion. As Straub, Loch, and Hill (2001) point out, there is good evidence that Arabs perceive time differently from other cultures, as, for example, Americans. Arabs tend to be polychronic, to see many strands of time moving at once, and to parallel-process work and tasks (Hall, 1984; Hall & Hall, 1990; Hall, 1976). Taking a culture-specific stance, an IT researcher would

20

be well served to measure the strength of particular cultural values held by the study participants. In cross cultural studies in Japan, it is very common to assume that all Japanese demonstrate the universal cultural characteristic of collectivism (Hofstede, 1980); yet we know that there will be variation in the strength of this characteristic and, in the case of highly entrepreneurial Japanese, it may not manifest itself at all. To generalize cultural characteristics across an entire nation of people is to flirt dangerously with what Robinson (1950) calls the “ecological fallacy,” where stereotypes are substituted for individualistic and idiosyncratic traits. Therefore, an in situ measurement of culture is appropriate. Cultural values of an individual can be calibrated by the extent to which a study participant identifies with a social group in certain settings. The settings can be stimulated through such methods as policy scenarios. Policy capturing offers the advantage of having respondents evaluate a set of feasible real world alternatives. A methodological technique reviewed by Webster and Trevino (1995), it has been used in IS and management research (Daft, Lengel, & Trevino, 1987; Straub & Karahanna, 1998; Straub et al., 2001; Zmud, Lind, & Young, 1990). The scenario and metrics below are merely illustrative of how a theory-based approach to measuring a specificcultural characteristic could be operationalized. It is not meant to be a considered attempt at possible scale items. Please note that the Arab sense of time is embedded in the scenario by showing how the top manager is able to parallel process several task at the same time (Straub, Loch, and Hill, 2001). Scenario An associate of a top manager, with whom you are having a formal meeting, walks into the office and soon there is a side meeting taking place between the associate and the top manager. The top manager’s other assistant engages you in further discussion until the top manager is able to resolve the other matter and return to the discussion. The side meeting transpired over a 10 minute period. Which sub-cultures should respondents be queried about with respect to this setting? Sub-cultures for ethnicity, nationality, religion, organizations, urban-rural, etc. are some of the social groups that many cross-cultural researchers are interested in, and are very likely relevant in this cultural setting. The key to creating an adequate surrogate for culture, we feel, is to measure the strength of identification with these salient social groups. Item x. In the context of the scenario you just read, to what extent do you identify with the following groups in determining how you feel about the top manger’s actions? Please use the following scale: 1 = Not at all, 2 = To a slight extent, 3 = To a moderate extent, 4= To a large extent, 5= To a very large extent.

Jan - Mar 2002


Vol. 10, No. 1

• • • • • • • • • • •

Members of your organization Members of your profession or occupation Devout Muslims Egyptians Arabs Devout Coptic Christians City-dwellers Europeans Educated persons Well-to-do persons Villagers

generalizability of the approach across cultural features, across populations, and across technologies (in the case of IT research). Having culture measured on scales rather than via assumption is a stronger scientific test and one might predict that some relationships that were not significant in past work would suddenly become so, once culture was more carefully delineated.

CONCLUSION

Assigning values to each of these sub-cultures, a respondent is, in effect, indicating which group values are most personally relevant in this context. Thus, these values resemble layers of an onion, with the most important values being closest to the core, which are the highest scored values. One may expect the extent to which one identifies with a particular behavior or circumstance (i.e., the importance of the value) will vary across context. Thus, for each individual, the layers of the onion do not have a fixed sequential order. Rather, they resemble a virtual onion in which layers exchange places depending on the specific behavior (Evaristo, Karahanna et al. 2000). The thickness of each layer signals the level of similarity between the individual and the cultural group that espouses that specific value. Finally, aggregating responses across respondents enables researchers to infer the values espoused by a certain cultural group. Measuring the impact of culture on IT effects may also be complicated by the extent to which an individual identifies with persons in the scenario and this, therefore, may also be probed, perhaps even prior to the various social group identifications immediately above. Item y. To what extent do you identify with the manner and actions of the top manager? Using the [semantic differential] scales below, place an “X” to represent the extent to which you identify with the top manager. • Seems perfectly usual ................. Seems unusual to me. • Seems polite ................................ Seems impolite • Seems acceptable ........................ Seems unacceptable. • Makes me identify ...................... Makes me not identify with the top manager .................. with the top manager.

