Members' Responses to Organizational Identity Threats: Encountering and Countering the 'Business Week' Ratings.

AN9611141411;asq01sep.96;2004Oct07.15:00;v2.1

Members' Responses to Organizational Identity Threats: Encountering and Countering the Business Week Rankings 

This research investigates how organizational members respond to events that threaten their perceptions of their organization's identity. Using qualitative, interview, and records data, we describe how members from eight "top-20" business schools responded to the 1992 Business Week survey rankings of U.S. business schools. Our analysis suggests that the rankings posed a two-pronged threat to many members' perceptions of their schools' identities by (1) calling into question their perceptions of highly valued, core identity attributes of their schools, and (2) challenging their beliefs about their schools' standing relative to other schools. In response, members made sense of these threats and affirmed positive perceptions of their school's identity by emphasizing and focusing on their school's membership in selective organizational categories that highlighted favorable identity dimensions and interorganizational comparisons not recognized by the rankings. Data suggest that members' use of these categorization tactics depended on the level of identity dissonance they felt following the rankings. We integrate these findings with insights from social identity, self-affirmation, and impression management theories to develop a new framework of organizational identity management.[*]

Out of the four billion people on earth, everyone in our class must be in at least the most fortunate two-tenths of one percent. But we figure if this school were ranked first or second instead of ninth [by the Business Week survey], we'd be in the top one-tenth of one percent, so we're all pissed off.

--Stanford MBA, responding to the 1990 Business Week rankings of U.S. business schools (quoted in Robinson, 1995: 189; emphasis in original)

An organization's identity reflects its central and distinguishing attributes, including its core values, organizational culture, modes of performance, and products (Albert and Whetten, 1985; Dutton and Dukerich, 1991; Whetten, Lewis, and Mischel, 1992). For members, organizational identity may be conceptualized as their cognitive schema or perception of their organization's central and distinctive attributes, including its positional status and relevant comparison groups (Dutton and Penner, 1993; Kramer, 1993; Dutton, Dukerich, and Harquail, 1994). Consequently, external events that refute or call into question these defining characteristics may threaten members' perceptions of their organization's identity (Dutton and Dukerich, 1991). For example, journalists have recently criticized the socially responsible firm, the Body Shop, for exploiting the very populations it was supposedly serving. Such criticisms may threaten members' perceptions of what the organization is and what it stands for.

The purpose of this paper is to describe how organization members respond to such identity-threatening events, which represent a symbolic and sensemaking dilemma for organization members that is distinct from most previously studied organizational image threats (i.e., events that threaten members' perceptions of an organization) (cf. Weick, 1993; Elsbach, 1994; Dutton, Dukerich, and Harquail, 1994). While most existing research on organizational image management has focused on how formal spokespersons use impression management tactics to improve external perceptions of the organization following controversies arising from what an organization has done (i.e., verbal accounts following public health crises, scandals, and accidents) (Sutton and Callahan, 1987; Marcus and Goodman, 1991; Elsbach, 1994), we propose that organizational identity threats cause organization members to use cognitive tactics to maintain both personal and external perceptions of what their organization is or stands for.

Dutton, Dukerich, and Harquail (1994) recently proposed that it is important to distinguish between two types of organizational identity perceptions: (1) members' perceived organizational identity (i.e., what members themselves believe are the central, distinctive, and enduring attributes of their organization) and (2) their construed external identity (i.e., what members think outsiders believe are the central, distinctive, and enduring attributes of their organization). From the standpoint of the present analysis, it is important to note that perceived organizational identity and construed external organizational identity are both cognitive representations held by individual members, and both may be affected by external attributions of organizational identity. When we speak of members' identity perceptions, therefore, we are referring to both their perceived organizational identity and their construed external identity. Similarly, when we speak of tactics used to affirm identity perceptions, we are referring to tactics that affirm both perceived and construed identities.

Using this conceptual perceptive, as well as insights from social identity theory (Ashforth and Mael, 1989), self-affirmation theory (Steele, 1988), and impression management theory (Tedeschi, 1981; Tedeschi and Melburg, 1984; Elsbach, 1994), we develop a framework of members' responses to organizational identity threats that emerged from a qualitative examination we did of business school members' responses to Business Week magazine's rankings of U.S. business schools. As we describe below, our preliminary examination of these responses, as reported in the popular business press (Putka, 1990; Hall, 1990; Hay, 1992), in casual conversations with colleagues and students at a few top business schools, and in an initial interview with survey founder John Byrne, suggested that the Business Week rankings threatened many members' perceptions of their school's central and distinctive attributes, i.e., their school's identity.

The Business Week Rankings As an Organizational Identity Threat

Business Week magazine has ranked the top-20 U.S. business schools every two years since 1988. The survey evaluates business schools on two primary criteria: (1) recent Master of Business Administration (MBA) graduates' satisfaction with the school and (2) recruiters' satisfaction with recent graduates of the school. Business Week uses a composite score of these two dimensions to evaluate and rank the top-20 U.S. business schools. Our preliminary readings and interviews suggested that, prior to the

Business Week rankings, the absence of a dominant standard for evaluating business schools, along with the wide variety of possibly distinct niches, enabled schools to decide which identity attributes were important and with whom they should be compared. By imposing an ostensibly objective and uniform metric for evaluating all U.S. business schools, the Business Week rankings dramatically disrupted the status quo that these schools had long enjoyed, creating an organizational identity threat for some institutions. Early interviews suggested that this new ranking metric posed a two-pronged threat to many members' perceptions of their business schools' identities. First, the survey often devalued central and distinctive dimensions of a school's identity. Second, in many instances, it challenged members' prior claims about the positional status of their schools.

Devaluing core identity dimensions. Our pilot investigations suggested that the Business Week rankings threatened some members' perceptions of their organization's identity by calling into question the merit or importance of core, distinctive, and enduring organizational traits associated with their institutions. For example, in emphasizing MBAs' satisfaction with teaching as a primary evaluative criteria, Business Week implicitly challenged the value of many schools' longstanding research mission. Moreover, according to several business school admissions directors, as the rankings received increasing attention and achieved greater legitimacy with students, recruiters, and alumni, dimensions of a school's identity that were not included in the survey became perceived as less important and perhaps even irrelevant as indicators of a school's performance or quality. As one MBA student put it, "I don't need 'balanced excellence' [between teaching and research] in my program. I came here to benefit from research, not support it." Notwithstanding the shortsightedness of this view, by excluding certain historically important and enduring institutional characteristics, the Business Week survey threatened the core identities of many business schools, even those ranked near the top.

Such threats to identity value may be, in some respects, a unique consequence of the particular methodology used to generate the Business Week rankings. Our preliminary investigation suggested that, prior to the first Business Week survey, many business schools took pride in touting what they perceived as valuable attributes and distinctive competencies. For example, MBA catalogs from the University of Chicago and Stanford University business schools frequently drew attention to their excellent research faculty, noting their impressive record of scholarly accomplishments. By contrast, descriptions of other prominent business schools, such as Dartmouth and Cornell, frequently alluded to their close-knit communities and "user-friendly" MBA cultures. Although other surveys of business schools have existed for some time (e.g., Barron's Profiles of American Colleges, Peterson's Guide to American Colleges, Brecker and Merryman Inc.'s Survey of U.S. Business Schools, and U.S. News and World Report), none were perceived to have clear dominance or compelling merit within the business school community. In addition, the criteria for evaluating schools differed: Some drew distinctions between private vs. public institutions; others collapsed across such distinctions (cf. Hay, 1992). As a result, schools could select which surveys and which attributes to emphasize in their own evaluation.

The Business Week rankings dramatically changed all of this. They implied that "top" schools had to have national stature and an elite ranking among recently graduated MBAs and their recruiters. By framing this as an objective measure (i.e., based on over 6,000 responses), the survey implied that other ranking schemes (e.g., those based purely on deans' ratings of schools' academic quality) were overly narrow and of dubious validity. Thus, members of schools with strong regional or research reputations suddenly found that these core identity dimensions had little influence in determining their school's ranking.

Challenging claims of positional status. The rankings also challenged--and in some instances outrightly repudiated--members' prior claims about the relative standing or status of their school among U.S. business schools. In particular, they challenged the credibility of many schools' assertions that they were a leading, cutting-edge, or "top-tier" institution. As John Byrne, the creator and editor of the Business Week survey, noted, "For years and years there were probably 50 [business] schools that claimed that they were in the top 20 and probably hundreds that claimed they were in the top 40. . . . What the Business Week Survey does is eliminate the ability of some schools to claim that they are in a top group."

Members who had always considered their schools to be elite institutions suddenly found them ranked out of the top five or the top ten and categorized instead as "merely" a top-20 school. Perhaps the most dramatic example of this was the Harvard Business School, which had often been characterized in the press and by chief executive officers as the leading business school in America. Suddenly, Harvard found itself only a "runner up" in what one scholar aptly called "the Business Week beauty contest." Such threats to positional status are important to an organization's identity because they threaten the perceived favorability of comparisons with its peers. As Frank (1985: 7) noted, many of the rewards or goods for which individuals and organizations compete are positional goods, "sought after less because of any absolute property they possess than because they compare favorably with others in their own class. A 'good' school, for example, is sought less for its absolute quality than for its high rank among schools in general."

To better understand conceptually what such organizational identity threats might mean to organizational members and how we would expect members to respond to them, we next examined organizational and individual-level research on identity threats. This research is grounded in several theoretical perspectives, including impression management theory (Tedeschi, 1981), self-affirmation theory (Steele, 1988), and social identity theory (Ashforth and Mael, 1989).

Conceptualizing Members' Responses to Organizational Identity Threats

As noted, organizational identities define members' perceptions of their organization's central, distinctive, and enduring traits (Dutton, Dukerich, and Harquail, 1994). Individuals also maintain perceptions of their own social identity, which is a measure of their self-concept defined by their association and affiliation with various social groups (see Tajfel, 1982; Brewer and Kramer, 1986; Turner, 1987; Hogg and Abrams, 1988, for overviews). At a cognitive level, individuals' social identities are assumed to be organized in terms of multiple, hierarchically organized categories, including social categories based on such things as age, gender, and race, as well as institutional and organizational affiliations. Thus, members' perceptions of their organization's identity may have a direct effect on their perceptions of their own social identities.

Effects of organizational identity threats. Because of these psychological links between organizational and social identities, organizational scholars have increasingly argued that individuals attach considerable significance to their organization's identity (Brown and Williams, 1984; Ashforth and Mael, 1989; Kramer, 1993). A person can acquire a more positive social identity through association with organizations that have positive identities (Mael and Ashforth, 1992; Dutton, Dukerich, and Harquail, 1994) because "the attributes that comprise an organization's identity, by association, are transferred to individuals who work there" (Dutton and Penner, 1993: 103). Conversely, events that threaten the organization's identity constitute a threat to members' own social identities. According to this view, organizational identity threats create a predicament for organizational members. In response, theorists have suggested that members will be motivated to protect and affirm positive perceptions of their organization's identity to restore and affirm a positive social identity (Dutton, Dukerich, and Harquail, 1994). Yet, to date, theorists have not examined how members carry out such identity affirmations.

