This paper experimentally examines the effects of passing (versus revealing) a contextually devalued identity on performance-related self-confidence. An experimental scenario was developed on the basis of the results of a pilot study. Studies 1 and 2 (total N ¼ 255) experimentally manipulate passing versus revealing a contextually devalued identity, to an ingroup or an outgroup partner. The results show that, although passing makes participants believe that their partner has more positive expectations of them, it also undermines performance-related self-confidence. Moreover, the results show that negative self-directed affect (i.e., guilt and shame) mediated the negative effect of passing on performancerelated self-confidence. Copyright # 2006 John Wiley & Sons, Ltd.Members of devalued or stigmatized groups are frequently exposed to negative stereotypes and expectations (e.g., Crocker, Major, & Steele, 1998). These negative stereotypes and expectations can lead to self-fulfilling prophecies, resulting in low self-confidence (Biernat, Crandall, Young, Kobrynowicz, & Halpin, 1998;Cadinu, Maass, Frigerio, Impagliazzo, & Latinotti, 2003;Kray, Thompson, & Galinsky, 2001;Lord & Saenz, 1985;Schmader & Johns, 2003;Stangor, Carr, & Kiang, 1998;Steele & Aronson, 1995). Members of devalued groups can choose to break this self-fulfilling cycle by hiding their devalued group membership and passing as members of a more valued group (Goffman, 1963;Jones et al., 1984;Katz, 1981;Tajfel, 1981). Although this strategy should lead others to associate the self with more positive expectations, it is as yet unclear whether or not it successfully raises self-confidence. In this paper we empirically demonstrate that passing as a member of a more valued group is not an effective way to raise one's self-confidence, and provide evidence for the psychological process responsible for this effect. In this way, we demonstrate that self-confidence among members of devalued groups does not only depend on the external expectations of the self people perceive others to have, as it may be negatively affected by the very strategies aimed at improving those expectations. PASSING AS AN IDENTITY MANAGEMENT STRATEGYPast research on the effects of the negative stereotypes and expectations held about members of specific groups mainly focused on the effects of visible group memberships (see also Quinn, 2004). However, people often have a choice of whether or not to reveal their (devalued) group membership to others. This choice is clearly present when group memberships are not visible (as sexual orientation, certain mental and physical illnesses, and devalued professions; Holmes & River, 1998;Lee & Craft, 2002), but even identities which are usually visible may in some contexts be hidden, allowing for selfpresentation on the basis of a more positive identity (e.g., when interacting through the internet; Jaffe, Lee, Huang, & Oshagan, 1999). Passing as a member of a more positively evaluated group is thus an identity management strategy available to ...