Table 2.

Associations between bi+ visibility strategies and dimensions of identity, minority stress, and health

	Separate regression analyses	Simultaneous regression analyses
	Bi+ visibility strategies (aggregated across types)	Gender-based visual displays	Community engagement	Direct communication	Indirect communication	Public behavioral displays
	β	β	β	β	β	β
Discrimination (heterosexual)	.39**	.17**	.09	.08	.11*	.13*
Discrimination (gay/lesbian)	.29**	.08	.03	.16**	.04	.08
Internalized bi-illegitimacy	−.09*	−.01	.07	−.13*	−.13*	.11*
Anticipated binegativity	−.06	.22**	−.03	−.34**	.12*	−.02
Internalized binegativity	−.19**	.05	.002	−.25**	−.06	.04
Identity affirmation	.50**	.09	.16**	.38**	−.01	.02
Identity centrality	.48**	.22**	.15**	.31**	.01	−.05
Anxiety	.15**	.07	−.13*	−.05	.26**	.07
Depression	.11*	.04	−.11	.002	.18**	.04
Alcohol use problems (Rate Ratio)	1.08	.49	1.06	.99	.91	1.18**
Marijuana use problems (Rate Ratio)	1.10	1.10	.93	1.07	1.04	.98
Cigarette use (Odds Ratio)	1.22	.93	1.17	.94	.89	1.34*

In the simultaneous regression analyses, the five types of bi+ visibility strategies were entered as simultaneous predictors. In the separate regression analyses, the only predictor was bi+ visibility strategies (aggregated across type). All of the analyses controlled for age, sex/gender, sexual identity, and race/ethnicity. All of the coefficients are betas with the exceptions of the coefficients for alcohol and marijuana use problems (which are rate ratios) and cigarette use (which are odds ratios).

^*p < .05;

^**p < .001.