Expectation States Theory
Expectation States Theory
EXPECTATION STATES THEORY
Expectation states theory is a well-established, ongoing research program investigating various aspects of group interaction. The focus is on small, task-oriented groups; the central interest is in both the processes through which group members assign levels of task competence to each other and the consequences this assignment has for their interaction. Originating as a single theory developed by Joseph Berger (1958), expectation states theory has grown to include various branches sharing a core of basic concepts, definitions, and propositions about group interaction processes, as well as methodological and metatheoretical assumptions. Thus, the program in fact contains not just one theory but several. (Unless otherwise specified, the expression "expectation states theory" refers here to the entire program rather than to any particular theory within it.) Expectation states theory has received strong support from extensive empirical research.
Two key concepts in the program are "status characteristics" and "performance expectations." A status characteristic is any valued attribute implying task competence. Such characteristics are viewed as having at least two levels (e.g., being either high or low in mechanical ability, being either male or female), one carrying a more positive evaluation than the other. They are also defined as varying from specific to diffuse, depending on the range of their perceived applicability. For instance, mechanical ability is usually considered to be relatively specific, or associated with well-defined performance expectations. Gender, on the other hand, is frequently treated as diffuse, or carrying both limited and general performance expectations. The "diffuseness" refers to the fact that since there is no explicitly set limit to the expectations, the characteristic is viewed as relevant to a large, indeterminate number of different tasks. Other attributes commonly treated as diffuse status characteristics are ethnicity, skin color, socio-economic class, level of education, organizational rank, age, and physical attractiveness.
Performance expectations link status characteristics to observable behavior. Thus, levels of these characteristics are associated with levels of competence and corresponding expectations that, in turn, determine what is known as "the power and prestige order of the group." This concept refers to a set of interrelated behaviors, namely, the unequal distribution in the offer and acceptance of opportunities to perform, the type of evaluations received for each unit of performance, and the rates of influence exerted among group members. Note, then, that performance expectations are distinguished from evaluations of units of performance: while the latter are evaluations of a single act, the former refer to the level of competence that a person is predicted to exhibit over a number of such performances. Once formed, expectations tend to be stable, since the behaviors that make up the power and prestige order of a group operate in a way that reinforces the status quo.
Research in expectation states theory follows a situational approach. Propositions are formulated from the point of view of an individual (also referred to as "self" or "focal actor") who performs a given task with a partner ("other"). Performance expectations are therefore relative to a specific pair of actors, and an individual is not seen as holding high (or low, or medium) expectations for self but, rather, an "expectation state" that is defined in terms of self and other. In other words, the focal actor is said to hold expectations for self that are either higher than, lower than, or equal to, the expectations he or she holds for the partner. Such expectations are also conceptualized as relative to a particular situation. Thus, for example, with a different partner or in the context of a different task, a person's expectation state could vary. It should also be noted that an expectation state is a theoretical construct, not an observable phenomenon. In particular, although expectations are seen as reflecting a person's beliefs about the distribution of task competence in the group, they are not assumed to be self's conscious calculations of advantage (or disadvantage, or equality) in this respect. Rather, what is proposed are models to be used to predict the focal actor's behavior, and this person is to be seen only as if he or she performed the operations specified in the models.
Propositions in expectation states theory also have been formulated within well defined scope conditions, or clauses specifying the limits within which the propositions apply. Since scope conditions are part of the theory, they are expressed in abstract terms (as are the key concepts of "status characteristic," "performance expectations," and "power and prestige order"). Three important scope conditions are that the focal actor is assumed to: (1) value the task in question (he or she views it as having two or more possible outcomes, none due exclusively to chance, and considers one of these to be success and the other failure); (2) be task-oriented (motivated to do the task well); and (3) be collectively oriented (prepared to accept the partner's ideas if they are thought to contribute to the task solution). In other words, the third scope condition specifies that solving the task must be more important to the focal actor than any other considerations (such as having his or her individual contributions accepted, or being liked by the partner).
The situational approach and the use of scope conditions are part of an overall theory-construction strategy whereby the initial focus is on the simplest contexts, which are studied with the aid of minimum assumptions. Complexities are then added gradually as knowledge accumulates. More recent expectation states research has, for example, extended the investigation from dyads to larger groups. Similarly, the scope condition of collective orientation has been relaxed in some of these studies, with results showing that status still affects expectations under such circumstances.