Global information management researchers frequently examine the issue of culture and how it affects the international deployment of information systems (Gallupe & Tan, (1999). But cross-cultural studies are rife with problems, some of which hinge, seemingly, on the inability of the field to operationalize “culture,” even provisionally. An additional complication lies in the view that there is homogeneity in the espoused cultural values across all individuals in a specific culture. We propose that a possible solution to the problem is: (1) to adopt a theory-based conceptualization and measurement of an individual’s culture and (2) to measure the strength of particular cultural features as part of the data gathering in positivist research. Thus, we adopt a more complex view of culture that avoids reliance on assumed espoused cultural values. This view acknowledges the multiplicity of diverse cultural influences received by an individual which contributes to a unique idiosyncratic set of values. Towards this end, social identity theory provides a useful, candidate theoretical framework to identify, integrate, measure, and aggregate these influences. The implications of accepting this theory-based definition of culture are apparent. There are methodological implications and practical implications. Methodologically, researchers will need to inquire as to the identification priorities of respondents and subjects in addition to the context area queries or domain area queries that are the main point of study. From a practical point of view, the collection of data in any group may be drawn from numerous cultures and subcultures, some of which will be overlapping. Practically speaking, much future cross-cultural work will have to be flexible, inventive and adaptable to accommodate this fluidity in the sampling procedures and the operational framing of culture that emerges.

ENDNOTES

NEXT STEPS A fruitful, future scientific project would be to create a cultural instrument and examine the extent to which various cultures and sub-cultures could be teased out of the data. Nomological validity tests could be implemented to ensure that the cultures being identified were, indeed, influencing outcomes in an expected manner. Rigorous validation would help to establish the viability of the approach. Follow-up work by researchers could also test the

1 Hofstede (1984) defines values as “a broad tendency to prefer certain states of affairs over others” (p.18). 2 This characterization of individualism versus collectivism is essentially the same as that of Hofstede. Universalism implies a belief that something that is “true” and “good” can be universally defined, and therefore applicable everywhere. Particularism, on the other hand, prevails with a belief that unique circumstances and relationships determine what is “right” and “good.” In neutral cultures, people are more reserved, keeping emotions from being



21

out of the control in the workplace – which would be considered unprofessional. Affective cultures tend to display emotions more openly in all venues, and may consider a neutral culture as intentionally deceitful. In the specific and diffuse relationships dimension the idea is that an individual may have an inner private space and a larger public space. The difference between specific and diffuse cultures is the relative size of these “spaces.” In a specific culture, there is a tendency to have larger public areas and smaller private areas. In achievement-based cultures, people are evaluated based on their performance in their functional role. In ascriptive cultures people who naturally have certain attributes are admired and therefore gain status. Internal versus external orientation refers to the idea that there are two guides to action. Action can either be determined by inner-directed judgments, decisions, and commitments or by the signals, demands, and trends of the outside world. Time perspective refers to the idea that cultures can view time in two ways. Activities can be done as quickly as possible in the shortest possible sequence of passing time or synchronized so that completion is coordinated. 3 Laurent (1991) found, however, that corporate culture was skin deep in comparison to the intensity of national culture. 4 Cultural patterns, value orientations, and pattern variables are used interchangeably. 5 Although not covered here in greater depth, Herskovits (1955) offers a definition that is neither complex nor abstract: “The man-made part of the human environment” (p. 305). As might be expected, Skinner (1971) viewed culture as a complex series of reinforcement contingencies moderated by a schedule of rewards.

REFERENCES Brown, R. J. (1996). Tajfel’s Contribution to the Reduction of Intergroup Conflict. In W. P. Robinson (Ed.), Social Groups and Identities: Developing the Legacy of Henri Tajfel. Oxford, UK: Butterworth-Heinemann. Cook, T. D., & Campbell, D. T. (1979). Quasi-Experimentation: Design and Analysis Issues for Field Settings. Boston: Houghton Mifflin. Daft, R. L., Lengel, R. H., & Trevino, L. K. (1987). Message Equivocality, Media Selection, and Manager Performance: Implications for Informational Systems. MIS Quarterly, 11: 355-366. Deaux, K. (1984). From Individual Differences to Social Categories. Analysis of a Decade’s Research on Gender. American Psychologist, 39(2): 105-116. Erez, M., & Early, C. P. (1997). The Transplanted Executive. New York, NY: Oxford University Press. Evaristo, J. R., Karahanna, E., & Srite, M. (2000). Culture and Information Systems. Proceedings of the Global Information Technology Management (GITM) World Conference, Memphis, TN. Ford, C. S. (1942). Culture and Human Behavior. Scientific Monthly, 55: 546-547. Gallupe, R. B., & Tan, F. B. (1999). A Research Manifesto for Global Information Management. Journal of Global Information Management, 7(3): 5-18. Geertz, C. (1973). The Interpretation of Cultures. New York, NY: Basic Books. Gefen, D., & Straub, D. W. (2001). Perceived Cultural Similarity and Technology Acceptance: Georgia State University Working Paper.

