Schools should provide the opportunity for a support

This preview shows page 7 - 8 out of 10 pages.

LimitationsThis study has several limitations. Although we used the bestsampling strategies available to reach stigmatized populations(Diamond, 2003), the results cannot establish causality and cannotbe generalized to all gender-nonconforming youth in other settingsoutside California. The data collection was retrospective, whichdoes not allow for measurements to be taken at unique data points(Frazier, Tix, & Barron, 2004). The order of measurements in thesurvey may have led to measurement bias because participantswere asked to report retrospectively on prevalence of LGBTschool-related victimization prior to being asked about their cur-rent life situations. This order of questions may have promptedrespondents to report more negative psychosocial adjustment. Ourmethods attempt to establish temporal order by asking participantsto report retrospectively on gender nonconformity and victimiza-tion while reporting current life adjustment. Although this is apotential concern, prior research has found that results of retro-spective reports of school bullying are stable over time, a findingthat gives us confidence that reports of adolescent school victim-ization were not overly influenced by young adult mental health(Rivers, 2001b). Another limitation of our construct of LGBTschool victimization and our test of the minority stress model isthat we do not have a measure of expectations of victimization;those who expect more victimization may report more victimiza-tion experiences.Our focus on school victimization as the sole context for ourmeasure of LGBT-related victimization and violence is limited. Amore comprehensive approach to studying the mechanisms thatplace LGBT and gender-nonconforming youth at greater risk forconcurrent and later psychosocial maladjustment would includeexperiences of victimization and rejection from multiple contexts(e.g., family, community, work). Our measurement of gendernonconformity is also limited in that it was assessed only with asingle item. Future work could examine the associations amonggender nonconformity, victimization experiences, and adjustment2For example, the Safe Schools Improvement Act (H.R. 2262), cur-rently under consideration by Congress, is the first proposed federal schoolantibullying law that includes enumerated categories. Currently 10 U.S.states have enumerated school antibullying laws designed to protect stu-dents based on sexual orientation and gender identity or expression.Table 3Tests of Equivalence of Covariance, Variance, Latent Correlations, and MeansModel2dfP2dfpConstraint tenableHomogeneity of variances and covariances134.67114.0514.3015.05YesEquality of variances127.59104.057.225.05YesEquality of correlations128.96109.058.5910.05YesEquality of means166.83116.018.7710.05Yes1586TOOMEY, RYAN, DIAZ, CARD, AND RUSSELL

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture