Skip to main content
. Author manuscript; available in PMC: 2022 Jun 4.
Published in final edited form as: J Adolesc Health. 2020 Oct 17;68(6):1155–1161. doi: 10.1016/j.jadohealth.2020.08.034

Table 3. The multinomial regression model of dating violence classes predicted by sexual orientation, gender identity, and gender nonconformity (N = 71,843).

High DV victimization versus no/low DV DV victimization and perpetration versus no/low DV Verbal victimization and perpetration versus no/low DV Moderate DV victimization and perpetration versus no/low DV
Estimate (SE) RRR   p Estimate (SE) RRR   p Estimate (SE) RRR    p Estimate (SE) RRR    p
Sexual orientation (ref: heterosexual)
   Gay or lesbian   –.95 (.16) 2.59 < .001 –.94 (.31) 2.56 .002 –.24 (.29) 1.27   .410  –.41 (.32) 1.51   .194
   Bisexual –1.36 (.07) 3.90 < .001 –.65 (.24) 1.92 .007 –.43 (.15) 1.54   .003  –.63 (.18) 1.87   .001
   Questioning   –.42 (.11) 1.53 < .001 –.39 (.27) 1.47 .156   .09 (.17)   .92   .613  –.21 (.22) 1.23   .338
   Transgender (ref: nontransgender)   –.00 (.10) 1.00    .983 –.73 (.25) 2.08 .004 –.01 (.23) 1.01   .982  –.22 (.24) 1.25   .362
   Gender nonconformity – .26 (.03) 1.30 < .001 –.21 (.07) 1.23 .003 –.08 (.04) 1.08   .054 – .18 (.05) 1.20 < .001
   Biological sex (ref: female)  1.05 (.06)   .35 < .001 –.07 (.12) 1.07 .567   .79 (.07)  .46 < .001    .19 (.09)   .83   .031

Controlling for age and race/ethnicity. Unstandardized estimates and relative risk ratios (RRR). Statistically significant estimates (p < .05) are shown in bold.

DV = dating violence.