Most existing research on the management of organizational perceptions is grounded in an impression management theory perspective (e.g., Sutton and Callahan, 1987; Marcus and Goodman, 1991; Elsbach and Sutton, 1992). This research has identified a variety of externally directed tactics used by spokespersons who are responsible for justifying organizational performance and practices to various constituents to whom the organization is accountable (Staw, McKechnie, and Puffer, 1983; Salancik and Meindl, 1984; Elsbach, 1994). Yet most of this work has focused on one-time crises (i.e., an oil spill, a plane crash, a food safety crisis) that threatened the legitimacy of important practices or procedures (i.e., the clean-up protocol for an oil spill, the meat inspection process at a meat packing plant). By contrast, no studies have examined members' responses to events that challenge members' perceptions of the value of an organization's central, distinctive, and enduring attributes. Such events may pose greater threats to members' social identities and self-concepts than to their perceptions of the organization's external legitimacy (Dutton and Penner, 1993). In turn, these events may lead members to engage in self-directed identity affirmations, rather than externally directed excuses or justifications of the event, as predicted by impression management theories. While current organizational research does not discuss members' responses to organizational identity threats, social identity research provides some hints about how members might respond to such threats.

Responses to organizational identity threats. Social identity theorists suggest that people use cognitive tactics to maintain positive perceptions of their social identities. Turner's (1987) research on self-categorizations, for example, suggests that people can affect their social identities by selectively highlighting those social categories (e.g., triathlete, mother, female, professor) that most accentuate or contribute to a positive identity in a given situation. When membership in one category implies a negative social identity or is identity-threatening, a person can restore a positive sense of self by selectively increasing the salience of other unthreatened or untarnished categories (Hogg and Abrams, 1988). While organizational membership has not been studied in this context, this research suggests that, if organizational membership (i.e., Stanford professor) is viewed as an undesirable or threatening categorization, people may choose to describe themselves in terms of their professional affiliations (i.e., social psychologist) or nonwork social groups (i.e., a masters swimmer).

Another consequence of highlighting selective categorizations is that it influences the salience of interpersonal or intergroup comparisons.[1] As social identity theorists have noted, people can enhance their social identities by highlighting their membership in categories that are widely viewed as high status in comparison with other categories (Tajfel and Turner, 1979; Hogg and Abrams, 1988; Hinkle and Brown, 1990). Members of low-status groups may improve their relative status by selecting different groups with which to be compared. For example, studies have shown that, in response to threats to self-esteem, people sometimes invoke categorization schemes that highlight downward social comparisons to those who are worse off on some dimension (i.e., "I may have breast cancer, but at least I didn't have a double mastectomy") (Wood, Taylor, and Lichtman, 1985). In other cases, they may invoke categorization schemes that highlight similarities to highly performing others to improve perceptions of their status (Wheeler, 1966).

A variety of laboratory experiments provide evidence that highlighting selective categorizations and social comparisons affects people's self-perceptions (e.g., Tajfel, 1969; Kramer and Brewer, 1984; Gaertner et al., 1989, 1990). This notion also receives strong support from recent and closely related research on self-affirmation (Steele, 1988) and constructive social comparison processes (Goethals, Messick, and Allison, 1991). The portrait of the individual that emerges is that of a cognitively flexible, adaptive, and opportunistic social perceiver, one who responds to identity-threatening events by highlighting personal membership in select social categories to make salient his or her positive identity attributes, favorable status among peers, and favorable similarity or uniqueness relative to others. Yet this research has not discussed members' attempts to highlight selectively their organization's membership in favorable categories following events that threaten their organization's identity, nor do current models of social identity or self-affirmation describe any of the cognitive tactics members may use to affirm perceptions of their organization's identity following events that call into question or devalue their organization's central, distinctive, and enduring traits.

The above shortcomings in organizational theories of impression management and psychological theories of social identity, coupled with our preliminary findings about members' responses to the Business Week rankings, suggest that our understanding of the relationship between organizational identities and members' social identities may be enhanced by an investigation of members' responses to organizational identity-threatening events. We describe such an investigation in the following sections of this paper.

METHODS

We conducted a study of business school members' responses to the 1992 Business Week rankings of U.S. business schools. Although we reviewed data from the 1988 and 1990 Business Week rankings as well, we focused on the 1992 Business Week rankings because (1) complete records data were not available from earlier rankings, and (2) we were interested in assessing current organizational members' contemporaneous reactions to the rankings. We collected data for this study from January 1993 through December 1993.

Business Schools

We restricted our sample to schools in the top 20 because it was the major category emphasized by the Business Week survey and because schools falling outside the top 20 were merely grouped together as the "second twenty" schools. Thus, focusing on the top 20 allowed us to track a school's exact movement in rank and to examine how members responded to an institution's change in ranking.

We also focused our study on schools that we believed experienced a representative range of identity threats in response to the rankings. On prima facie grounds, we reasoned that the level of organizational identity threat would reflect both a school's absolute rank (e.g., being in or out of the top-ten category), as well as its change in rank (e.g., movement up or down in the top-20 rankings).

We initially chose twelve schools that represented a range of possible absolute ranks (high, medium, and low ranking), as well as a range of changes in rank (significant moves up, down, or relative stability in ranking between the 1990 and 1992 surveys). To narrow our sample to a smaller set of schools for more intensive study, we then collected evidence of initial reactions (e.g., claims that the rankings did not reflect the school's true identity, claims that a school should have been ranked higher, etc.) from various members of these schools. We examined responses to the survey in the schools' MBA newspapers and in local newspapers.

Based on the availability and credibility of this evidence, we subsequently selected eight business schools for more intensive examination. We summarize their rankings and movement in rank in Table 1. These schools were (1) Carnegie-Mellon University's Graduate School of Industrial Administration, (2) University of Texas' Graduate School of Business, (3) Stanford University's Graduate School of Business, (4) Cornell University's Johnson Graduate School of Management, (5) University of California's Haas School of Business, (6) Northwestern University's J. L. Kellogg Graduate School of Management, (7) University of Chicago's Graduate School of Business, and (8) Indiana University's Graduate School of Business.

Respondents

We interviewed a total of 43 respondents from the eight schools. To get a representative range of reactions, and because we were interested in how organizational members throughout the organization responded to the survey, we interviewed a cross section of business school members. We selected individuals who we thought would have generally high levels of identification with their organizations (i.e., their social identities would be strongly affected by their association with their school) and, as a consequence, would be interested in and affected by the Business Week rankings. Thus, from each business school we interviewed (a) two faculty members from the management area of the business school, (b) an MBA student editor or publisher of the MBA newspaper, (c) a dean or assistant dean highly involved with the MBA program, and (d) a director of public relations or communications for the business school. This sample thus included, but was not limited to, those in image-management roles.

Data Sources

Interviews. We conducted semistructured interviews lasting 30-60 minutes with each respondent. All interviews were either tape-recorded or hand-transcribed. We first asked respondents a series of questions concerning their perception of their school's identity, its unique attributes, and how it compared with other top-20 business schools.

We then asked them to describe their general reactions to the rankings and others' reactions they had observed (including reactions from school administrators, faculty, and students). Because we were specifically interested in individuals' personal reactions to the rankings, we framed the questions to elicit their own opinions and candid responses. For example, we asked respondents to explain why they thought their school had achieved its current rank. We also asked them to provide their opinions of the responses given by other schools, if they knew of any. Finally, we asked respondents to give their opinions about the long-term effects of the Business Week rankings.

Records data. To provide a benchmark that would help us better interpret and calibrate individuals' responses to the rankings and to get an accurate definition of each school's enduring identity dimensions, we obtained records data that were published prior to the first Business Week rankings in 1988 and covering the four years between that first ranking and the 1992 rankings. These sources included the 1987-88, 1989-90 and 1991-92 MBA program catalogs for each of the eight schools and Business Week's 1991 edition of the book, The Best Business Schools. We searched each of these sources for statements about a school's unique and defining characteristics.

We also used records data to obtain information about the eight schools' responses following the 1992 rankings. We searched for stories about the rankings in each school's MBA newspapers and its alumni magazines published during the six months following the rankings. Additionally, we searched for stories concerning the 1992 rankings in national and local newspapers published from cities near the eight schools (e.g., the Chicago Sun Times, the Chicago Tribune, the New York Times, the Los Angeles Times, the San Francisco Chronicle and Examiner, the Philadelphia Inquirer, and the Dallas Morning News). A few schools also provided internal documents and memos related to their reactions and responses to the rankings. A total of 47 different publications provided responses to the 1992 rankings.

Data Analysis

Our qualitative data analysis followed an iterative approach of traveling back and forth between the emerging theory, the existing literature, and our data (Glaser and Strauss, 1967; Eisenhardt, 1989). Based on our previous observations about business schools' intense interest in the Business Week rankings, our initial goal was to determine the specific ways in which these rankings affected members of business schools.

Early iterations. In our first iterations, we coded respondents' statements about their initial reactions to the rankings and the positiveness and strength of those reactions. We collected 162 statements of initial reactions to the rankings from respondents and records data (between 15 and 27 per school). Most respondents' initial reactions reflected some degree of cognitive distress related to the rankings' attributions about a school's identity. The degree of cognitive distress appeared to be related to the strength of respondents' statements about the rankings. As a rule of thumb, we interpreted data as indicating strong distress if all respondents or records data from a category (i.e., students, administrators, or faculty) made similar statements revealing cognitive distress about the rankings. Moderate evidence was indicated if only about half of the respondents indicated cognitive distress.

We also analyzed respondents' and records data statements that offered explanations of the rankings. We collected 554 statements from interview and records data describing reasons for and responses to the rankings. This analysis produced a preliminary typology of tactics that members from the eight schools used in response to the Business Week rankings. These tactics primarily took the form of organizational categorizations (i.e., highlighting a school's membership in a category). These categorizations appeared to be used to affirm their perceptions of their organization's identity to make sense of the rankings. For example, in responding to a poor ranking, a member might make a statement highlighting the "product" categories that defined his or her school but may have hurt it in the rankings, such as "we're a research-oriented school" or "we're dedicated to producing technical MBAs." Through further analysis, we defined four primary cognitive tactics members used for identity affirmation and sensemaking: (1) selective categorizations to highlight alternate identity attributes, (2) selective categorizations to highlight alternate comparison groups, (3) selective categorizations to excuse a ranking, and (4) selective categorizations to justify a ranking. Four researchers analyzed the data during these iterations to define these tactics. Two researchers confirmed the classification of each identity-affirmation or sensemaking tactic. Strong evidence for a tactic was indicated if most of the respondents or records data in a category (i.e., students, administrators, or faculty) used the tactic.

