Public opinion tends to be stable. Once formed, attitudes are persistent and endure over time at both the individual and the aggregate levels. Attitudes toward marriage equality, however, have changed rapidly in recent years. This article posits that this is partly due to people learning that other members of their in-groups are supporters; they then alter their own opinions to be consistent with those of other in-group members. The authors tested this theory using a set of randomized survey experiments that shared identities as fans of professional football. When fans learnsometimes unexpectedly-that other fans or athletes are supporters of marriage equality, they are motivated to agree in order to further normalize their membership in those sports-fan groups. P ublic opinion tends to be stable. Once formed, attitudes are persistent and endure over time at both the individual and the aggregate levels. Attitudes toward marriage equality are a notable exception. In 1988, only 11% of the US public supported the legalization of same-sex marriage; by 1996, that support had increased to 27% and, by 200627% and, by , to 35%. By 2013 showed that a majority of Americans supported marriage equality, and the latest polls put national support near the 60% mark. However, identity groups have changed their opinions at diff erent rates (Public Religion 2014 ). To better understand these attitudinal shifts, this article explores how individual-level identities affect attitudes toward Lesbian, Gay, Bisexual, and Transgender (LGBT) rights.Several theories regarding public-opinion change may explain this rapid opinion shift. Allport's ( 1954 ) contact theory suggested that personal contact with LGBT individuals should prompt people to reconsider previously held prejudices. More Americans today are aware that they have personal contact with a member of the LGBT community (Public Religion 2014 ). Other theories point to cohort replacement, secularization, and a broader cultural shift (Banauch 2012 ).We argue that when individuals realize that a member of one of their in-groups communicates support for LGBT rights, their own support increases. This occurs regardless of an individual's sexual orientation and/or gender identity, even if that in-group is based on an unrelated identity (e.g., support for a professional sport or a sports team). In other words, a primed in-group identity (even if unrelated to being LGBT) should increase support for LGBT rights, provided that the shared identity is suffi ciently strong.Previous research generated three relevant models that outline the conditions under which individuals may undergo preference or attitude changes: Zaller's ( 1992 ) Receive-Accept-Sample model; Sherif, Sherif, and Nebergall's ( 1965 ) Social Judgment Theory;and Petty and Cacioppo's ( 1986 ) Elaboration Likelihood Model. These three models indicate that the fi rst step in successful attitude change is a willingness to receive, listen to, and process new information. By emphasizing a common identity, senders of comm...