22

Glenn, E. (1981). Man and Mankind: Conflict and Communication Between Cultures. Norwood, NJ: Ablex Publishing Corporation. Hall, E. (1984). The Dance of Life: Another Dimension of Time. New York: Doubleday. Hall, E., & Hall, M. R. (1990). Understanding Cultural Differences. Yarmouth, Maine: Intercultural Press, Inc. Hall, E. T. (1976). Beyond Culture. Garden City, NY: Doubleday. Hampden-Turner, C., & Trompenaars, F. (1993). The Seven Cultures of Capitalism. New York, NY: Doubleday. Herskovits, M. J. (1955). Cultural Anthropology. New York, NY: Alfred A. Knopf. Hill, C., Loch, K., Straub, D. W., & El-Sheshai, K. (1998). A Qualitative Assessment of Arab Culture and Information Technology Transfer. Journal of Global Information Management, 6(3 (Summer)): 29-38. Hill, C. E., Straub, D. W., Loch, K. D., Cotterman, W. W., & El-Sheshai, K. M. (1994). Arab Culture and the Transfer of Information Technology: A Culture-Driven Model. Paper presented at the International Federation of Information Processing, WG 9.4 Conference, Havana, Cuba. Hofstede, G. (1980). Culture’s Consequences: International Differences in Work-Related Values. Beverly Hills, CA: Sage. Hofstede, G. (1984). Culture’s Consequences: International Differences in Work-Related Values. Newbury Park, CA: Sage. Hofstede, G. (1991). Cultures and Organizations: Software of the Mind. London: McGraw-Hill. Hofstede, G., & Bond, M. H. (1988). The Confucius Connection: From Cultural Roots to Economic Growth. Organizational Dynamics, 16(4): 5-21. Hofstede, G., Neuijen, B., Ohayv, D. D., & Sanders, G. (1990). Measuring Organizational Cultures: A Qualitative and Quantitative Study Across Twenty Cases. Adminstrative Science Quarterly, 35: 286-316. Hogg, M. A. (1996). Social Identity, Self-Categorization, and the Small Group. In E. H. W. J. H. Davis (Ed.), Understanding Group Behavior. Mahwah, NJ: Lawrence Erlbaum Associates. Kluckholn, C. (1951). The Study of Culture. In D. Lerner, & H. D. Lasswell (Eds.), The Policy Sciences. Stanford, CA: Stanford University Press. Kluckholn, C., & Leighton, H. D. (1946). The Navaho. Cambridge, MA: Harvard University Press. Kluckholn, C., & Strodbeck, F. L. (1961). Variations in Value Orientations. Evanston, IL: Row and Peterson. Kroeber, A. L. (1949). Man and His Works. New York, NY: Alfred A. Knopf. Kroeber, A. L. (1952). The Nature of Culture. Chicago, Il: Chicago University Press. Kroeber, A. L., & Kluckholn, C. (1952). Culture: A Critical Review of Concepts and Definitions. New York, NY: Vintage Books. Lachman, R. (1983). Modernity Change of Core and Peripheral Values of Factory Workers. Human Relations, 36: 563-580. Laurent, A. (1991). In N. J. Adler (Ed.), International Dimensions of Organizational Behavior. Boston, MA: PWS-Kent Publishing. Martin, J. (1992). Cultures in Organizations: Three Perspectives. Oxford, UK: Oxford University Press. Moran, R. T., & Stripp, W. G. (1991). Successful Interna-

Jan - Mar 2002


Vol. 10, No. 1

tional Business Negotiations. Houston, TX: Gulf Publishing Company. Morris, C. W. (1956). Varieties of Human Value. Chicago, IL: Chicago University Press. Murdock, G. P. (1965). Culture and Society. Pittsburgh, PA: The University of Pittsburgh Press. Parsons, T., & Shils, E. E. (1951). Toward a General Theory of Action. New York, NY: Harper & Row Publishers. Robinson, W. S. (1950). Ecological Correlations and the Behavior of Individuals. American Sociological Review, 15: 351157. Rokeach, M. (1973). The Nature of Human Values. New York, NY: Free Press. Schein, E. H. (1985). Organizational Culture and Leadership: A Dynamic View. San Francisco, CA: Jossey-Bass. Schein, E. H. (1999). The Corporate Culture Survival Guide: Sense and Nonsense about Culture. San Francisco, CA: JosseyBass Publishers. Sheldon, R. C. (1951). Some Observations on Theory in the Social Sciences. In T. Parsons, & E. A. Shils (Eds.), Toward a General Theory of Action. New York, NY: Harper & Row Publishers. Skinner, B. F. (1971). Beyond Freedom and Dignity. New York, NY: Alfred A. Knopf. Straub, D. W., & Karahanna, E. K. 1998. Knowledge Worker Communications and Recipient Availability: Toward a Task Closure Explanation of Media Choice. Organization Science, 9(2):