Middle iterations. In our next iterations, we analyzed data about the eight schools' identities during the five years prior to the 1992 Business Week rankings. We did this to provide a benchmark against which to examine members' identity affirmations or sensemaking responses. We searched for members' statements about their school's unique and defining characteristics from each school's 1987-88, 1989-90, and 1991-92 business school catalogs and biographies in Business Week's 1988 and 1990 editions of The Best Business Schools. We focused on statements from members that roughly fit the prototypes: "our school is an X type of school," "our school is different from most schools on dimension X," "a central dimension of our school is X," or "we have always been a type X school." We collected a total of 844 identity statements from the eight schools (between 60 and 200 statements from each school). We grouped these statements iteratively to determine the central, distinctive, and enduring dimensions of each school prior to the 1992 rankings. We retained only those dimensions that consistently appeared in publications over the five years analyzed (i.e., were included in every publication), including the year prior to the first Business Week rankings.

Later iterations. In our final iterations we looked for relationships between members' reactions to the rankings and their identity-affirmation or sensemaking responses. Extrapolating from individual-level research, we anticipated that members' perceptions of identity threats would predict the strength of their responses (Steele, 1988). We thus searched for trends in the quantity and quality of cognitive responses for schools whose members exhibited a high or low degree of threat to identity perceptions. We also searched for other predictors and trends in members' identity affirmations.

FINDINGS

As background, we first summarize our findings on the eight business schools' preexisting and enduring identities. We then summarize our findings about members' perceptions of the Business Week rankings as an organizational identity threat and describe the cognitive tactics members used in responding to these threats. Finally, we attempt to explicate the relationship between the magnitude of perceived threat and use of categorization strategies to respond to that threat.

Evidence of Enduring Organizational Identities

Our analysis of preexisting identity statements revealed that each school maintained a set of core identity dimensions over the six-year period we examined (i.e., 1987-1992). Table 2 presents a summary of all eight schools' core identity attributes and dimensions. As this table shows, a few schools maintained fairly narrowly defined identities (e.g., Northwestern was defined by only three attributes: group culture, elite performer, and continuous improver), while others were more broadly defined (e.g., Texas was defined by ten distinct identity dimensions). Further, many of the core attributes and distinctive competencies associated with these schools are quite different from those that are implied to be important by Business Week. Berkeley and Texas, for example, highlighted the fact that their MBA programs were embedded in a large public institution with sound academic and research values. Similarly, Cornell and Dartmouth highlighted their programs' small, user-friendly cultures as valued dimensions. As we will show, the perceived value of these identity dimensions--which were not emphasized by the Business Week rankings--played an important role in members' cognitive responses to those rankings.

Evidence of Perceived Organizational Identity Threats

Our data suggest that business school members' preexisting perceptions of their schools' identities were threatened by the Business Week rankings. In this paper, we view cognitive distress about the rankings as evidence of identity threats, and all of the respondents in our sample expressed some level of cognitive distress over Business Week's characterization of their school. We found it useful to conceptualize this cognitive distress as a kind of "identity dissonance," which reflected cognitive dissonance related to the disparity or inconsistency between members' perceptions of their organization's identity (e.g., their perception that their school is "a dynamic, still growing program") and the identity attributed to it by the Business Week survey (the assertion that the school "is standing still"). Evidence of members' identity dissonance is summarized in Table 3.

Degree of threat. Not surprisingly, the more severe degrees of threat, indicated by high levels of identity dissonance, were associated with disappointment with the rankings, either in terms of the absolute ranking achieved, a fall in the rankings, or an insufficient rise in the rankings (i.e., a rise to a position still far below a school's expected rank). In these instances, the rankings may have refuted specific claims members made about their schools' prestige or status. Such contradictions may have been especially distressing to business school members because they not only threatened members' perceptions of their school's positional status (see below) but also suggested that they had misrepresented their school in a self-serving manner. Thus, members of Berkeley, Carnegie-Mellon, Stanford, Cornell, and Texas expressed higher levels of dissonance in response to the rankings than members of Chicago, Indiana, and Northwestern.

Type of threat. In line with our pilot study findings, we also found that the rankings posed two types of organizational identity threats: (1) threats to the value of core attributes and (2) threats to perceived positional status of the organization. Threats to the value of a school's core identity attributes reflected respondents' reactions to the discrepancy between their beliefs about the value of their institution's core attributes and the value attributed to them by the survey (e.g., Business Week's assertion--or, in some instances, even implication--that a school lacked an important attribute or that it possessed "irrelevant" attributes). Threats to a school's perceived status, by contrast, reflected reactions to the discrepancy between members' beliefs about their institution's positional status and that assigned to it by the Business Week rankings.

Threats to the value of core-identity dimensions. Our data suggest that many respondents felt that Business Week either underestimated the importance of, or overlooked entirely, key attributes of their organization's core identity. A respondent from Carnegie-Mellon, for example, claimed that Business Week had completely overlooked the innovativeness of its program:

What other programs tend to be doing this year, we may have been doing 5 or 6 or 7 years ago. The mini-semester system that Wharton is doing, and Texas is doing, we started 8 years ago. U.S. News and World Report was touting Stanford for a new course in design of manufacturing and marketing new products when they were ranked #1. . . . we started that 6 years ago.

Similarly, Stanford respondents complained that the survey failed to recognize its excellence in faculty research. As one put it, "What bothers me is the need to quantify all this down to one number. They end up measuring things that aren't important. For example, we've spent time talking to Business Week about including faculty research in the survey." Finally, a Berkeley respondent expressed frustration over the survey's neglect of student culture by claiming, "An enormous part of what makes our school special is not susceptible to quantification. Students have risen to major roles which would be unheard of at other schools."

It is important to note that even members of highly ranked or highly improved schools seemed to perceive the survey results as threatening and displayed signs of distress over the fact that the rankings overlooked or devalued cherished dimensions of their school's identities. For example, one Indiana respondent, noting its highly innovative program, complained, "Business Week has detailed none of the changes that have taken place within the IU MBA program over the past two years." Similarly, rather than dwelling on positive attributes that Business Week had attributed to its program, a Northwestern respondent complained that the rankings failed to recognize its continuous improvement, i.e., "I was a little upset about the Business Week article when we went through it. It made us sound like we were standing still" [emphasis added]. Finally, several Chicago respondents claimed that the ranking's focus on teaching was perceived as a threat to the value of faculty research. As one respondent put it, "Some think the whole thing is ridiculous and are hostile that we pay more attention to it than things like research."

Threats to perceived organizational status. Evidence also suggested that members from lower-ranked schools experienced varying levels of dissonance regarding their beliefs about their school's position in the rankings. Many respondents reacted with statements of disbelief and denial, claiming that their school's ranking was not indicative of its true stature. For example, one Berkeley respondent asserted, "I look at some of the schools, and I have a hard time believing, from what I know of colleagues and what I know of the schools, that they really belong ahead of us. So in that sense I'm in denial." Similarly, a Carnegie-Mellon respondent noted, "The students were disappointed. Certainly the second-year MBAs took the view that we're better than that." Finally, a Texas spokesperson reported, "Students were upset; many said, 'I applied to a top 20 school, and it's not a top 20 school'."

Overall, our data indicate that members did not passively accept their Business Week ranking, nor, however, did they completely discount it. Because the rankings were open to multiple and conflicting interpretations and value orientations, they generated interpretive ambiguity and dissonance and motivated members to engage in sensemaking behavior. As Weick (1995: 100) argued, occasions that seem incongruous or that "violate expectational frameworks" prompt organizational sensemaking. Along these lines, we found that members used several distinct strategies to restore and affirm their positive perceptions of organizational identity following the Business Week rankings.

Responses to Organizational Identity Threats

Organizational members' primary response to threats posed by the Business Week rankings was to make salient their school's membership in selective and favorable social groups through (1) categorizations that highlighted positive identity attributes not emphasized by the rankings and (2) categorizations that highlighted favorable social comparisons not emphasized by the rankings. It is important to note that members were not attempting to place their schools in new categories but, instead, were highlighting their school's preexisting membership in categories that the rankings did not recognize.

Members appeared to use these categorization tactics for two purposes: (1) to affirm positive aspects of their school's identity that the rankings had neglected and (2) to make sense of and explain why their school achieved a specific, disappointing ranking. Responses used for the first purpose turned attention away from the rankings and focused it on favorable aspects of the school's identity that members felt should be included in an evaluation of business schools. Responses used for the second purpose directed attention toward the ranking itself and showed how it was misleading in its representation of the school and ignored aspects of the school's identity that were more important than the criteria used in the survey. Members' use of these categorization tactics for these purposes are summarized in Table 4 and described below.

Affirming organizational identities. Two important attributes of members' responses indicated that they were designed, at least in part, to affirm organizational identities. First, a majority of members' selective categorizations highlighted cherished attributes of the organizations' enduring identities that were neglected by the rankings, suggesting to us that members were attempting to categorize their organizations selectively along core dimensions in the same way that individuals categorize themselves to affirm positive social identities (Steele, 1988). As Steele (1988: 291) noted, following identity-threatening events, "[The perceived integrity of the self may be restored] by affirming and sustaining valued self-images. To be effective these images must be at least as important to the individual's perception of self-adequacy as are the negative images inherent in the threat."

The second indication we found that members' responses were designed as identity affirmations was that even members of schools that did well in the rankings responded by selectively categorizing their schools along central identity dimensions not recognized by the rankings. In these cases, members had no reason to manage organizational images, since the favorable rankings, in objective terms, were image-enhancing. To the extent that the survey overlooked or minimized other cherished dimensions, however, they seemed to motivate members to affirm neglected facets of their organization's identity.

Tactic 1: Selective categorizations highlighting alternate identity attributes. As noted earlier, researchers have shown that, when responding to personal identity threats, people can enhance their self-worth by highlighting positive dimensions of their identities that are unrelated to the threat (Steele, 1988). As Ashforth and Mael (1989: 35) suggested, "Individuals have multiple, loosely coupled identities, and inherent conflicts between their demands are typically not resolved by cognitively integrating the identities, but by ordering, separating, or buffering them." Similarly, theorists have proposed that organizations may emphasize identity dimensions that portray them in the most favorable light (Albert and Whetten, 1985; Dutton and Dukerich, 1991). As Albert and Whetten (1985: 269) proposed, "In those cases in which a distinctive identity is prized, one might expect organizations to select uncommon dimensions of inter organizational comparison as well as uncommon locations along more widely employed dimensions."

