Sociology Of Groups Research Paper

These small-scale collective phenomena, in turn, influence the behavior of individuals and their interactions (see Fig. 1). Some sociological work concentrates on relation 1, some on relation 2, and some on 1 and 2, and other combinations. Relation 3b allows explicit feedback mechanisms involving both levels (as in expectation states theory, see below). Psychological studies often focus on the direct influence of interaction (or the presence of others) on attitude change, productivity, satisfaction, self-esteem, depression, etc., occasionally allowing for feedback mechanisms (3a and 3b). The extra arrows in Fig. 1(i) indicate exogeneous factors (such as genetic dispositions, rules, established categories, status differences).

1. Historical Background Of Small Group Research

The scientifically serious investigation of groups in the sociological tradition started with anthropological studies of tribes, clans, and families in the nineteenth and early twentieth centuries. Sumner, Simmel, and Cooley were early codifiers of that kind of research. The most important turn in this development occurred in the 1930s. Within the span of a few years, four great books appeared which had two very important messages in common: that groups are constituted by a cognitively mediated functional interdependence between individuals, i.e., people who are dependent on each other for achieving a goal interact, and something important happens to their minds and the way their minds operate when they interact. In addition, the actual situation in which their action is placed matters most for behavior (as opposed to traits or other cross situational constants). In 1934 Mead’s (then already posthumous) Mind, Self and Society appeared. Two years later, Lewin’s Principles of Topological Psychology and Sherif’s The Psychology of Social Norms came out. A year later, Parsons’ The Structure of Social Action made its appearance. Merton’s early work on reference groups was also quite influential.

From this time, group studies continued in two traditions—experimental and field studies—and both traditions bore the mark of this convergent conception of groups. Under the leadership of Lewin and Sherif, ‘group dynamics’ as the experimental study of groups was then called, achieved great heights. Field studies had Mayo’s Hawthorne studies and Whyte’s famous Street Corner Society as models. Both traditions owe their success to the simultaneous consideration of the three fundamental interdependencies in groups: functional (related to goal achievement), cognitive (related to common categorizations, beliefs, and knowledge) and structural (related to the configurations of social relations) (see Lindenberg 1997 for more detail on this and other aspects of this research paper).

However, the precise nature of the interdependencies was not yet well investigated and in the late 1950s and early 1960s, specialized groups of studies on each of these interdependencies became dominant, with interesting crossovers between psychology and sociology. Homans (1950, 1958), a sociologist, combined the field and the experimental traditions in the study of functional interdependencies in exchange relations. Inspired by Homans, two psychologists, Thibaut and Kelley (1959), worked out the study of small groups as the study of functional interdependencies. Their book is to this day one of the best on the subject. It influenced Homans’ further work on the subject (Homans 1961). Cognitive interdependencies were elaborated by Sherif and turned into a major strand of group research called ‘social identity theory’ by Tajfel (1982) and Turner (1982), both psychologists but very influential among sociologists and in turn influenced by sociological symbolic interactionism. In a related development, Moscovici, a psychologist, elaborated Durkheim’s theory of collective representations (see Flick 1998). Structural interdependencies were more formally elaborated by the psychologist Cartwright and the mathematician Harary and turned into a full-blown network approach by sociologist White and his followers (see White et al. 1976).

2. Bringing Different Kinds Of Interdependencies Back Together Again

This section deals briefly with the field study approach and in some more detail with experimental studies. No attempt is made to be complete. Rather, the overview is selective. The selection criteria are first, that the studies should be substantively important. Second, they must mainly focus on group processes comprising small groups (i.e. face-to-face interaction with at least one common goal), small networks, and face-to-face encounters. Third, the studies must be seriously concerned with the integration of at least two of the three fundamental interdependencies. Fourth, they must cluster around a program of investigation with some continuity.

With regard to field studies, work inspired by Goffman was chosen. Goffman’s studies provided a new integration for field studies in sociology (excluding Mayo-type studies). With regard to experimental (and simulation) studies, expectation states theories, various groupings of social exchange theories, and studies of friendship and cognitive networks were selected. Studies of social dilemmas are covered elsewhere. Helpful discussions of a broader scope can be found in Cook et al. (1995) and Foschi and Lawler (1994). Due to space restrictions, the reader will at times be referred to bibliographies and collections rather than individual works.

