Daily Sexual Behavior, Sexual Esteem, and Body Image in Transgender and Cisgender Individuals

Figures & data

Table 1. Participant characteristics and group differences on education, occupation, housing, and number of children.

		Transgender group	Cisgender group	Group difference
Education	No high school	1 (2%)	0 (0%)	X^2(1, N = 99) = 2.12, p = .347, Cramer’s V = .146
	High school degree	21 (45%)	29 (56%)
	College or university degree	25 (53%)	23 (44%)
Occupation	Student	18 (38%)	13 (25%)	X^2(1, N = 99) = 2.09, p = .352, Cramer’s V = .145
	Employed	17 (36%)	24 (46%)
	Unemployed	12 (26%)	15 (29%)
Housing	Alone	16 (34%)	19 (37%)	X^2(1, N = 99) = 5.07, p = .166, Cramer’s V = .226
	Parents/family	13 (28%)	20 (38%)
	Partner	12 (26%)	12 (23%)
	Friends/student housing	6 (13%)	1 (2%)
Number of children	0	38 (81%)	39 (75%)	X^2(1, N = 99) = 1.77, p = .779, Cramer’s V = .134
	1	3 (6%)	3 (6%)
	2	2 (4%)	6 (12%)
	3	3 (6%)	3 (6%)
	4	1 (2%)	1 (2%)

Table 2. Descriptive statistics on study variables.

	Transgender group M (SD), range, skewness, kurtosis	Cisgender group M (SD), range, skewness, kurtosis
Sexual esteem	4.25 (1.17), 1.28–6.38,−.20, −.01	4.83 (1.11), 1.84–7,-.03, −.45
Body image	4.68 (1.36), 1.03–6.75,−.71, .26	5.38 (1.30), 1.46–7.00,-.88, .54
Sexual openness	3.15 (1.34), 1.00–6.19,.36, −.17	3.70 (1.57), 1.00–6.90,.19, −.76
	%	%
Sexual activity	10	16
Intimate activity	30	34
Masturbation	38	31

M = mean, SD = standard deviation. Percentages present the percentage of days on which sexual behavior was reported.

Table 3. Multilevel coefficients (β) and standard errors (SE) representing the effects of sexual behavior and sexual openness on sexual esteem and body image on a within- (i.e. daily) and between-subject level.

	Sexual esteem β (SE)		Body image β (SE)
Predictor	Main effects
Cisgender versus transgender identity	1.21 (.45) **		1.64 (.70) *
Relationship versus single	.42 (.23)		−.42 (.37)
	Within	Between	Within	Between
Sexual activity	.03 (.10)	−1.04 (.96)	.03 (.10)	−1.38 (1.46)
Intimacy	.21 (.08)*	.89 (.49)	.07 (.09)	1.64 (.75)
Masturbation	−.07 (.06)	.21 (.51)	.10 (.07)	.28 (.78)
Sexual openness	.12 (.02) ***	.68 (.12) ***	.06 (.02) **	.51 (.18) **
Cis x Sexual activity	.16 (.13)	1.52 (1.43)	−.12 (.15)	1.50 (2.17)
Cis x Intimacy	−.16 (.12)	−1.00 (.67)	.08 (.13)	−.94 (1.03)
Cis x Masturbation	−.01 (.09)	−.59 (.71)	−.21 (.09) *	−.82 (1.10)
Cis x Sexual openness	−.01 (.03)	−.20 (.15)	−.01 (.03)	−.20 (.24)

*** = p < .001; ** = p < .01; * = p < .0.

Figure 1. The daily effect of having masturbated or not (on the x-axis; values represent subject-centered values) on body image (on the y-axis) differed significantly between transgender (left panel) and cisgender (right panel) participants (β = −0.21, SE = 0.09, t(1508) = −2.23, p = .026). Shadings indicate 95% confidence intervals. Note that while the figure suggests there is a positive effect in transgender individuals and a negative effect in cisgender individuals, none of these effects in themselves were significant, and only the interaction reached a significance of p < .05.

Table 4. Multilevel coefficients (β) and standard errors (SE) representing the effects of sexual esteem and body image on sexual behavior and sexual openness on a within – (i.e. daily) and between-subject level.

	Sexual activity β (SE)		Intimacy β (SE)		Masturbation β (SE)		Sexual openness β (SE)
Predictor	Main effects
Cisgender versus transgender identity	−1.12 (1.68)		−.30 (1.79)		2.56 (1.59)		−.48 (.96)
Relationship versus single	1.69 (.41) ***		4.10 (.46) ***		−1.10 (.37) **		−.58 (.23) **
	Within	Between	Within	Between	Within	Between	Within	Between
Sexual esteem	.37 (.21)	.10 (.30)	.75 (.19) ***	−.27 (.32)	.42 (.16) **	.99 (.28) ***	.55 (.08) ***	.90 (.17) ***
Body image	.20 (.20)	.24 (.27)	.05 (.17)	.66 (.27) *	.03 (.14)	.06 (.24)	−.01 (.07)	−.07 (.15)
Cis x Sexual esteem	.63 (.30) *	.34 (.45)	.17 (.27)	.45 (.47)	−.15 (.22)	−.24 (.40)	−.04 (.11)	.27 (.25)
Cis x Body image	.12 (.30)	−.12 (.40)	.28 (.25)	−.50 (.42)	−.03 (.22)	−.41 (.36)	.19 (.11)	−.12 (.22)

*** = p < .001; ** = p < .01; * = p < .05.

Note: Since multilevel logistic regression models are used to predict sexual activity, intimacy, and masturbation, the estimated regression coefficients reported for these variables should be interpreted on the log odds ratio scale. A positive (negative) coefficient corresponds to a higher (lower) probability for the event of interest.

Figure 2. The daily effect between sexual esteem (on the x-axis; values represent subject-centered values) and the chance of engaging in sexual activity (on the y-axis; values represented on a logit scale) differed significantly between transgender (left panel) and cisgender (right panel) participants (β = 0.63, SE = 0.30, z = 2.11, p = .035). Shadings indicate 95% confidence intervals. Note that the positive effect of daily sexual esteem on the chances of engaging in sexual activity was significant in cisgender participants only.