Our data indicate similarly that when respondents felt that the rankings undervalued distinctive and central dimensions of their schools that they believed should be considered when evaluating a business school, they selectively categorized their schools along these alternate but cherished identity dimensions. Respondents from Berkeley, for example, categorized their school as a "public management" program, implying that it was distinct from other business schools. They also noted that their program was entrepreneurial and catered to the needs of West Coast students better than other schools, including those ranked higher by Business Week. As one Berkeley respondent noted in arguing the importance of their entrepreneurial dimension, "We really value our entrepreneurial culture. If the Haas emphasis on high-tech and entrepreneurship were to change, the school would lose its identity and competitive advantage" [emphasis added]. One Stanford respondent similarly noted that the Stanford Business School was uniquely oriented to the entrepreneurial needs of its students, which the respondent felt was unappreciated or underestimated by the Business Week survey, i.e., "More Stanford MBAs have non-Fortune 1000 interests, choosing instead, smaller and entrepreneurial ventures, as well as public management, and non-profit firms. . . . Some of the things that improve rankings are part of what we don't want to change" [emphasis added]. Finally, one Texas respondent categorized his school in terms of its regional standing, noting that its MBA program catered to "regional labor markets" better than other schools and that regional standing was a more important metric for evaluating its program than its standing in a survey that compared all schools across the nation.

Members of the more highly ranked schools also used this strategy. They seemed to feel that their schools' identities were threatened because the Business Week rankings did not recognize some positive or distinctive dimension of their school, even though it rated them highly on other dimensions. Chicago respondents, for example, emphasized that, despite their high student satisfaction ratings, they had not given up their highly valued academic values: "the core of the place has not been altered. It was never our intention to say our solid social science traditions are irrelevant" [emphasis added]. In all of these instances, members' selective highlighting of alternative categories made positive organizational identity attributes salient and, by implication, affirmed positive perceptions of organizational and social identities.

Tactic 2: Categorizations highlighting alternate comparison groups. As noted earlier, social identity researchers have suggested that people may accentuate intergroup differences that reflect positively on the group to which they belong (Tajfel and Turner, 1979). As Hogg and Abrams (1988: 23) noted: "By differentiating ingroup from outgroup on dimensions on which the ingroup is at the evaluatively positive pole, the ingroup acquires a positive distinctiveness, and thus a relatively positive social identity in comparison to the outgroup." Following unfavorable social comparisons, for example, people may invoke comparisons based on other, more flattering dimensions (Salovey and Rodin, 1984) or dimensions on which they appear to have an advantage (Taylor, Wood, and Lichtman, 1983). In much the same way, we found that following the Business Week rankings, many business school members selectively categorized their schools in ways that placed them in more favorable interorganizational comparison groups. This strategy seemed to affirm both their perceptions of core identity dimensions and their perceptions of the school's positional status.

First, we found that many business school members used categorizations that increased the salience of identity dimensions that were also held by well-respected and highly ranked schools but were neglected or undervalued by Business Week. By categorizing their institutions in terms of attributes of well-respected schools, these members affirmed the value of their organization's core identity dimensions and implied that their school should be compared favorably with other schools in the same way that highly ranked schools were (Tajfel, 1969; Tajfel and Wilkes, 1973). In effect, these categorizations allowed an institution and its members to "bask in the reflected glory" of another's achievement (cf. Cialdini et al., 1976). For example, in categorizing their school as a regional leader, some Texas respondents proposed that this put their school in the same category as more highly ranked Michigan. They also implied that this categorization would provide them with positive social comparisons to others in their region: "We are considered to be the best in our region. . . like Michigan, which is a very powerful regional school, and is also of national stature with Stanford, Harvard, and Wharton. So that's how we'd like to be seen. We'd like to be a school that totally dominates a region, and yet is not known as only a regional school." Similarly, members of Stanford affirmed the value of having an elite student body by suggesting this categorization put them in the same class as #3-ranked Harvard. As one member noted, "We've got the brightest students in the country. . . . I think 1/3 of our students turned down Harvard to come here." This strategy was also used by members of schools that did well in the rankings. Even members of #2-ranked Chicago attempted to affirm the value of their school's academic identity by equating themselves with other well-respected research institutions. As one respondent claimed, "We're a top research institution. I think of us in the same academic league as Harvard and Stanford."

Organization members also used these types of categorizations to affirm their preexisting perceptions of their school's high positional status among U.S. business schools. For example, members from Berkeley and Texas selectively categorized their organizations as public institutions to change their organization's relevant comparison group to a smaller set of highly ranked public institutions. From the reference point of this category, both schools compared more favorably than they had in the Business Week rankings. As one Berkeley respondent aptly noted, "In its market, Berkeley does a better job than most schools. But [Business Week's] throwing the Fords and the Chevys and the Porsches in the same mix" [emphasis added]. Carnegie-Mellon respondents suggested, similarly, that they should be assessed against other technical business schools. As one respondent noted, "It's really not fair. It's like judging apples and oranges, and we're not the same type of school as many others" [emphasis added]. In these examples, members suggested that comparison groups consisting of few schools possessing similar characteristics were more valid than the comparison groups based on Business Week's single performance measure (i.e., customer satisfaction); thus, selective categorizations narrowed or reduced the size of the organization's relevant comparison group. These findings are similar to previous results showing that people prefer performance comparisons with others on similar dimensions that they believe are related to or predictive of performance, rather than comparisons with similar performance outcomes directly (Wheeler and Zuckerman, 1977; Miller, 1982). Zanna, Goethals, and Hill (1975), for example, found that competitive swimmers preferred comparisons based on age, experience, and recent training, rather than solely on performance. Researchers suggest people prefer comparisons based on these related attributes because they allow them to better determine the meaning of outcomes (e.g., whether a favorable comparison is due to ability or training) and to feel more satisfied with those outcomes (Wood, 1989).

Explanations and sensemaking. While many respondents used categorizations primarily to affirm positive organizational identities by highlighting valued identity dimensions or comparison groups that they believed should be included in business school evaluations, some went beyond identity affirmation, using these same types of categorizations to explain their ranking. In particular, members of several low or disappointingly ranked schools described their schools in terms of categories that, they suggested, prevented them from achieving a higher ranking (i.e., an excuse) or were more important than categorizations necessary to achieve a higher ranking (i.e., a justification) (Schlenker, 1980; Tedeschi, 1981). These responses directed attention toward the methodology of the survey and implied that it misrepresented a school's true stature by ignoring these important dimensions. While we recognized these explanatory responses as fitting traditional forms of impression management (i.e., they resemble excuses and justifications), in most cases organizational members appeared to use them to make sense of the rankings and improve their own, rather than others' perceptions of their school's identity. As Weick (1993: 158) suggested, justifications may serve a sensemaking role for people who are motivated by feelings of dissonance (e.g., due to the Business Week rankings) to reexamine their organization and its identity, i.e., "Justification is often the result of focused attention that reveals new properties of a situation that unfocused attention missed." In such sensemaking contexts, selective categorizations appeared to be used primarily to bolster the credibility and believability of organizational excuses and justifications for the rankings.

Tactic 1: Categorizations highlighting alternate identity dimensions. Some business school members used categorizations to make salient valued and core identity dimensions, not emphasized by the rankings, to suggest that "they had good reason" for their poor ranking. In this way, members justified their poor showing in the survey by claiming their school maintained identity dimensions that were more important than student or recruiter satisfaction. At Carnegie-Mellon, for example, members suggested that their school's single-minded emphasis on innovation justified its poor showing on student satisfaction. One respondent noted, "We had an ethic here that placed a very high premium on innovation and creativity, but the notion of being good listeners and gathering information from students was not a part of that culture. We didn't have time for that." In a similar example, Berkeley respondents highlighted their school's entrepreneurial, West Coast identity as a distinctive categorization that justified its lower ranking: "Certain criteria in the rankings cannot be met by all schools. Berkeley's placement statistics at East Coast or Midwest firms will never be as good as, say, Harvard's, because of our students' preferences for entrepreneurial, West Coast firms." Finally, several Stanford informants noted that their distinction of recruiting and training the brightest students in the country may have actually hurt their rankings with recruiters looking for multiple placements from schools at which they interview. These informants suggested that maintaining their "elite student" distinction justified their disappointing ranking. As one informant put it, "A lot of our problems in the rankings came from recruiters' perceptions that you can't get Stanford students. They know we have very selective students. . . and we don't want to change that."

In using these types of explanations, members suggested that a low ranking based on one dimension (i.e., recruiter satisfaction) may have actually affirmed their schools' positive organizational identity on an alternate and potentially more important dimension (i.e., incredibly selective student body well-adapted to the independent, entrepreneurial careers its students would pursue). This finding parallels that observed in experimental research showing that people may sometimes affirm or admit to a negative but less central dimension of their personal identity if it simultaneously enhances a more positive, global dimension (Swann, 1987; Steele, 1988).

Tactic 2: Categorizations highlighting alternate comparison groups. Members also suggested that the comparison groups imposed on them by the Business Week survey hurt their institution's ranking. In most of these cases, members used categorizations to highlight alternate comparison groups, not emphasized by the rankings, to make the excuse that the ranking "wasn't their fault" (Tedeschi, 1981). Respondents from Berkeley and Texas, for example, used the fact that their schools were public institutions as an excuse for their poor showing. They claimed that their schools had responsibilities for funding basic research and for educating state taxpayers that prevented them from competing with higher-ranked, private schools. As one Texas respondent put it, "I personally don't think a public institution, with its multiple missions, can compete with private schools like Stanford, Harvard, or Northwestern." Similarly, a Berkeley respondent argued, "It would be much easier to be a top-10 business school if we had an agreement like at Harvard, where they're a separate professional school. There's always going to be that research aspect [because we're a state supported school]." Finally, a Texas respondent argued that the fact that they were a regional program had a negative impact on recruiters' perceptions: "There's a stereotype about us that we're only a regional school. . . . I do think the regional issue probably hurts us with recruiters."

In other cases, members used such categorizations to show that their school was similar to more highly ranked schools (i.e., they highlighted inclusion in a more prestigious comparison group) to justify their ranking. For example, Berkeley respondents noted that their students' entrepreneurial preferences put them at the same disadvantage as Stanford: "We have the same problem as Stanford, 40% of our graduates want to stay in California . . . so [East Coast] recruiters naturally get a little defensive [when we don't accept their offers]." In a similar fashion, Stanford respondents categorized their institution as a research-oriented school to highlight its similarity to other research institutions that had been poorly ranked. As one Stanford publication noted, "MIT's Sloan School, ranked twelfth in the most recent Business Week survey, has made a major commitment and contribution to research and education in manufacturing technology. On a 10 year time horizon, I believe these efforts, and others like them will yield important benefits to business. . . . We believe our own agenda should be set, in part, by field-based research." These categorizations and comparisons implied that an inferior ranking was not indicative of an inherently defective or poorly run program but, rather, of a program that adhered to valued ideals that were representative of other esteemed institutions. In this way they excused or justified their ranking in this particular survey. This behavior corresponds to research showing that, following poor performance, people may compare themselves to similarly performing others to highlight the commonness of their performance, especially among highly respected others, and to imply that such performance should not be attributed to them personally or viewed as unique or distinctive (Alicke, 1985).