The core of the theory investigates the performance expectations an individual forms about self and other in settings where the two are engaged in the joint solution of a task. As indicated previously, the focal actor is assumed to value the task and to be both task-oriented and collectively oriented. A valued task implies that competence at the task is more desirable than incompetence, and therefore that levels of task ability constitute a status characteristic for that actor. Basically, the assignment of levels of this characteristic and corresponding performance expectations may occur in one of two ways: directly, from actual evaluations of task performance, or indirectly, on the basis of other status characteristics of the performers that self perceives to be relevant to the task. For simplicity, I first consider situations where expectations are formed exclusively in one fashion or the other. These two ways comprise the major branches of expectation states theory, the former focusing on the effects of performance evaluations and the latter on the effects of other status information. These branches and their development are described below. (Because of space limitations, the emphasis in this article is on the basic ideas, and specific references can only be given for a sample of the theoretical and empirical work done in the program. This sample consists of key earlier pieces and a selection of the more recent research. It should be noted, however, that expectation states theory has been a collaborative effort from its beginnings and that a considerable number of researchers have worked in it. For references to their individual contributions at the various stages of the program, see the reviews and edited collections, as well as the examples of research, mentioned at the end of this article.)
Initial work in the first branch of the theory investigated the formation of performance expectations in situations where two persons begin their interaction as status equals (i.e., they have no information from either inside or outside the group that would enable them to assign different levels of task competence to each other). In this case, the assignment of levels of competence that eventually occurs is the result of a generalization from evaluations of units of performance. These evaluations may be made by the performers themselves as they resolve disagreements between them regarding the correct solution to the task (Berger and Conner 1969; Berger and Snell 1961) or by a "source" or third party with the right to evaluate the performers (Crundall and Foddy 1981; Webster 1969; Webster and Sobieszek 1974). Moreover, the evaluations may or may not be made through the use of objective criteria. Various other topics were later investigated within this branch. These include the effects on expectations of the type of evaluations received (e.g., either positive or negative; either highly or moderately positive), the consistency of these evaluations (e.g., across performances or across sources), and the types of attributes that confer evaluative competence on a source (such as level of task competence or vicarious exposure to the task). Other recent work has dealt with the effects of applying either strict or lenient standards for competence (as well as for lack of competence) to the processing of evaluations. The study of second-order expectations (those based on self's perceptions of the expectations held by the partner) also has received attention in this program, and is of direct relevance to this branch.
The second branch of the theory is primarily concerned with situations where the actors differ in status. It is in this branch, which has come to be known as "status characteristics theory," that a large proportion of expectation states research has been conducted. This work started with the investigation of situations in which two persons are differentiated with respect to a single diffuse status characteristic (Berger et al.  1972; Moore 1968). Furthermore, this difference constitutes the only information that the focal actor has about self and other, and he or she treats the attribute as a diffuse status characteristic (i.e., attaches different evaluations to its various levels and holds different performance expectations, specific as well as general, for each level).
The basic proposition in this branch states that unless the focal actor believes the characteristic to be irrelevant to the task at hand, he or she will use the status difference between self and other to organize their interaction. In other words, performance expectations for the task will be the result of importing status information from outside the group. This process is known as "status generalization." Let us consider gender as an example (with "man" carrying a more positive valuation than "woman") and assume that the focal actor is a man and that his partner is a woman. The theory predicts different expectations, depending on what the individual believes the sex linkage of the task to be. If the task is perceived as masculine (one at which, in general, men do better than women), self will use this information to infer that he is the better of the two. If he has no information about its sex linkage (perceives the task as neither associated with nor dissociated from sex differences), the "burden-of-proof principle" will apply: those of lower status will be considered to have inferior task competence, unless they demonstrate the opposite. Accordingly, also in this situation, the man will expect to be more competent at the task than his female partner. On the other hand, the theory predicts that he will not form expectations of his own superiority if the task has been explicitly dissociated from gender. Finally, if the valued task is perceived to be one at which women generally excel, the prediction is that the male actor will consider himself inferior to his female partner.
