Skip to main content
. Author manuscript; available in PMC: 2020 Jun 1.
Published in final edited form as: J Res Adolesc. 2019 Jun;29(2):357–368. doi: 10.1111/jora.12409

Table 3.

Coefficients from Logistic Regression Models Predicting Separation of the Mother and Stepfather by Wave III

Model 1 Model 2 Model 3
Religious Attendance a
 Attends with both parents −.58 * (.27) −.58 * (.27) −.53 + (.29)
 Attends with one parent −.33 (.31) −.32 (.31) −.43 (.32)
Control Variables
Age −.09 (.06) −.11 (.06) −.09 (.06)
Female −.28 (.22) −.29 (.23) −.23 (.23)
Race b
 Black   .93 ** (.28)   .88 ** (.29)   .81 * (.30)
 Hispanic   .13 (.34)   .12 (.32)   .16 (.35)
 Other   .70 * (.32)   .69 * (.33)   .55 (.36)
Mother’s education c
 High school   .05 (.30)   .06 (.30)   .06 (.32)
 Some college −.67 (.39) −.66 (.39) −.68 (.41)
 Bachelor’s or higher   .25 (.42)   .25 (.41)   .13 (.41)
Logged family income   .02 (.14)   .01 (.14)   .03 (.15)
Years in stepfamily −.06 ** (.02) −.06 * (.02) −.09 ** (.02)
Total # siblings −.02 (.10) −.01 (.10) −.02 (.10)
Any half-siblings   .04 (.23)   .04 (.22)   .08 (.23)
Any stepsiblings −.28 (.39) −.28 (.39) −.31 (.40)
Shared activities   .02 (.03)   .02 (.04)
Family dinners −.04 (.04) −.03 (.04)
Family Relationships
Stepfather-adolescent closeness −.04 (.16)
Mother-adolescent closeness   .28 (.18)
Mother-stepfather relationship −.66 *** (.15)
Family belonging −.04 (.17)

Note: Note: N = 1,523. Missing data are imputed. Estimates are based on weighted data. Standard errors (in parentheses) are adjusted for weighting, clustering, and stratification.

a

Omitted reference group is adolescent attends with neither parent. Results from rotating the omitted religious attendance group revealed no significant differences between attends with one parent and attends with both parents.

b

Omitted reference group is White.

c

Omitted reference group is less than high school.

+

p = .07,

*

p <.05,

**

p < .01,

***

p < .001.