Relationship between Members' Responses and Level of Perceived Threat

The above findings suggest that categorization processes are useful both to help organizational members make sense of threats to organizational identity (for themselves and external audiences) and to reduce the perceived threat to their own social identities. Thus one might expect that the greater the perceived threat to organizational and social identities, the greater would be members' use of categorization processes. In line with this argument, we observed relationships between members' level of identity dissonance and their use of the selective categorization strategies we have identified.

Categorizations highlighting alternate identity attributes and justifications. First, our data revealed that schools in which members experienced the greatest identity dissonance were also schools in which members used the greatest proportion of available categorizations to affirm the value of their core identity or to justify their poor ranking. As can be seen in Table 4, members from Berkeley, Carnegie-Mellon, Stanford, and Texas, who had the highest dissonance (as reported in Table 3), referred to an average of half of their total available bases of categorization to affirm their schools' identities. In addition, members from all of these schools highlighted valued identity dimensions as justifications for their ranking. In contrast, members from Chicago, Indiana, and Northwestern, who had the lowest dissonance, averaged only about a quarter of their available categorization schemes and did not use categorizations as excuses or justifications. Members from Cornell, who experienced moderate dissonance, also used a small portion of its available categorizations.

By affirming a positive identity on many different dimensions, members may have had a better chance of relieving the dissonance related to a single dimension (i.e., status or prestige related to rank). In this respect, Steele (1988) has proposed that people may tolerate inconsistencies in their feelings about a specific dimension of their self-concept by affirming other valued dimensions. By contrast, respondents who experienced less identity dissonance had less motivation to affirm organizational identities by highlighting alternate identity dimensions. The favorable rankings, themselves, affirmed members' perceptions of their organization's identity, especially identity perceptions related to rank or customer satisfaction. Thus members of these schools may have emphasized only those identity dimensions that were obviously congruent with their top ranking, to minimize the chance of revealing disconfirming evidence. These findings are congruent with research showing that to confirm their self-identities, people may conspicuously display self-verifying identity cues, such as titles, labels, and physical appearance (Swann, 1987), and surround themselves with others who confirm their self-views (Swann and Pelham, 1987).

Categorizations highlighting alternate comparison groups and excuses. Our data also suggest that members experiencing moderate to high levels of identity dissonance were most likely to highlight organizational categorizations that increased the salience of alternative interorganizational comparisons to affirm their organization's identity or provide an excuse for its ranking. Thus, members from Berkeley, Carnegie-Mellon, Cornell, Stanford, and Texas categorized their schools in terms of smaller and more specialized groups and claimed that the rankings unfairly lumped them into more generalized comparison groups. This strategy may have been attractive to those experiencing higher levels of dissonance because it allowed members to improve their school's relative standing on important identity dimensions, including those emphasized by the rankings, by placing it in a comparison group in which it ranked higher than the Business Week survey indicated. These higher rankings appeared consonant with members' previous perceptions of their organization's identity. This behavior is consonant with research demonstrating that people will often choose comparison groups that show them as superior to others on some cherished dimensions (Campbell, 1986).

By contrast, of the three schools whose members reported low dissonance, only Chicago invoked alternative comparisons. Members experiencing low identity dissonance may have preferred not to alter their organization's comparison groups because those changes usually meant categorizing their organization into a smaller group, whose top performers were viewed as less prestigious than top performers of larger groups. Members seemed willing to categorize their organizations into more exclusive groups only if they could not achieve a high ranking in a more inclusive group. This suggests that members paid attention to the relative prestige invoked by interorganizational comparisons and that they sometimes preferred being a big fish in a big pond to being the biggest fish in a much smaller pond.

DISCUSSION

Our inductive analysis of how members respond to organizational identity threats suggests several important insights. First, we found evidence that many business school members perceived the Business Week rankings as a threat to their organization's identity, even when their school was highly ranked, because it devalued central and cherished identity dimensions and refuted prior claims of positional status. Second, we found that organization members commonly used selective categorizations to reemphasize positive perceptions of their organization's identities, for themselves and external audiences, by highlighting identity dimensions and interorganizational comparison groups not emphasized by the rankings. These tactics functioned to deflect members' attention away from threatened dimensions of their organization's identity, rather than addressing threatened dimensions directly. Finally, we found that members' identity management tactics were related to their level of perceived threat, as indicated by variation in their expressed dissonance over the rankings.

One can use the metaphor of a microscope, as it is used to view material on a slide, to visualize the above organizational identity management tactics and the identity affirmation and sensemaking functions they serve. Using this metaphor, selective categorization can be likened to increasing or decreasing magnification to manipulate the field of view available to the perceiver and, by implication, the inclusiveness or exclusiveness of comparisons, as shown in Figure 1. Thus, the first strategy, of highlighting cherished identity attributes not recognized by the rankings, is like using a high level of magnification to focus on a single organization and moving the slide around to highlight or make salient alternate facets of that organization's identity. Moving toward positive identity attributes and away from tarnished attributes can help members establish a more positive overall perception. In contrast, a lower level of magnification can focus on a set of organizations. Thus, the second strategy, of highlighting alternate comparison groups, involves moving the slide at this lower level of magnification to focus on different subsets of organizations and their interrelationships. By placing the organization in a more diffuse, broader visual field, its relationship to other organizations and its perceived similarity and distinctiveness can be manipulated. By focusing on a social comparison group in which the organization has high status, members can affirm their perceptions of their organization's relative value and prestige. In both cases, the principle of perceptual contrast influences perceived attractiveness, distinctiveness, and the commonality of objects (cf., Tajfel, 1969; Cialdini, 1984). As the microscope metaphor suggests, selective categorization is a gestalt-like process, in which elements of figure and ground come forward and recede in response to the perceiver's motives or goals. Accordingly, positive perceptions depend on their context.

The prominent use of affirmation and sensemaking processes suggested by this metaphor contrasts with previous work on organizational impression management, which suggests that people primarily employ repair tactics when responding to identity-threatening predicaments. While the present research does show evidence of excuses and justifications in response to organizational identity threats, it also suggests that members can attenuate or mitigate organizational identity threats simply by making salient other legitimate and competing dimensions along which the organization should be evaluated or construed. Thus, rather than having to defend, deny, or explain a particular external claim--and perhaps have to change their perception of the organization's identity in response to it--members can emphasize other ways in which the organization is intrinsically good or functioning well. Similarly, rather than respond directly to external attributions about an organization's status relative to other organizations, members can invoke alternate categorization schemes that make salient the ways in which the organization is different from or better than other organizations with which it is being compared.

The significance of these findings rests on several unique attributes of the Business Week rankings that differentiate them from other forms of external attribution that can threaten an organization's identity (such as stock price forecasts or industry expert polls). First, a distinctive feature of Business Week's attributions about business schools is that they are heavily dependent on the subjective interpretations of only one subset of the organization's total membership--and a subset that consisted of short-term, temporary members at that (i.e., second-year MBA students). The Business Week rankings also did not include any evaluation inputs from longer-term members, such as faculty and administrators, nor did they take into consideration a number of important, central traits of these institutions, such as research productivity, and contributions to other missions of the university, such as undergraduate teaching, that had been used in prior evaluations and rankings. Because of these characteristics, the Business Week rankings could fluctuate greatly, depending on the idiosyncratic approbations and grievances of a small and narrow sample of organizational members, all of whom were soon to become ex-members. Yet the rankings carried all the weight of a long-term measure that had important reputational and self-esteem implications for current and more permanent members. As a consequence, these members may have felt helpless to influence this important measure and may have perceived their identities to be more threatened by the rankings than by other forms of external attribution.

At the same time, compared with other kinds of external threats studied in the impression management literature, the Business Week rankings may have seemed less life-threatening to the specific organizations we studied because the rankings did not attack widely accepted perceptions of their legitimacy (i.e., the appropriateness of their structures, procedures, and goals). All of the schools we examined were ranked in the upper echelons of U.S. business schools, clearly enjoying high status and prestige relative to other business schools throughout the United States. In this respect, the threat posed by the rankings was substantially different from the kinds of external threats that have affected many other industries, such as the tobacco and cattle industries (Rosenblatt, 1994; Elsbach, 1994). In these industries, companies are facing intense public scrutiny about the legitimacy of their products and even their right to exist. In contrast, the Business Week rankings merely attacked or called into question members' perceptions of the value and distinctiveness of a school's central identity dimensions. As a result, members may have felt that the Business Week rankings provided somewhat misleading and incomplete characterizations of their institutions but were not completely wrong (i.e., they fairly accurately report the perceptions of a small group of students about a narrow range of student-related concerns such as teaching and recruitment). The Business Week rankings thus represented a strong organizational identity threat, but a rather weak organizational legitimacy threat. Consequently, members may have been more motivated to make sense of the rankings to affirm their preexisting perceptions of their organization's identity rather than to explain or justify the rankings to external audiences. Our findings about these types of responses have numerous theoretical and managerial implications.

Theoretical Implications

Organizational identity theory. Our findings have a number of implications for organizational identity theory. First, they provide support for recent arguments that a significant psychological interdependence exists between individuals' social identities and their perceptions of their organizations' identities. As Dutton and Dukerich (1991: 550) proposed, "The relationship between individuals' senses of their organizational identity and image and their own sense of who they are and what they stand for suggests a very personal connection between organizational action and individual motivation." These arguments imply that because members' own social identities and self-esteem are intimately connected to the identity and reputation of their organizations, they care about how their organizations are described and also how they compare with other organizations. In accord with such arguments, our results document the cognitive distress or identity dissonance people experience when they think their organization's identity is threatened by what they perceive as inaccurate descriptions or misleading (and, by implication, unfair) comparisons with other organizations.

Second, our findings extend current theories of organizational identity by showing that in response to such cognitive distress, members may restore and affirm positive self-perceptions by affirming alternate dimensions of their organization's identity or highlighting their organization's membership in alternate comparison groups. These responses are quite distinct from results reported in research on social identity and self-affirmation. On the basis of those studies, we would expect people to attempt to restore positive self-perceptions and social identities by highlighting other personal or social categorizations (i.e., membership in other organizations) as a means of distancing themselves from the tarnished identity of their business schools. Yet we observed almost no evidence of this sort in our study. These results show that organizational identity affirmation is distinct from previously defined self-affirmation and social identity repair tactics because of its emphasis on affirming organizational traits rather than individual traits.