2.1 Fieldwork Studies

From the beginning of the 1960s, fieldwork studies on face-to-face groups have received their main theoretical guidance from Goffman (1961). He also put short-lived groups, or ‘encounters’, on the map. There is a relatively small but very productive circle of contributors to this kind of research (a few important examples are Anderson 1978, Cahill and Lofland 1994, Fine 1987, MacLeod 1995). These studies are concerned with conformity, status, group divisions, identities, and their combination in impression management. They are rich in descriptive detail. Aspects of functional and cognitive interdependencies are clearly present in Goffman’s work.

2.2 Experimental Studies

2.2.1 Expectation States Theories. Generic expectation states theory is concerned with the formation and maintenance of power and prestige structures in face-to-face task groups (see literature in Berger and Zelditch 1993). Following the tradition of Bales in the 1950s, power and prestige are conflated and their structure is conceived in terms of behavioral aspects (opportunities and influence) rather than the distribution of scarce resources. People are higher on the power and prestige structure, the more opportunities to act they receive, the more often they take advantage of these opportunities, and the more often they win disagreements if and when they occur.

The basic mechanism that creates this structure is pairwise interaction (between, say, Ego and Alter) that is governed by performance expectations. It is the (unobservable) structure of these performance expectations which keeps generating the observable power and prestige structure. When Ego has positive performance expectations towards Alter, he or she will give more action opportunities to Alter and Alter will more likely take up each opportunity offered. Alter has influence over Ego. It is assumed that there is an evaluation bias. Ego is more likely to judge Alter’s performance to be positive if the performance expectation was positive and vice versa. And if Ego nonetheless voices disagreement, he or she is more likely to eventually yield to Alter if the performance expectation was positive to begin with. Such evaluation biases greatly enhance the stability of the power and prestige structure (provided the task endures).

The theory is meant to apply to task groups. For this reason, functional interdependence is by definition the starting point. Here, a joint task and face-to-face interaction is what makes a collection of individuals into a group. Each group member’s contribution influences a group score (mostly outcome interdependence, say, linking bonus to agreement in answers). There are also cognitive aspects. For the theory to apply, group task achievement must be a salient goal in the interaction with other members (‘collective orientation’). The central cognitive element in the theory is a performance expectation. It presupposes other cognitive elements, such as goal criteria and causal beliefs about how likely a particular characteristic or behavior is to lead to success or to failure. Cognitive interdependence comes about when the performance expectations (and, prior to that, the goal criteria and causal beliefs) are being shared. Given the evaluation bias, the interdependence means that performance expectations of different group members interlock (see Moore 1985).

Functional and cognitive interdependencies are closely interlinked in this theory. This link is made even stronger through extensions of the theory to include external status characteristics (see Berger and Zelditch 1993) and the generation of status characteristics from nominal distinctions (see Webster and Hysom 1998). Task groups are embedded in a wider culture in which such status and competence beliefs exist (such as ‘men are better at mathematics’). People seem to follow the maxim ‘when in doubt, assume that a salient characteristic is task-relevant’ (the so-called ‘burden of proof principle’). With regard to gender, this favors men in general tasks as well as in ‘masculine’ tasks, and it favors women only in stereotypically ‘feminine’ tasks. Through the link of performance expectations with worthiness, there is also a link of expectation structures to conceptions of just rewards, which is worked out in the ‘status value theory’ of distributive justice (see Berger and Zelditch 1993).

Ridgeway has shown that, in turn, such cultural beliefs can be explained by the diffusion of status belief-generating mechanisms in small groups. When interaction partners in task groups are doubly dissimilar in that they differ in valuable resources (say high pay level) and in a salient nominal distinction (such as race or gender), then a process of misattribution will causally link having the valuable resource with the nominal distinction and eventually with assumptions of competence, worthiness, and esteem. The group consensus on this attribution generalizes to the idea that ‘most people’ believe it to be so (Ridgeway et al. 1998). Group members are likely to transfer such beliefs to other group contexts, thereby spreading and reinforcing them across groups and turning them into cultural status beliefs. Similar mechanisms are now invoked to explain legitimation and delegitimation of power and prestige structures in groups (see Berger et al. 1998).