It is important to emphasize that the above predictions assume that gender is a diffuse status characteristic for the focal actor (and that, in this example, "man" carries a higher value than "woman"). If gender is not such a characteristic, the propositions do not apply. Because of the inclusion of this scope condition, the theory is able to incorporate cultural, historical, and individual differences. Thus, through this condition, it incorporates the fact that although gender is a diffuse status characteristic in most societies, it is more so in some than in others. The extent to which this is the case also may vary within a given society and from one historical period to another. Furthermore, the theory reflects the fact that even in strongly sexist societies there are likely to be individuals who are less sexist than the majority, and perhaps some who are not sexist at all. (This may be due, for example, to variations in socialization practices within a given culture, and/or to the different degrees of success of these practices). Note that the theory assumes that women, as well as men, may treat gender as a diffuse status characteristic and devalue women's performances. (For a theoretical account of the social construction of status beliefs, see Ridgeway 1991).
Let us then reanalyze the earlier example from the point of view of the female actor, and assume that she considers gender to be a diffuse status characteristic to the same degree as her male partner does. The theory predicts that she will then form expectations that are exactly complementary to those of her partner. Thus, if she either views the task as masculine or relates it to sex differences through a burden-of-proof process, she will consider herself the less competent of the two performers. (She will not form such expectations if she sees the task as dissociated from gender, and will believe herself to be the better performer if she views the task as feminine.) However, as with most status systems, those persons who benefit from the existing order can be expected to be more supportive of it than those who do not, socialization practices notwithstanding. Thus it could be that men tend to meet this scope condition more than women do. This is not an issue about the theory itself but an empirical matter regarding the particular instances to which it applies. Note, also, that while the preceding discussion has been limited to gender for the sake of providing an example, the comments are meant to apply to any diffuse status characteristic. In general, the strength of the link between status and expectations is a function of the extent to which the attribute is a status characteristic for the focal actor.
Performance expectations formed through status generalization may, of course, be affected by other factors. Accordingly, the basic ideas of the initial formulation of status characteristics theory have been expanded and elaborated in several directions. In particular, these include the study of situations where actors possess more than one status characteristic. These may be either specific or diffuse, or either consistent or inconsistent with each other in terms of the levels of competence they imply. Furthermore, as a set, they may either equate or differentiate the performers to various degrees (Berger et al. 1977). This extended formulation specifies the conditions under which the available status items will become salient (or "activated"), as well as the rules by which they will be processed. It is proposed that all (and only) activated items will be used and that they will be combined according to the following principles: inconsistent information will have more impact than consistent information, and each additional item of consistent information will have less weight than it would by itself. Furthermore, this version includes referent actors (nonparticipating actors who serve as objects of comparison for the performers) and their role in the formation of expectations.
The two main branches of the program have not developed independently of each other. In fact, the branches touch and intertwine at various points, as expectations are formed on the basis of both performance evaluation and status information. For example, the extent to which the focal actor accepts the evaluations from a source is often affected by the diffuse status characteristics (and their levels) of everyone involved in the situation. Of particular interest is the resilience of status-based expectations when these are contradicted by actual performance evaluations. This resistance shows itself in, for example, the higher level of performance that lower-status actors have to achieve in order to escape the effects of status generalization. The inconsistency between performance evaluation and status information is of course a special case of the extension of status characteristics theory to more than one attribute. A recent theoretical proposal in this area links the resilience to the use of double (or even multiple) standards for both competence and incompetence that protect higher-status actors and penalize lower-status ones.
Other expectation states research, also of relevance to both branches, has included a rich variety of topics. For example, theoretical and empirical work has investigated the relationship between performance expectations and reward expectations, the status-value view of distributive justice, the transfer of expectations across actors and situations, and the legitimation of status positions. The latter work examines how and why individuals come to view the status hierarchies in which they are involved as right and proper. Also of special interest is the analysis of the way in which task settings generate emotional reactions and how these, in turn, affect performance expectations. Other important areas have been: The extension of status characteristics theory to include self-fulfilling effects of status on performance; the study of the role that status cues (such as accent, dress, demeanor) have in the assignment of task competence; and the work on how the attribution of personality characteristics (such as friendly, rigid, outgoing) and moral characteristics (such as honest, fair, selfish) affects this assignment and is, in turn, affected by it.