Third, the results of this study imply that identity affirmation is distinct from externally directed image management. The most compelling evidence for this distinction came from several instances in which clearly image-enhancing outcomes (e.g., receiving a top ranking) were construed by organizational members as identity threatening because they implied that other central and valued dimensions of their organization were unimportant or undervalued. If image-management concerns were the only factor motivating the selective categorizations we observed, then members had little reason in these instances to manage their organization's images or defend its general prestige or status. Instead, however, we found that members' responses were directed at highlighting the value of key organizational dimensions they perceived as neglected or undervalued. Such responses seemed to have had more to do with attempting to reaffirm to themselves a school's positive identity in light of the rankings than with simply enhancing their school's image to external audiences. In advancing this argument, we should emphasize that it is not possible to disentangle completely self-affirmation from self-presentational explanations for our findings. Nor, frankly, is it necessary to do so. As numerous theorists have previously argued, these two motives are not logically incompatible and, in fact, both motives are typically present in such situations (e.g., Tetlock and Manstead, 1985). Selective categorization processes, when they are used as part of members' public accounts, are probably directed simultaneously at shaping and enhancing both their own and others' perceptions of their organization's identity.

Finally, and perhaps most importantly, the present research contributes to our understanding of the specific cognitive tactics that organizational members use to maintain and affirm organizational identities. Our findings suggest that by selectively highlighting organizational categorizations, members are capable not only of affirming positive organizational identities when some dimensions are threatened, but also of influencing theirs and others' perceptions of the validity of favorable and unfavorable categorizations and social comparisons (e.g., the validity of organizational rankings).

Throughout our study, we were struck with the pervasive and creative use of selective categorization processes to maintain positive perceptions of an organization's identity. One example, brought to our attention after we completed our study, involved a school that heard it was to be ranked as one of the top 50 U.S. business schools in the next U.S. News and World Report. A memo from the dean sent to all faculty and staff proclaimed that the school was the "youngest public school" ever to be so ranked (emphasis added). The memo further read, "With the possibility of TV cameras wandering our halls, I would like to encourage you to wear business attire on that day."

Social identity and self-affirmation theories. While a primary goal of our research was to contribute to organizational theory by importing insights from social identity and self-affirmation research, our findings return something to these theories as well. First, our findings suggest that one way people can protect and affirm their own social identities is by selectively categorizing their organizations. We found that most members elected to categorize their organization selectively rather than categorize themselves in a different way. As noted above, these findings run contrary to what social identity, self-affirmation, and impression management theories have usually found. They also support individual-level findings suggesting that people often tolerate inconsistencies in self-perceptions by affirming other valued dimensions of themselves (Steele, 1988). Organizational categorization thus provides another route to self-affirmation, and one not addressed in any previous research. Broadly construed, our findings thus highlight the adaptive role of organizational categorization as a route to self-esteem and social identity maintenance, especially with respect to influencing individuals' perceptions of the relative status or prestige of their organizations. As Frey and Ruble (1990: 168) have argued, "healthy [psychological] functioning may depend on the ability to exhibit flexibility in the choice of evaluative comparisons in order to maintain a sense of competence and high self-esteem." To paraphrase Frank (1985), even though organizational members cannot always choose the best pond for themselves and their institutions, they at least have considerable cognitive flexibility in creating the perception that they are a reasonably large-sized frog in a reasonably good pond.

Impression management theory. Finally, our findings also have important implications for theories of organizational impression management. First, our findings explicate some of the conditions necessary to motivate impression management. Our findings suggest that, following an organizational identity threat, if alternate organizational categorizations are available and organizational legitimacy is not threatened, sensemaking motivations may be greater than impression management motivations. Thus the rather weak threat to organizational legitimacy posed by the Business Week rankings, coupled with the ready availability of organizational recategorization as a low-cost identity affirmation strategy, may explain why we observed so little evidence in our data of what affirmation theorists call spontaneous attributional search and generation of causal accounts (see Weiner, 1986). If people can make adequate sense of a threat and resolve dissonance surrounding it simply by affirming alternate identity dimensions that already exist and are readily available, then the need to generate detailed causal explanations for the event may be considerably attenuated. Identity affirmation processes thus may sometimes cause individuals to cut short such attributional search. As attribution theorists have frequently noted, generating complete causal attributions is an effortful process, sometimes likened to conducting a multivariate analysis of variance inside one's head. People engage in such processes only occasionally. Our results shed some light on the conditions under which people might be motivated to do so: when alternative selective categorizations are not available as a means of sensemaking and when organizational legitimacy is threatened.

Yet in those cases in which there is a motivation for impression management, our findings also suggest that organizational members may include selective organizational categorizations in their excuses and justifications to bolster the credibility of these externally directed accounts. Highlighting selective categorizations reminds both organizational members and outside audiences of long-held identity dimensions that should be considered in forming perceptions of the organization. If audiences agree that these identity dimensions are valuable, such categorizations may provide evidence of a school's favorable identity and thus make excuses and justifications more believable (Elsbach, 1994). Recognizing organizational categorizations as potential evidence or "content" for verbal accounts is a new addition to theories of organizational impression management.

A theory of organizational identity management. Theory and research on organizational identity is still in its early stages, and explicating the basic cognitive processes that constitute the underpinnings of identity is an important first step in integrative theory linking organizational and individual identity processes. Our findings can be summarized in a new framework of organizational identity management that integrates them with insights from social identity and impression management theories. According to this framework, when organizational members perceive that their organization's identity is threatened, they try to protect both personal and external perceptions of their organization as well as their perceptions of themselves as individuals. As Table 5 indicates, this identity management perspective builds on and integrates findings from the current study with insights from social identity theory and impression management theory in several ways. First, it indicates that members are motivated by both self-affirmation concerns and impression-management concerns when responding to threats to their organization's identity. Second, it notes that identity management appears to follow events that devalue or refute identity dimensions members cherish both as a part of the organization's enduring identity and as a part of their own social identities. Third, it portrays organizational identity management as a pervasive activity that is carried out not only by organization members who formally occupy public relations roles but also by any member who identifies with the organization. Finally, it describes tactics used in identity management that involve both selective organizational categorizations (commonly used in self-affirmation models) and verbal accounts (commonly used in impression management models).

Managerial Implications

The results of this research have a number of practical implications. First, categorization processes can contribute to important sensemaking activities in organizations (Weick, 1995). As March (1994: 71) observed, "Organizations shape individual action both by providing the content of identities and rules and by providing appropriate cues for invoking them." Categorization processes can play a central role in this process, especially in terms of linking and helping rationalize organizational members' cognitions and actions (Weick, 1993; March, 1994). Organizational categorizations may allow leaders to focus members' attention on what they should be doing and why. The findings from this study thus suggest a rich set of tools for managers involved in the symbolic management of their organizations. As Pfeffer (1981: 26) noted, "Every organization has an interest in seeing its definition of reality accepted . . . for such acceptance is an integral part of the legitimization of the organization and the development of assured resources." Much of the research on this symbolic management process over the past ten years has focused on the use of verbal accounts, including excuses, explanations, and justifications (e.g., Elsbach and Sutton, 1992; Ginzel, Kramer, and Sutton, 1993; Elsbach, 1994) and the manipulation of causal attributions (e.g., Salancik and Meindl, 1984) to explain identity-threatening organizational events, but categorization processes may offer advantages over verbal accounts.

Our findings also suggest that managers may use categorization processes for symbolic management and sensemaking, both with respect to helping people inside the organization make sense of what their organization is about and in explaining it to external constituents and audiences. Many of the identity threatening predicaments that have been studied in prior organizational research were unique, organization-specific events. For example, the explosion of the space shuttle Challenger, the Union Carbide Bhopal crisis, and the Exxon Valdez oil spill were distinctive events that threatened the identity of a single organization. In contrast, the Business Week rankings of U.S. business schools simultaneously affected multiple organizations, forcing comparative appraisals along multiple dimensions. In situations having this greater complexity, selective categorizations and interorganizational comparisons can be powerful and flexible tactics for organizations and their members to maintain positive identities. By selectively directing and focusing attention, categorization processes heighten the salience of some dimensions while deflecting attention away from others. They can be used therefore to influence perceptions of positivity and negativity, similarity and dissimilarity, uniqueness and distinctiveness, or commonality and difference. In this respect, they function much like other kinds of general framing processes that have been found to influence the perceived positivity and negativity of events (Kahneman and Tversky, 1984; Brickman and Bulman, 1985).

Finally, our findings suggest that categorization processes may help organizations to change or reshape their identities. Strategic management theorists have recently proposed a theory of "strategic dissonance," suggesting that managers may purposefully take advantage of distress related to incongruencies between an organization's strategic intent and managers' strategic action. In their discussion of managing strategic dissonance, these theorists imply that organizational identity management may play a role. They suggest that "Top management must use information that is generated by strategic dissonance when trying to discern the true new shape of the company. . . . It must be a realistic picture grounded in the company's distinctive competencies--existing ones or new ones that are already being developed. . . . Getting through the period of immense change requires reinventing--or perhaps rediscovering--the company's identity" (Burgelman and Grove, 1996: 20). Our findings support and extend these notions by suggesting that managers may use categorization processes following strategic dissonance to reinvent or rediscover their firm's identity.

Study Limitations

There are some limitations to our study that need to be acknowledged, in particular, issues of external validity and the generalizability of our results. First, our study focused on schools that, in absolute terms, had generally fared rather well in the Business Week rankings. All of the schools in our sample were relatively elite performers relative to the population of business schools throughout the U.S. For these highly ranked schools, their legitimacy was not directly threatened by the rankings. As a result, the impact of the threat on the schools we studied may have been very different from other business schools, especially those who failed even to obtain a ranking. For business schools that were ranked much lower by Business Week (or not ranked at all), the survey results could threaten their legitimacy and even their very existence. A small, regional business school, for example, might have claimed to be a good school comparable to other elite institutions, but if it was not even ranked by the survey, external constituencies might begin to question whether the university should continue to fund an MBA program and compete for increasingly scarce MBA students. As Meyer Feldberg, dean of Columbia's Business School, noted, as long as the survey creates a contest for status and reputation, there is the implicit long-term threat that "the strong schools will get stronger and the weak will get wiped out" (quoted in Fombrun, 1996: 267, emphasis added). Similarly, some identity threats may strike so close to the central or core identity of an institution that selective categorization tactics alone may be insufficient. For example, when the space shuttle Challenger exploded, NASA's identity as the "can do" organization was severely damaged. Once such core identity attributes are tarnished, categorizations may not only be inadequate, they may be dysfunctional and approach denial.

In evaluating the generalizability of our arguments about the motives underlying selective categorization, we need to emphasize that our analysis has focused almost exclusively on identity affirmation and restoration following threat. But there may be other motives for using categorization strategies, including the desire for accurate self-assessment and self-improvement. Researchers have long argued that there are many reasons for engaging in categorization processes, including the desire to obtain information that will contribute to realistic self-appraisal and to generate informative and useful social comparisons (Tajfel, 1969; Brewer, 1991). On balance, however, we were struck by how infrequently such goals or motives were expressed in our interviews. Instead, the preponderance of our data pointed toward more defensive (i.e., self-protective) uses of categorization and comparisons. Repeatedly in our interviews we observed evidence of organizational members attempting to reduce the dissonance between their perceptions of their organization's identities and the identity attributed to them by the Business Week survey results.

