This dissertation consists of four journal length articles that examine manifestations and correlates of institutionalized heterosexism in high school policies, programs, and social environments as perceived by lesbian, gay, bisexual, and questioning (LGBQ) students. One article presents a theoretical framework for operationalizing and assessing heterosexism in schools and three articles report on secondary analyses of Internet survey data that were collected from 987 LGBQ high school students in 1997 and from 2076 LGBQ students in 2000.

The first study examines the impact of recent changes on student experiences of heterosexism in high schools. Student perceptions of policies, programs, and anti-LGBQ harassment in school were described. There were significant increases between 1997 and 2000 in both policies and programs. For a subset of 230 students nested within 32 districts, student perceptions of programs, but not policies or perceived harassment, were significantly related to district-level reports of policies and programs.

The second study examines relationships between perceived heterosexism in high school policies and programs, the prevalence and tolerance of anti-LGBQ harassment, and victimization rates among LGBQ students. There were moderate correlations between perceived policies, programs, and harassment. Perceived programs were more closely associated with anti-LGBQ harassment in schools than were policies. Victimization was related to perceived programs and harassment. Perceived harassment partially mediated effects of programs on victimization, but programs predicted victimization even after controlling for perceived harassment. Moderating effects of demographic characteristics (i.e., gender, race, outness, and access to LGBQ youth groups or community centers) were explored.

The third study utilized a cluster analysis to discover distinct constellations of perceived policies, programs, and harassment. Several youths endorsed a pattern of responses reflecting consistently Hostile or Welcoming climates across the policy, program, and harassment domains. The most common profile reflected Invisibility (few inclusive policies or programs and little harassment). Cluster membership was associated with personal victimization experiences, self-esteem, and suicidality in predictable ways in both cohorts. In each article, implications for research, theory, and educational policy are discussed.