No review of expectation states theory would be complete without a discussion of three features that, together with those mentioned earlier, serve to characterize this program. First, a sizable portion of the research has used a standardized experimental situation involving two subjects (Webster and Sobieszek 1974, Appendices 1 and 2). The utilization of an experimental approach clearly is well suited to the program's analytical strategy of investigating the effects of a few variables at a time, and the standardized setting has contributed to cumulativeness in the research findings by allowing comparability across studies. In this setting, the task consists of a series of trials, each involving a binary choice regarding visual stimuli, with subjects allowed to interact only indirectly (as opposed to face-to-face) with each other. The situation consists of two phases. In the first phase, performance expectations are formed on the basis of information controlled by the experimenter. This may include evaluations of each person's task performance, cues regarding the participants' other status characteristics, or a combination of the two. In the second phase, each trial consists of an individual initial choice, the experimentally controlled communication of the partner's choice, and a final individual choice. In most trials, the communication indicates that the partner disagrees with the subject. The dependent variable of central interest is the amount of influence (one of the components of the power and prestige order of the group) that a partner exerts on the subject in the resolution of those disagreements. Several versions of this setting, including computerized ones, have now been developed. (Additional bases of empirical support for the theory include direct evidence from research conducted in discussion groups, classrooms, and various other settings—such as those mentioned below—as well as indirect evidence from work originating in several other theoretical traditions. These studies corroborate and extend the knowledge gained through the use of the standardized setting. For a review of the different types of support for the theory, see Berger et al. 1989).
Second, the program has been characterized by a strategy of stating propositions as part of deductive systems. These systems have been formalized to various extents, and different techniques (such as stochastic models, a Bayesian approach, graph theory, fuzzy set theory) have been used. The most comprehensive of these formulations is the graph theoretic model of the extended version of status characteristic theory (Berger, Fisek, Norman, and Zelditch 1977). This model has also been used to formalize other areas of the program, such as the formation of expectations in groups where members initially are status equals, and the legitimation and delegitimation of status hierarchies.
Third, from its beginnings, researchers in the program have shown a definite interest in applications and interventions designed to understand and alleviate specific social problems. A large portion of this work has included the development of techniques to reduce the effects of status generalization, particularly those based on skin color and ethnic background, in classroom settings (Cohen 1972, 1982; Cohen and Lotan 1997; Entwisle and Webster 1972). It is important to note that this research has progressed in close association with theoretical work; thus applications and interventions have both stimulated and benefited from the development of abstract formulations.
Expectation states theory continues to generate strong interest among researchers in group processes and related areas, and to develop in several directions. For example, the refinement, expansion, and integration of various aspects of the program are currently under way. There are also some exciting new areas of research, such as the study of the relationship between status and various other factors in the assignment of task competence, namely affect (emotions and sentiments), formal authority, power, dominance, and social identity. Furthermore, the notion of an "expectation state" has been generalized from its original reference to status-based processes, to incorporate those concerning affect and control. Finally, the research methodology has been extended to comprise a wide range of settings and tasks, including face-to-face contexts, larger groups, interactions where selected aspects of formal organizations have been re-created, vignette studies, computerized tasks, and settings where actors are not required to be collectively oriented.
For reviews and assessments of the expectation states program at various stages, see Berger (1988); Berger, Rosenholtz, and Zelditch (1980); Berger, Wagner, and Zelditch (1985, 1989); Meeker (1981); Ridgeway and Walker (1995); Wagner and Berger (1993); Webster and Foschi (1988a). For edited collections of work on various topics, see Berger, Conner, and Fisek (1974); Berger and Zelditch (1985; 1998b); Szmatka, Skvoretz, and Berger (1997); Webster and Foschi (1988b). For examples of the more recent research, see Balkwell, Berger, Webster, Nelson-Kilger, and Cashen (1992); Cohen and Zhou (1991); Foschi (1996); Foschi, Lai, and Sigerson (1994); Gerber (1996); Lovaglia and Houser (1996); Lovaglia, Lucas, Houser, Thye, and Markovsky (1998); Riches and Foddy (1989); Ridgeway (1989); Ridgeway, Boyle, Kuipers, and Robinson (1998); Ridgeway and Johnson (1990); Shelly and Webster (1997) Stewart and Moore (1992); Troyer and Younts (1997); Wagner, Ford, and Ford (1986); Webster and Whitmeyer (1999). For discussions of the program's methodological and metatheoretical assumptions, see Berger and Zelditch (1998a); Berger, Zelditch, and Anderson (1972); Cohen (1989, esp. chap. 6); Zelditch (1969).
Balkwell, James W., Joseph Berger, Murray Webster, Jr., Max Nelson-Kilger, and Jacqueline Cashen 1992 "Processing Status Information." Advances in Group Processes 9:1–20.