Conclusion

We have portrayed selective categorization processes primarily as useful cognitive tactics for helping organizational members both make sense of organizational identity threats and affirm positive organizational and social identities. But categorization processes are basic cognitive processes that can contribute to a number of other important organizational processes and outcomes, including shaping perceptions of emerging opportunities and future ventures. Reger and her colleagues (1994), for example, noted the important role organizational identity beliefs play in a constructive change process. As Reger et al. (1994: 33) asserted, "Organizational members interpret new management initiatives through their existing mental models. . . . A particularly powerful mental model is the set of beliefs members hold about the organization's identity." We view categorization processes as one of the fundamental building blocks of such mental models.

We have argued that using selective categorization processes creatively can help organizations decide not only where emerging opportunities lie, but also what the appropriate and useful responses to them are. Further, we have noted how these same processes help organizational members understand their organizations and their relationship to the external environment. Our framework of identity management thus provides the foundation for a more general model that emphasizes the functional role categorization processes play in attempts by organizations and their members to make sense of themselves and their environments.

* We wish to thank Susan Ashford, Jim Baron, Marilynn Brewer, Michael Cohen, Jane Dutton, Janet Dukerich, Ben Hanna, Joanne Martin, Michael Morris, Jeff Pfeffer, Bob Sutton, Michelle Walker, Jim Walsh, Janet Weiss, and Dave Whetten for their comments and contributions to this paper. Thanks also go to Keith Murninghan and the three anonymous ASQ reviewers for their susbtantial and constructive comments. Earlier versions of this research were presented at the National Academy of Management annual meetings, University of Michigan;s Graduate School of Business, Cornell University's Johnson Graduate School of Management, University of Texas' Graduate School of Business, and Stanford Asilomar Conference on Organizations. We are grateful for comments provided by participants at those conferences. Finally, we wish to thank the business school members who generously gave their time for this study. This study was supported by funding from Emory University's Goizueta Business School, Standford University's Graduate School of Business, and a James and Doris McNamara Faculty Fellowship.

Table 1

Business Schools Selected for Study

Place in rankings Change in rank, 1990 to 1992 Top tier Middle tier Bottom tier Moved up Chicago Indiana 4right arrow2 15right arrow8 Little change Northwestern Stanford Cornell 1right arrow1 9rightarrow7 16right arrow14 Berkeley 19right arrow18 Carnegie-Mellon Texas 9right arrow17 18right arrowout

Table 2.

Dimensions of Business School Identities[*]

Legend for Chart: A - School B - Identity Dimension C - Example A B C Berkeley 1. Participatory culture "At Cal, students really get involved in everything" 2. Diverse students "Haas' richest asset is the diversity of its students" 3. Creative students "We select students who are intrinsically more creative" 4. Global program "special opportunities to examine . . . managing in a global environment" 5. Interdisciplinary "I don't believe any other B-school has as many programs that are a program product of real joint planning with other units on campus" 6. Entrepreneurial "Haas differs from other top schools in . . . entrepreneurship" 7. Public institution "There is no other public institution of such quality" 8. Renowned university "part of one of the world's preeminent institutions of teaching/research." 9. High value "A top dollar education for a bargain basement price" 10. Elite internationally "Internationally recognized for our quality of faculty and programs" Carnegie-Mellon 1. Small/friendly culture "The thing I remember most is the size and intimacy of the place" 2. Rigorous/technical "There is one word to describe the school--intense" 3. Innovative program "innovation and the transfer of knowledge are part of our heritage" 4. Elite performer "[We] rank among the most respected of business schools in the world" Chicago 1. Quantitative program "We're going to keep our pocket protectors" 2. Academic values "At Chicago we're interested in theory" 3. Elite performer "for decades one of the leaders in graduate business education" 4. Innovative program "the home of many of this century's innovations in business education"' 5. Research institution "an integral part of a major research institution" Cornell 1. Small/friendly culture "Because [Corneli's] small, you actually get to know people" 2. Teaching values "[faculty have] made a commitment to teaching" 3. Strong alumni "not-so-secret weapon for graduates--an alumni network of 170,000" 4. Renowned university "[The business school is] . . . Part of a world-recognized institution" 5. Top-tier school "The respected position of the Johnson School . . . was important to me" Indiana 1. Friendly culture "This isn't a snake pit, and a lot of schools are. It's a nice place to be" 2. Participatory culture "MBA students at Indiana have a tradition of involvement" 3. Teaching values "Professors at Indiana are willing to teach well" 4. Work ethic "our graduates are willing to roll up their sleeves and do the work" 5. Innovative "Indiana is truly a leader embracing change" Northwestern 1. Group culture "[Kellogg's] driving esprit de corps is unique among business schools" 2. Elite performer "[in] a select group ranked outstanding by scholars and practitioners" 3. Continuous improver "many other top schools appear complacent by comparison" Stanford 1. Friendly culture "A spirit of cooperation is an integral part of the culture" 2. Balanced program "Teaching is a natural companion to research" 3. Public management "A special feature of the Stanford MBA is its Public Management program Program" 4. Elite students [listed as a distinctive characteristic] "the world class students" 5. Diverse culture "People here come from all types of backgrounds" 6. Research values "The engine of the Stanford MBA is the faculty's research" Texas 1. Diverse culture "At Texas, there is no such thing as a typical business student" 2. Participatory culture "One of the best aspects is how involved students are" 3. Academic values "students are genuinely concerned about receiving a good education" 4. Public institution "[MBAs have] opportunities unique to a large university" 5. Top-tier performer "Having established a position of national prominence, [the b-school] continues its commitment to professional excellence in business education" 6. High-value program "For the price of education, there is not a finer school in the country" 7. Regional program "A top-20 business school located in the southwest" * Based on identity statements from business school catalogues printed in 1987-88, 1989-90, and 1991-92, and Business Week's The Best Business Schools, printed in 1991.

Table 3

Evidence of Identity Dissonance Following the Business Weak Rankings[*]

Legend for Chart: A - Ranking event B - High dissonance C - Moderate dissonance D - Low dissonance Berkeley Stability at bottom A/F/S "doesn't reflect true rank" Carnegie-Mellon Significant, unexpected move to bottom, Chicago Small but significant move at top. A/F/S/ "It was a travesty" a/f "I just don't see how [Northwestern] can be consistently ranked #1" Cornell Insignificant move in middle. Indiana Significant, unexpected move to top. A/S "unhappy out of top 10" f/s "Business Week detailed none of the changes that have taken place" Northwestern Stability at top. a/f "It made us sound like we were standing still." Stanford Disappointing stability in middle. A/F/S "most don't think we were ranked properly" Texas Significant move down to our of top-20 A/F/S "I applied to a top 20 school and this is not a top 20 school' * A, a strong or moderate evidence from administrators: F, f = strong or moderate evidence from faculty; S, s = strong or moderate evidence from students. Strong evidence indicates that all of the respondents in a category made similar statements; moderate evidence indicates that about half the respondents made such statements.

Table 4

Selective Categorization Tactics in Response to Organizational Identity Threats[*]

Legend for Chart: A - Alternate Identity Dimensions B - Alternate Comparison Groups C - Type D - Group Berkeley Identity High value A Entrepreneur AFS West Coast schools affirmations Entrepreneur AFS Public A Public schools Public A Participatory AFS Sensemaking Participatory AS Public A Public schools Entrepreneur/ A creative Carnegie-Mellon Identity Innovative AFS Rigor/tech A Quantitatively oriented schools affirmations Rigor/tech F Small AS Innovative AS Sensemaking Innovative A Chicago Identity Academic AFS Academic A Research-oriented schools affirmations Innovative A Cornell Identity Small AFS Small AFS Small schools affirmations Sensemaking Small AFS Small schools Indiana Identity Innovative AFS affirmations Northwestern Identity Continuous AFS affirmations improvement Stanford identify Public Mgmt AFS Elite student AFS Top-tier schools affirmations Research AFS Research AFS Research-oriented schools Texas identity Public AF Regional AS Public schools affirmation Regional A Public AF Public schools Academic A Sensemaking Regional AS Public A Public schools Regional AF Southern schools

* A, F, S. = strong evidence from administrators, students, and faculty, respectively. Strong evidence indicates that most of the respondents or records data from that category (i.e. administrators, faculty, or students) used that tactic.

Table 5

Theoretical Perspective on Perception Management

Legend for Chart: A - Perception Management Variable B - Theoretical Perspective--Organizational impression management C - Theoretical Perspective--Social identity maintenance D - Theoretical Perspective--Organizational identity management[*] 1. Primary target of perception management External perceptions of an organization's legitimacy (Elsbach, 1994) Individuals' perceptions of self based on their association with social groups (Ashforth and) Mael, 1989). Members' and audiences' perceptions of the organization's identity and members' perceptions of self based on their affiliation with the organization. 2. Primary motivation for perception management Event that casts doubt on the organization's legitimacy (Marcus and Goodman, 1991). Context or event that highlights a person's association with a negatively viewed social group (Hogg and Abrams, 1988). Event that devalues or disputes cherished identity dimensions that are part of the organization's enduring identity and part of the member's social identity. 3. Who manages perceptions A member spokesperson (Sutton and Callahan, 1987.) Any person who identifies with the group (Abrams and Hogg, 1990) Any member who identifies with the organization. 4. Tactics used Verbal accounts. Display of legitimate organizational structures and procedures Self-categorizations, Social comparisons. Organizational categorizations. Verbal accounts containing organizational categorizations.

[*] Based on findings from the current study.

CHART: Figure 1. Identity affirmation and sensemaking tactics.

REFERENCES

Abrams, Dominic, and Michael A. Hogg 1990 Social Identity Theory. Constructive and Critical Advances. New York: Springer-Verlag.

Albert, Stuart, and David A. Whetten 1985 "Organizational identity." In B. M. Staw and L. L. Cummings (eds.), Research in Organizational Behavior, 7: 263-295. Greenwich, CT: JAI Press.

Alicke, Mark D. 1985 "Global self-evaluation as determined by the desirability and controllability of trait adjectives." Journal of Personality and Social Psychology, 49:1621-1630.

Ashforth, Blake, and Fred Mael 1989 "Social identity and the organization." Academy of Management Review, 14: 20-39.

Brewer, Marilynn B. 1991 "The social self: On being the same and different at the same time." Personality and Social Psychology Bulletin, 15: 475-482.

Brewer, Marilynn B., and Roderick M. Kramer 1986 "The psychology of intergroup attitudes and behavior." Annual Review of Psychology, 43: 219-243.

