Andreas Lubitz, a 27-year-old German pilot employed by Germanwings, a subsidiary of the airline Lufthansa, struggled with severe depression—a condition that he felt he could not share with his co-workers (Eddy, Bilefsky, and Clark 2015). His depression was bad enough that he elected to take medical leave during his aviation training before returning to complete his pilot’s license. Presumably he hoped his condition would improve, but it didn’t. On March 24, 2015, he killed himself and 149 other people by intentionally crashing an Airbus A320 into the French Alps.
The consequences of withholding personal information in the workplace are rarely so dramatic. Yet the dilemma that Lubitz faced was familiar to many. We must often contend with difficult personal matters in the course of carrying out our work. This creates a dilemma. On one hand, keeping such information to ourselves is unpleasant and cognitively depleting (Lane and Wegner 1995; Critcher and Ferguson 2014) while disclosing it to another person can be emotionally cathartic (Pennebaker 1997). On the other hand, we are compelled by a psychological need to be liked by others (James 1890; Srivastava and Beer 2005) and cognizant of the fact that the negative evaluations of coworkers have a direct bearing on our employment prospects (Schneider 1986).
Accordingly, insofar as certain information marks one as substantially less desirable to others (Goffman 1963: 3), its disclosure in the workplace threatens both our self-esteem and our remunerative potential. Following Goffman’s distinction between the discredited and the discreditable—people with undesirable traits that are visible versus those that are concealable—I refer to such information as discrediting (Goffman 1963: 4). The extent to which information can actually discredit a person exhibits numerous contingencies, varying substantially based on aspects of the focal individual and the social environment in which he or she is embedded (Quinn and Chaudoir 2009). The key point for the purposes of this research, however, is that the potential discloser considers the information discrediting and guards it carefully as a result.
The tension between the compulsion to disclose and the rationale for withholding discrediting information receives substantial attention from scholars (see Pachankis 2007 for an excellent review). However, at present our understanding of this tension remains circumscribed due to a heavy emphasis on the role of the individual—a consequence of the fact that the study of disclosure has been, in recent years, primarily the domain of psychologists. As a result, we know little about the way in which the social structure of relationships that surround the dyad of discloser and confidant influences the probability of disclosure, as well as the consequences therein. Yet this element is critical to building an accurate theory of disclosure, since a person’s reputation—which he risks when revealing discrediting information—arises from the exchange of information between people in his social network who are themselves connected (Coleman 1988). Accordingly, to understand disclosure one must also understand the social structure in which a discloser operates.
I investigate the influence of workplace social networks on the disclosure of discrediting information with a multi-method empirical approach that includes the analysis of survey data, a laboratory experiment, and qualitative interviews. The use of multiple methods strengthens the empirical insights that I am able to derive (Fine and Elsbach 2000; Small 2011). In Chapter 1, I develop a theory of disclosure that addresses a conceptual puzzle. Disclosure occurs when a discloser trusts that a confidant will not share his discrediting information with any third parties. Trust tends to inhere in strong ties, or ties between two people who are emotionally close (Marsden and Campbell 1984). Yet strong ties are often transitive in that two people who are close friends are likely to have other close friends in common (Simmel 1908; Heider 1958; Kossinets and Watts 2006). Thus, while trust at the dyadic level can be expected to promote disclosure, its structural concomitant—the density of relations around a strongly-connected dyad—should exert the opposite effect, insofar as a confidant who shares many third-party ties with a discloser is ideally positioned to circulate the discloser’s discrediting information should he or she wish to (Davis 1973, Granovetter 1985: 492).
I test this idea in Chapter 2 with two empirical studies. In the first, an analysis of a social network survey that I administered to the employees of a consulting firm, I show that the presence of a dense social structure around a dyad discourages disclosure over and above the positive effect of tie strength. Because these data are cross-sectional, however, I also design and implement a laboratory experiment in order to test whether social network density itself, as opposed to an omitted variable that is associated with it, is responsible for inhibiting disclosure. By manipulating subjects’ perceptions of the social structure surrounding them and a confederate I confirm experimentally that density itself reduces the probability of disclosure.
In Chapter 3 I summarize my findings and describe opportunities for future research. My findings contribute to social capital theory by showing that a dense social structure does not universally enhance knowledge sharing (e.g., Kadushin 2012). Instead, when the information in question renders its object vulnerable, network density serves as a source of interpersonal “friction” that can impede knowledge exchange (Ghosh and Rosenkopf 2014). Additionally, I advance scholars’ theoretical understanding of disclosure by expressly incorporating the effect of social structure beyond the dyad. This approach, which stands in contrast to social psychological treatments that focus exclusively on discloser and confidant (e.g., Pennebaker 1990), increases theoretical precision with respect to the probability of disclosure occurring. With that said, it represents the very beginnings of a comprehensive theory of disclosure that has yet to develop.
Accordingly, drawing from the disclosure literature and 30 in-depth qualitative interviews that I conducted during the Spring of 2016, I discuss three different research projects that, when pursued, will significantly advance scholarly understanding of disclosure’s antecedents and consequences. I first discuss the fact that most investigations of disclosure are agnostic to the egregiousness of the discrediting information in question (e.g., Kelly 2002). This element is essential to specify, however, since information that is more egregiously discrediting is less likely to be disclosed at baseline, irrespective of individual or structural factors. Next, I argue that the opportunity for emotional catharsis, which extant literature suggests is the primary factor that motivates disclosure, is insufficient to explain the full range of disclosure in the workplace. Instead, I suggest that people may disclose for two additional and strategic reasons: As a form of anticipatory impression management and to increase intimacy with another. This represents an important theoretical step insofar as it more precisely identifies the factors that motivate disclosure. Finally, I address the perspective of the confidant, and explore the antecedents of “betrayal,” that is, of a confidant sharing a discloser’s discrediting information with third parties. This approach has the potential to shed light on the way in which discrediting information travels through a social network in spite of incentives that should inhibit its transfer (e.g., Kurland and Pelled 2000). In each case, I derive testable hypotheses that constitute fruitful avenues for future research.