Berger, Joseph 1958 "Relations Between Performance, Rewards, and Action-Opportunities in Small Groups." Ph.D. diss., Harvard University.
——1988 "Directions in Expectation States Research." In Murray Webster, Jr., and Martha Foschi, eds., Status Generalization: New Theory and Research. Stanford, Calif.: Stanford University Press.
——, Bernard P. Cohen, and Morris Zelditch, Jr. 1966 "Status Characteristics and Expectation States." In Joseph Berger, Morris Zelditch, Jr., and Bo Anderson, eds., Sociological Theories in Progress, vol. 1. Boston: Houghton Mifflin.
——, Bernard P. Cohen, and Morris Zelditch, Jr. 1972 "Status Characteristics and Social Interaction." American Sociological Review 37:241–255.
——, and Thomas L. Conner 1969 "Performance Expectations and Behavior in Small Groups." Acta Sociologica 12:186–198.
——, Thomas L. Conner, and M. Hamit Fisek, eds., 1974 Expectation States Theory: A Theoretical Research Program. Cambridge, Mass.: Winthrop.
——, M. Hamit Fisek, Robert Z. Norman, and Morris Zelditch, Jr. 1977 Status Characteristics and Social Interaction: An Expectation-States Approach. New York: Elsevier.
——, Susan J. Rosenholtz, and Morris Zelditch, Jr. 1980 "Status Organizing Processes." In Alex Inkeles, Neil J. Smelser, and Ralph H. Turner, eds., Annual Review of Sociology, vol. 6. Palo Alto, Calif.: Annual Reviews.
——, and J. Laurie Snell 1961 "A Stochastic Theory for Self-Other Expectations." Technical Report no. 1. Stanford, Calif.: Laboratory for Social Research, Stanford University.
——, David G. Wagner, and Morris Zelditch, Jr. 1985 "Expectation States Theory: Review and Assessment." In Joseph Berger and Morris Zelditch, Jr., eds., Status, Rewards and Influence: How Expectations Organize Behavior. San Francisco: Jossey-Bass.
——, David G. Wagner, and Morris Zelditch, Jr. 1989 "Theory Growth, Social Processes, and Metatheory." In Jonathan H. Turner, ed., Theory Building in Sociology: Assessing Theoretical Cumulation. Newbury Park, Calif.: Sage.
——, and Morris Zelditch, Jr., eds., 1985 Status, Rewards, and Influence: How Expectations Organize Behavior. San Francisco: Jossey-Bass.
——, and Morris Zelditch, Jr. 1998a "Strategies of Theory Construction." In Joseph Berger and Morris Zelditch, eds., Status, Power and Legitimacy: Strategies and Theories. New Brunswick, N.J.: Transaction.
——, and Morris Zelditch, Jr., eds., 1998b Status, Power and Legitimacy: Strategies and Theories. New Brunswick, N.J.: Transaction.
——, Morris Zelditch, Jr., and Bo Anderson 1972 "Introduction." In Joseph Berger, Morris Zelditch, Jr., and Bo Anderson, eds., Sociological Theories in Progress, vol. 2. Boston: Houghton Mifflin.
Cohen, Bernard P. 1989 Developing Sociological Knowledge: Theory and Method, 2nd ed. Chicago: Nelson-Hall.
——, and Xueguang Zhou 1991 "Status Processes in Enduring Work Groups." American Sociological Review 56:179–188.
Cohen, Elizabeth G. 1972 "Interracial Interaction Disability." Human Relations 37:648–655.
——1982 "Expectation States and Interracial Interaction in School Settings." In Ralph H. Turner and James F. Short, Jr., Annual Review of Sociology, vol. 8. Palo Alto, Calif.: Annual Reviews.
——, and Rachel A. Lotan, eds., 1997 Working for Equity in Heterogeneous Classrooms: Sociological Theory in Practice. New York: Teachers College Press.
Crundall, Ian, and Margaret Foddy 1981 "Vicarious Exposure to a Task as a Basis of Evaluative Competence." Social Psychology Quarterly 44:331–338.
Entwisle, Doris R., and M. Webster, Jr. 1972 "Raising Children's Performance Expectations: A Classroom Demonstration." Social Science Research 1:147–158.
Foschi, Martha 1996 "Double Standards in the Evaluation of Men and Women." Social Psychology Quarterly 59:237–254.