2.2.2 Theories Of Exchange In Networks. Theories of social exchange by Blau (1964), Homans (1950), and Thibaut and Kelley (1959) have given rise to a number of theories on the influence of social structure on the distribution of exchange ratios or benefits within groups. The results of these theories are also relevant to the explanation of network effects on power advantages on a macro level (see, e.g., Burt 1992). Paradigmatic for these theories is Emerson’s (1976) concept of ‘network connection,’ which points to externalities between exchange relations. Two exchange relations are connected if and only if the exchange in one relation affects the exchange in the other (in terms of frequency and value). Thus, in an exchange network, the specific distribution of ties can have a strong impact on the distribution of benefits, but this influence depends on the nature of the network connections. For example, if there is a rule that one may exchange only with one other partner then the network is negatively connected. Division of labor creates positively connected networks in which exchange with one stimulates exchange with the other(s). Negative connections are easiest to trace both theoretically and empirically, and thus the bulk of research in this area is on negatively connected networks (an important exception is Burke (1997), who has developed an interesting link between network exchange and identities).

One prominent version of such a theory is exchange network theory (Cook and Yamagishi 1992). The distribution of benefits (or rates of exchange) within a network is explained on the basis of power (‘power/dependence’). In positively connected networks, the power process is determined by brokerage, i.e., by the ability to connect resources of various actors, and by the value of the focal resource to the actors. In a negatively connected network, power is based on the value and degree of availability of a focal resource from alternative sources. The more value Ego gets from an exchange with Alter compared to the best available exchange alternative, the more dependent Ego is on Alter and the more power Alter has over Ego. With repeated exchange in such negatively connected networks, the exchange ratio is predicted to shift toward the point at which neither of them could improve his or her outcome by exchange with another partner. Power advantages are thus structurally induced, given the ‘connectedness’ of the network.

A variant of this theory for coercive power has been elaborated by Molm (1997). Interestingly, Molm predicts and finds that power based on the ability to inflict damage on others is not structurally but strategically induced in order to extract rewards from the other. This opens exchange theory to a more explicit consideration of rational strategies, especially with regard to the balance between incentive to coerce and fear of retaliation. Lawler’s work is also relevant to this topic (see, e.g., Lawler 1992a).

Elementary theory (ET, sometimes also called ‘network exchange theory’) is also rooted in the Homans/Emerson tradition and it also explains exchange ratios on the basis of structurally induced power (see Berger and Zelditch 1993, pp. 323). The major difference with exchange network theory is Two fold. First, in reference to Weber, ET considers resistance to the exercise of power more explicitly than exchange network theory. Second, ET uses experiments in which the degree of ‘exclusion’ (i.e., the degree of negative connectiveness) varies. This second possibility gives it more flexibility to approximate various real life phenomena. However, the way resistance and the representation of exchange are worked out imposes strong restrictions, which reduces real life application to some degree (see Van Assen 2001). More recently, interesting links of ET with expectation states theory (Willer et al. 1997) and with game theory (see Willer and Skvoretz 1997) have been developed.

An additional grouping of theories on exchange should be mentioned. It is also related to the Homans Emerson tradtion but uses different elaborations. There have been long-standing efforts to explain structural network effects on power and exchange rates by means of game theory (especially using the concept of the ‘core’) by Bonacich and co-workers (see, e.g., Bonacich 1998). See Friedkin (1995) for yet a different model and Fararo and Skvoretz (see Berger and Zelditch 1993, pp. 416) for a comparison of various models. There is also a productive development of models that use exchange in networks for the explanation of power processes in collective decision making, mostly in small groups (see Stokman et al. 2000). Because these models are not tied to a particular experimental paradigm, they are more open to the consideration of other exchange models and strategies (such as Coleman’s model; see Coleman 1990) and to an explicit consideration of strategies of persuasion and challenges that accompany exchange strategies.

2.2.3 Friendship And Cognitive Networks. In recent years, dynamic friendship networks have received an increasing amount of attention. A new line of research has opened up by taking relatively simple behavioral assumptions and tracing their structuregenerating consequences through simulation and, at times, also empirical research. What is new is that the interdependence of simultaneous friendship choices is modeled. If choices go on simultaneously in a group of strangers, what structure will emerge from the interdependence of choices under various choice preferences and constraints? How does ‘group’ size affect the resulting structure(s)? How does heterogeneity of the choosing individuals affect the resulting structure(s)? Preferences have been made more complex by stepwise adding status preferences to similarity preferences, and by explicitly introducing friendship dissolution to friendship formation. As to influence, friends do not just have a preference for similarity, they also become more similar. Paradigmatic for this kind of work is Stokman and Zeggelink (1996) and work presented in Doreian and Stokman (1997). See Van der Bunt (1999) for empirical research. Work linking this kind of dynamic approach to issues of cooperation is exemplified by Flache and Macy (1996) and Flache and Hegselmann (1999).