160-175. Straub, D. W., Loch, K., & Hill, C. (2001). Transfer of Information Technology to the Arab World: A Test of Cultural Influence Modeling. Journal of Global Information Management, forthcoming. Tajfel, H. (1970a). Aspects of National and Ethnic Loyalty. Social Science Information, 9(3): 119-144. Tajfel, H. (1970b). Experiments in Intergroup Discrimination. Scientific American, 233(5): 96-102. Tajfel, H. (1978). Differentiation Between Social Groups. Cambridge, UK: Cambridge University Press. Triandis, H. C. (1972). The Analysis of Subjective Culture. New York, NY: Wiley. Trompenaars, F. (1993). Riding the Waves of Culture. London, UK: The Economist Books. Turner, J. C. (1982). Towards a Cognitive Redefinition of the Social Group. In H. Tajfel (Ed.), Social Identity and Intergroup Relations: 15-40. Cambridge, UK: Cambridge University Press. Tylor, E. B. (1871). Primitive Culture. New York, NY. Webster, J., & Trevino, L. K. (1995). Rational and Social Theories as Complementary Explanations of Communication Media Choices: Two Policy Capturing Studies. Academy of Management Journal, 38: 1544-1572. Zmud, R. W., Lind, M. R., & Young, F. W. (1990). An Attribute Space for Organizational Communication Channels. Information Systems Research, 1(4): 440-457.

Detmar W. Straub is the J. Mack Robinson Distinguished Professor of Information Systems at Georgia State University, Detmar has conducted research in the areas of e-Commerce, computer security, technological innovation, and international IT studies. He holds a DBA in MIS from Indiana and a PhD in English from Penn State. He has published over 80 papers in journals such as Journal of Global Information Management, Management Science, Information Systems Research, MIS Quarterly, Organization Science, Communications of the ACM, Journal of MIS, Information & Management, Communications of the AIS, Academy of Management Executive, and Sloan Management Review . Karen D. Loch is an Associate Professor and Director of the Institute of International Business in the J. Mack Robinson College of Business at Georgia State University, Karen’s current research interests span international IT studies, security and ethical concerns, and global ecommerce. She holds a Ph.D. in MIS from the University of Nebraska. Loch has published in journals such as Journal of Global Information Management, Communications of the ACM, MIS Quarterly, Information Systems Journal, Academy of Management Executive, Data Base, and Journal of Informatics and Education. She is a co-editor and contributor of Global Information Technology Education: Issues and Trends. She serves as Associate Editor for Journal of Global Information Management, The Journal of Global Information Technology Management, and as review board member for Information Resources Management Journal. Elena Karahanna is an associate professor of MIS at the Terry College of Business, University of Georgia. Her current research interests include the adoption, mandatory adoption, implementation, use, and infusion of information technologies, the effect of media choice and use on individuals and organizations, and cross-cultural issues. Her work has been published in Management Science, Organization Science, MIS Quarterly, Data Base, the Journal of Organizational Computing and Electronic Commerce, Information & Management, and elsewhere. Roberto Evaristo is an assistant professor at the University of Illinois, Chicago. He is currently involved in several projects related to the management of distributed projects in virtual organizations, with work done in Japan, USA and Europe. He has published in many outlets such as the International Journal of Project Management, Database, Information Technology and People, Journal of the Association for Global Strategic Information, Journal of International Information Management, International Information Systems, and Competitive Intelligence Review. Mark Srite is an assistant professor of MIS at the University of Wisconsin -Milwaukee. His current research interests include the acceptance, adoption, and use of information technologies, cross-cultural IT issues, and group decision-making. His work has been published in Decision Support Systems, and elsewhere. Mark Srite is an assistant professor of MIS at the University of Wisconsin -Milwaukee. His current research interests include the acceptance, adoption, and use of information technologies, cross-cultural IT issues, and group decision-making. His work has been published in Decision Support Systems and elsewhere.



23