——, Larissa Lai, and Kirsten Sigerson 1994 "Gender and Double Standards in the Assessment of Job Applicants." Social Psychology Quarterly 57:326–339.
Gerber, Gwendolyn L. 1996 "Status in Same-Gender and Mixed-Gender Police Dyads: Effects on Personality Attributions." Social Psychology Quarterly 59:350–363.
Lovaglia, Michael J., and Jeffrey A. Houser 1996 "Emotional Reactions and Status in Groups." American Sociological Review 61:867–883.
——, Jeffrey W. Lucas, Jeffrey A. Houser, Shane R. Thye, and Barry Markovsky 1998 "Status Processes and Mental Ability Test Scores." American Journal of Sociology 104:195–228.
Meeker, Barbara F. 1981 "Expectation States and Interpersonal Behavior." In Morris Rosenberg and Ralph H. Turner, eds., Social Psychology: Sociological Perspectives. New York: Basic Books.
Moore, James C., Jr. 1968 "Status and Influence in Small Group Interactions." Sociometry 31:47–63.
Riches, Phoebe, and Margaret Foddy 1989 "Ethic Accent as a Status Cue." Social Psychology Quarterly 52:197–206.
Ridgeway, Cecilia L. 1989 "Understanding Legitimation in Informal Status Orders." In Joseph Berger, Morris Zelditch, Jr., and Bo Anderson, eds., Sociological Theories in Progress: New Formulations. Newbury Park, Calif.: Sage.
——1991 "The Social Construction of Status Value: Gender and Other Nominal Characteristics." Social Forces 70:367–386.
——, and Cathryn Johnson 1990 "What Is the Relationship Between Socioemotional Behavior and Status in Task Groups?" American Journal of Sociology 95:1189–1212.
——, Elizabeth H. Boyle, Kathy J. Kuipers, and Dawn T. Robinson 1998 "How Do Status Beliefs Develop? The Role of Resources and Interactional Experience." American Sociological Review 63:331–350.
——, and Henry A. Walker 1995 "Status Structures." In Karen S. Cook, Gary A. Fine, and James S. House, eds., Sociological Perspectives on Social Psychology. Boston: Allyn and Bacon.
Shelly, Robert K., and Murray Webster, Jr. 1997 "How Formal Status, Liking, and Ability Status Structure Interaction: Three Theoretical Principles and a Test." Sociological Perspectives 40:81–107.
Stewart, Penni A., and James C. Moore, Jr. 1992 "Wage Disparities and Performance Expectations." Social Psychology Quarterly 55:78–85.
Szmatka, Jacek, John Skvoretz, and Joseph Berger, eds., 1997 Status, Network, and Structure: Theory Development in Group Processes. Stanford, Calif.: Stanford University Press.
Troyer, Lisa, and C. Wesley Younts 1997 "Whose Expectations Matter? The Relative Power of First- and Second-Order Expectations in Determining Social Influence." American Journal of Sociology 103:692–732.
Wagner, David G., and J. Berger 1993 "Status Characteristics Theory: The Growth of a Program." In Joseph Berger and Morris Zelditch, Jr., eds., Theoretical Research Programs: Studies in the Growth of Theory. Stanford, Calif.: Stanford University Press.
——, Rebecca S. Ford, and Thomas W. Ford 1986 "Can Gender Inequalities Be Reduced?" American Sociological Review 51:47–61.
Webster, Murray, Jr. 1969 "Source of Evaluations and Expectations for Performance." Sociometry 32:243–258.
——, and Martha Foschi 1988a "Overview of Status Generalization." In Murray Webster, Jr. and Martha Foschi, eds., Status Generalization: New Theory and Research. Stanford, Calif.: Stanford University Press.
——, and Martha Foschi, eds., 1988b Status Generalization: New Theory and Research. Stanford, Calif.: Stanford University Press.
——, and Barbara Sobieszek 1974 Sources of Self-Evaluation: A Formal Theory of Significant Others and Social Influence. New York: Wiley.
——, and Joseph M. Whitmeyer 1999 "A Theory of Second-Order Expectations and Behavior." Social Psychology Quarterly 62:17–31.
Zelditch, Morris, Jr. 1969 "Can You Really Study an Army in the Laboratory?" In Amitai Etzioni, ed., A Sociological Reader on Complex Organizations, 2nd ed. New York: Holt, Rinehart and Winston.