Brickman, Philip, and Ronnie J. Bulman 1985 "Pleasure and pain in social comparison." In J. M. Suls and R. L. Miller (eds.), Social Comparison Processes: Theoretical and Empirical Perspectives: 149-186. New York: Wiley.

Brown, Rupert, and Jennifer Williams 1984 "Group identification: The same thing to all people?" Human Relations, 37: 547-564.

Burgelman, Robert A., and Andrew S. Grove 1996 "Strategic dissonance." California Management Review, 38: 8-28.

Campbell, Jennifer D. 1986 "Similarity and uniqueness: The effects of attribute type, relevance, and individual differences in self-esteem and depression." Journal of Personality and Social Psychology, 50: 281-294.

Cialdini, Robert B. 1984 Influence: The New Psychology of Modern Persuasion. New York: Quill.

Cialdini, Robert B., Richard J. Borden, Avril Thorne, Marcus Randall Walker, Stephen Freeman, and Lloyd Reynolds Sloan

1976 "Basking in reflective glory: Three (football) field studies." Journal of Personality and Social Psychology, 34: 463-476.

Dutton, Jane E., and Janet M. Dukerich 1991 "Keeping an eye on the mirror: Image and identity in organizational adaptation." Academy of Management Journal, 34:517-554.

Dutton, Jane E., Janet M. Dukerich, and Celia V. Harquail 1994 "Organizational images and member identification." Administrative Science Quarterly, 39: 239-263.

Dutton, Jane E., and Wendy J. Penner 1993 "The importance of organizational identity for strategic agenda building." In J. Hendry and G. Johnson (eds.), Strategic Thinking: Leadership and the Management of Change: 89-113. New York: Strategic Management Society, Wiley.

Eisenhardt, Kathleen M. 1989 "Building theory from case study research." Academy of Management Review, 14: 532-550.

Elsbach, Kimberly D. 1994 "Managing organizational legitimacy in the California cattle industry: The construction and effectiveness of verbal accounts." Administrative Science Quarterly, 39: 57-88.

Elsbach, Kimberly D., and Robert I. Sutton 1992 "Acquiring organizational legitimacy through illegitimate actions: A marriage of institutional and impression management theories." Academy of Management Journal, 35: 699-738.

Fombrun, Charles J. 1996 Reputation: Realizing Value from Corporate Image. Boston: Harvard Business School Press.

Frank, Robert 1985 Choosing the Right Pond: Human Behavior and the Quest for Status. New York: Oxford.

Frey, Karin S., and Diane N. Ruble 1990 "Strategies for comparative evaluation: Maintaining a sense of competence across the life span." In R. J. Sternberg and J. Kolligan (eds.), Competence Considered: 167-189. New Haven: Yale University Press.

Gaertner, Samuel L., Jeffrey Mann, John F. Dovidio, and Audrey J. Murrell 1990 "How does cooperation reduce intergroup bias?" Journal of Personality and Social Psychology, 59: 692-704.

Gaertner, Samuel L., Jeffrey Mann, Audrey Murrell, and John F. Dovidio 1989 "Reducing intergroup bias: The benefits of recategorization." Journal of Personality and Social Psychology, 57: 239-249.

Ginzel, Linda E., Roderick M. Kramer, and Robert I. Sutton 1993 "Organizational impression management as a reciprocal influence process: The neglected role of the organizational audience." In L. L. Cummings, and B. M. Staw (eds.), Research in Organizational Behavior, 15: 227-266. Greenwich, CT: JAI Press.

Glaser, Barney, and Anselm L. Strauss 1967 The Discovery of Grounded Theory: Strategies for Qualitative Research. London: Wiedenfeld and Nicholson.

Goethals, George R., David M. Messick, and Scott T. Allison 1991 "The uniqueness bias: Studies of constructive social comparison." In J. Suls and T. A. Wills (eds.), Social Comparison: Contemporary Theory and Research: 149-176. Hillsdale, NJ: Erlbaum.

Hall, Carl T. 1990 "Why business school rating can rankle." San Francisco Chronicle, March 26: C1.

Hay, Tina M. 1992 "Education's love-hate relationship. Colleges abhor the rankings--but adore the attention." Case Currents, Sept.: 14-20.

Hinkle, Steve, and Rupert J. Brown 1990 "Intergroup comparisons and social identity: Some links and lacunae." In D. Abrams and M. A. Hogg (eds.), Social Identity Theory: Constructive and Critical Advances: 48-70. New York: Springer-Verlag.

Hogg, Michael A., and Dominic Abrams 1988 Social Identifications: A Social Psychology of Intergroup Relations and Group Processes. London: Routledge.

Kahneman, Daniel, and Amos Tversky 1984 "Choices, values, and frames." American Psychologist, 39: 341-350.

Kramer, Roderick M. 1993 "Cooperation and organizational identification." In J. K. Murnighan (ed.), Social Psychology in Organizations: 244-269. Englewood Cliffs, NJ: Prentice-Hall.

Kramer, Roderick M., and Marilynn B. Brewer 1984 "Effects of group identity on resource use in a simulated commons dilemma." Journal of Personality and Social Psychology, 46: 1044-1057.

Mael, Fred, and Blake E. Ashforth 1992 "Alumni and their alma mater: A partial test of the reformulated model of organizational identification." Journal of Organizational Behavior, 13: 103-123.

March, James G. 1994 A Primer on Decision Making. New York: Free Press.

Marcus, Alfred A., and Robert S. Goodman 1991 "Victims and shareholders: The dilemmas of presenting corporate policy during a crisis." Academy of Management Journal, 34: 281-305.

Miller, Carol T. 1982 "The role of performance-related similarity in social comparison of abilities: A test of the related attributes hypothesis." Journal of Experimental Social Psychology, 18: 513-523.

Pfeffer, Jeffrey 1981 "Management as symbolic action." In L. L. Cummings and B. M. Staw (eds.), Research in Organizational Behavior, 3: 1-52. Greenwich, CT: JAI Press.

Putka, Gary 1990 "Colleges rankled by how they're ranked." Wall Street Journal, March 23: B1.

Reger, Rhonda K., John V. Mullane, Loren T. Gustafson, and Samuel M. DeMarie 1994 "Creating earthquakes to change organizational mindsets." Academy of Management Executive, 8: 31-46.

Robinson, Peter 1995 Snapshots from Hell: The Making of an MBA. New York: Time-Warner.

Rosenblatt, Roger 1994 "How tobacco executives live with themselves." Business and Society Review, Spring: 22-34.

Salancik, Gerald R., and James R. Meindl 1984 "Corporate attributions as strategic illusions of management control." Administrative Science Quarterly, 29: 238-254.

Salovey, Peter, and Judith Rodin 1984 "Some antecedents and consequences of social-comparison jealousy." Journal of Personality and Social Psychology, 47: 780-792.

Schlenker, Barry R. 1980 Impression Management: The Self-concept, Social Identity, and Interpersonal Relations. Monterey, CA: Brooks/Cole.

Staw, Barry M., Pamela I. McKechnie, and Sheila M. Puffer 1983 "The justification of organizational performance." Administrative Science Quarterly, 28: 582-600.

Steele, Claude M. 1988 "The psychology of self-affirmation: Sustaining the integrity of the self." Advances in Experimental Social Psychology, 21: 261-302. New York: Academic Press.

Sutton, Robert I., and Anita L. Callahan 1987 "The stigma of bankruptcy: Spoiled organizational image and its management." Academy of Management Journal, 30: 405-436.

Swann, William B. 1987 "Identity negotiation: Where two roads meet." Journal of Personality and Social Psychology, 53: 1038-1051.

Swann, William B., Jr., and Brett W. Pelham 1987 "The social construction of identity: Self-verification through friend and intimate selection." Unpublished manuscript, Department of Psychology, University of Texas, Austin.

Tajfel, Henri 1969 "Cognitive aspects of prejudice." Journal of Social Issues, 25: 79-97.

1982 Social Identity and Intergroup Relations. Cambridge: Cambridge University Press.

Tajfel, Henri, and John C. Turner 1979 "An integrative theory of intergroup conflict." In W. G. Austin and S. Worchel (eds.). The Social Psychology of Intergroup Relations: 7--24. Monterey, CA: Brooks/Cole.

Tajfel, Henri, and A. L. Wilkes 1973 "Classification and quantitative judgment." British Journal of Social Psychology, 54: 101-114.

Taylor, Shelley E., Joanne V. Wood, and Rosemary R. Lichtman 1983 "It could be worse: Selective evaluation as a response to victimization." Journal of Social Issues, 39: 19-40.

Tedeschi, James T. 1981 Impression Management Theory and Social Psychological Research. New York: Academic Press.

Tedeschi, James T., and Valerie Melburg 1984 "Impression management and influence in the organization." In S. B. Bacharach, and E. J. Lawler (eds.), Research in the Sociology of Organizations. 3: 31-58. Greenwich, CT: JAI Press.

Tetlock, Philip E., and Antony S. R. Manstead 1985 "Impression management versus intrapsychic explanations in social psychology: A useful dichotomy?" Psychological Review, 92: 67-82.

Turner, John C. 1987 Rediscovering the Social Group: A Self-categorization Theory. Oxford: Basil-Blackwell.

Weick, Karl 1993 "Sensemaking in organizations." In J. K. Murnighan (ed.), Social Psychology in Organizations: 10-37. Englewood Cliffs, NJ: Prentice-Hall.

1995 Sensemaking in Organizations. Thousand Oaks, CA: Sage.

Weiner, Bernard 1986 An Attributional Theory of Motivation and Emotion. New York: Springer-Verlag.

Wheeler, Ladd 1966 "Motivation as a determinant of upward comparison." Journal of Experimental Social Psychology, Supplement, 1: 27-31.

Wheeler, Ladd, and Miron Zuckerman 1977 "Commentary." In J. M. Suls and R. L. Miller (eds.), Social Comparison Processes: Theoretical and Empirical Perspectives: 335-357. Washington, DC: Hemisphere.

Whetten, David, Debra Lewis, and Leann Mischel 1992 "Towards an integrated model of organizational identity and member commitment." Paper presented at the Academy of Management Annual Meeting. Las Vegas.

Wood, Joanne V. 1989 "Theory and research concerning social comparisons of personal attributes." Psychological Bulletin, 106: 231-248.

Wood, Joanne V., Shelley E. Taylor, and Rosemary R. Lichtman 1985 "Social comparison in adjustment to breast cancer." Journal of Personality and Social Psychology, 49: 1169-1183.

Zanna, Mark P., George R. Goethals, and Janice F. Hill 1975 "Evaluating a sex-related ability: Social comparison with similar others and standard setters." Journal of Experimental Social Psychology, 11: 86-93.

By Kimberly D. Elsbach, Emory University; Roderick M. Kramer, Stanford University