Networks are also cognitively represented which, in turn, contributes to the generation of joint cognitive and social structures. Research dealing with these aspects of networks link explicitly cognitive and structural interdependencies (see Carley 1986, Freeman 1994, Krackhardt 1987, Pattison 1994). Related research deals with the effects of cognitive interdependencies on learning, search, and general problem-solving strategies in groups and organizations (see Carley 1995, Edmonds and Dautenhahn 1999). Here functional interdependencies in terms of joint production are added explicitly (see also Troyer and Silver 1999). In the future, one can expect innovative impulses from these lines of research for sociological group research in general.

3. Evaluation

Sociological research on groups is lively, productive and, in the experimental fields, also cumulative. There is also a slow tendency to open up across the narrow borders of the research paradigms. However, there are a number of important shortcomings (see Lindenberg 1997 for more detail and for suggestions for new directions). First, field and experimental studies generally ignore each other. This is not a desirable state. Crossovers in the past have been very productive and should be encouraged. Second, the experimental paradigms carry such a strong weight that they hamper new theory formation. For example, in expectation states theories, the functional interdependence is experimentally restricted to outcome interdependence with regard to agreement. There is little or no task interdependence. In the exchange theories, exchanges are often limited to divisions of resource pools. Such exchanges lack many of the properties of the exchange of goods, information, and emotions. On the whole, there is little attention to the effects of joint production on expectations and exchange.

Third, functional interdependencies should have analytical primacy. This primacy derives from the fact that cognitive processes are strongly influenced by goals (see Gollwitzer and Moskowitz 1996) and structural effects only occur in interaction with goals (externalities) as has been seen in exchange network theory. Yet, in most exchange theories this primacy is lacking. Fourth, present research rarely includes ‘groupishness’ (i.e., categorization and framing effects) in its consideration of cognitive interdependencies (see Lawler 1992b as an exception). For example, in expectation states theory, ‘collective orientation’ is introduced as a scope condition, not as a variable. Related to this point is, fifth, that the distinction between small groups and small networks is nonexistent or ad hoc. There is little or no theory on the influence of groups on networks or vice versa, or on networks nested in groups. Sixth, the study of intergroup relations for the explanation of what goes on inside the group needs much more attention. A positive example of what is needed is Coleman’s classic study of the influence of interscholastic athletic competition on the local status structure in schools (see Coleman 1961).

Bibliography:

1. Anderson E 1978 A Place on the Corner. University of Chicago Press, Chicago
2. Berger J, Ridgeway C L, Fisek M H, Norman R 1998 The legitimation and delegitimation of power and prestige orders. American Sociological Review 63: 379–405
3. Berger J, Zelditch M (eds.) 1985 Status, Rewards, and Influence. Jossey-Bass, San Francisco
4. Berger J, Zelditch M (eds.) 1993 Theoretical Research Programs. Stanford University Press, Stanford, CA
5. Blau P 1964 Exchange and Power in Social Life. Wiley, New York
6. Bonacich P 1998 A behavioral foundation for a structural theory of power in exchange networks Social Psychology Quarterly 61: 185–98
7. Burke P J 1997 An identity model for network exchange. American Sociological Review 62: 134–60
8. Burt R S 1992 Structural Holes. The Social Structure of Competition. Harvard University Press, Cambridge, MA
9. Cahill S E, Lofland H (eds.) 1994 The Community of Streets. JAI Press, Greenwich, CT
10. Carley K 1986 An approach for relating social structure to cognitive structure. Journal of Mathematical Sociology 12: 137–89
11. Carley K 1995 Computational and mathematical organization theory: Perspectives and directions. Computational and Mathematical Organization Theory 1: 39–56
12. Coleman J S 1961 The Adolescent Society. Free Press, Glencoe, IL
13. Coleman J S 1990 Foundation of Social Theory. Harvard University Press, Cambridge, MA
14. Cook K S, Fine G A, House J S (eds.) 1995 Sociological Perspectives on Social Psychology. Allyn and Bacon, Boston
15. Cook K S, Yamagishi T 1992 Power in exchange networks: A power-dependence formulation. Social Networks 14: 245–66
16. Doreian P, Stokman F N 1997 Evolution of Social Networks. Gordon and Breach, Amsterdam
17. Edmonds B, Dautenhahn K (eds.) 1999 Special issue on social intelligence. Computational and Mathematical Organization Theory 5
18. Emerson R M 1976 Social exchange theory. Annual Review of Sociology 2: 335–62
19. Fine G A 1987 With the Boys. University of Chicago Press, Chicago
20. Flache A, Hegselmann R 1999 Rationality vs. learning in the evolution of solidarity networks. Computational and Mathematical Organization Theory 5: 97–127
21. Flache A, Macy M 1997 The weakness of strong ties: Collective action failure in a highly cohesive group. Journal of Mathematical Sociology 21: 3–28
22. Flick U (ed.) 1998 The Psychology of the Social. Cambridge University Press, Cambridge, UK
23. Foschi M, Lawler E J (eds.) 1994 Group Processes: Sociological Analyses. Nelson-Hall, Chicago
24. Freeman L C 1994 Interpersonal proximity in social and cognitive space. Social Cognition 12(3): 223–47
25. Friedkin N E 1995 The incidence of exchange networks. Social Psychology Quarterly 58: 213–22
26. Goffman I 1961 Encounters. Bobbs-Merrill, Indianapolis, IN
27. Gollwitzer P M, Moskowitz G B 1996 Goal effects on action and cognition. In: Higgins E T, Kruglanski A W (eds.) Social Psychology. Handbook of Basic Principles. The Guilford Press, London, pp. 361–99
28. Homans G C 1950 The Human Group. Harcourt, Brace & World, New York
29. Homans G C 1958 Social behavior as exchange. American Journal of Sociology 63: 597–606
30. Homans G C 1961 Social Behavior: Its Elementary Forms. Harcourt, Brace & World, New York
31. Krackhardt D 1987 Cognitive social structures. Social Networks 9: 109–34
32. Lawler E J 1992a Power processes in bargaining. The Sociological Quarterly 33: 17–34
33. Lawler E J 1992b Affective attachments to nested groups. American Sociological Review 57: 327–39
34. Lindenberg S 1997 Grounding groups in theory: Functional, cognitive, and structural interdependencies. Advances in Group Processes 14: 281–331
35. MacLeod J 1995 Ain’t No Making It. Westview, Boulder, CO
36. Molm L 1997 Coercive Power in Social Exchange. Cambridge University Press, Cambridge, UK
37. Moore J C 1985 Role enactment and self-identity. In: Berger J, Zelditch M (eds.) Status, Rewards, and Influence. Jossey-Bass, San Francisco, pp. 262–316
38. Pattison P 1994 Social cognition in context: Some applications of network analysis. In: Wasserman S, Galaskiewicz J (eds.) Advances in Social Network Analysis. Sage, Thousand Oaks, CA, pp. 79–109
39. Ridgeway C L, Heger Boyle E, Kuipers K J, Robinson D T 1998 Resources and interaction in the development of status beliefs. American Sociological Review 63: 331–50
40. Stokman F N, Van Assen M, Van Der Knoop J, Van Oosten R C H 2000 Strategic decision making. Advances in Group Processes 17: 131–53
41. Stokman F N, Zeggelink E P H 1996 ‘Self-organizing’ friendship networks. In: Liebrand W B G, Messick D M (eds.) Frontiers of Social Dilemma Research. Springer, New York, pp. 385–418
42. Tajfel H 1982 Instrumentality, identity and social comparisons. In: Tajfel H (ed.) Social Identity and Intergroup Relations. Cambridge University Press, Cambridge, UK, pp. 483–507
43. Thibaut J, Kelley H H 1959 The Social Psychology of Groups. Wiley, New York
44. Troyer L, Silver S D 1999 Institutional logics and group environments. Advances in Group Processes 16: 219–52
45. Turner J C 1982 Toward a cognitive redefinition of the social group. In: Tajfel H (ed.) Social Identity and Intergroup Relations. Cambridge University Press, Cambridge, UK, pp. 15–40
46. Van Assen M 2001 Two representations of negotiated exchange: A review and comparison. In: Van Assen M (ed.) Esssays on Actor Models in Exchange Networks and Social Dilemmas. Thela Thesis, Amsterdam
47. Van der Bunt G 1999 Friends by Choice: An Actor Oriented Statistical Network Model for Friendship Networks Through Time. Thela Thesis, Amsterdam
48. Webster M, Hysom S J 1998 Creating status characteristics American Sociological Review 63: 351–78
49. White H C, Boorman S A, Breiger R L 1976 Social structure from multiple networks: I. Blockmodels of roles and positions. American Journal of Sociology 81: 730–80
50. Willer D, Lovaglia M J, Markovsky B 1997 Power and influence: A theoretical bridge. Social Forces 76: 571–603
51. Willer D, Skvoretz J 1997 Games and structures. Rationality and Society 9: 5–35