Skip to main content

View full-text article in PMC

. Author manuscript; available in PMC: 2018 Jul 9.

Published in final edited form as: Soc Forces. 2013 Jan 12;91(3):1007–1034. doi: 10.1093/sf/sos192

Table A1.

Binary Logit Model Predicting College Completion

	Coefficient	Standard Error
Parental divorce	−0.349^*	0.177
Poor health at age 16	−0.305	0.290
Physical abuse^a
Rare	0.039	0.124
Often	−0.431^**	0.161
Emotional abuse^a
Rare	0.185	0.130
Often	0.140	0.151
Welfare	−0.738^*	0.302
Father’s education^b
High school	0.609^***	0.128
College	1.491^***	0.182
Missing on education	−0.069	0.211
Mother’s education^b
High school	0.636^***	0.128
College	1.275^***	0.204
Missing on education	−0.451	0.292
Father supervised	0.175	0.112
Father’s occupation^c
Upper-white-collar	0.544^***	0.136
Missing on occupation	0.414^*	0.183
Mother’s occupation^c
Upper-white-collar	0.312	0.189
Missing on occupation	−0.016	0.111
Age	0.010^*	0.004
Non-white	0.203	0.169
Female	−0.409^***	0.101
Log Likelihood	−1296.246
N	2,923

^*

p < .05;

^**

p < .01;

^***

p < .001

^a

Reference group = no abuse

^b

Reference group = less than high school

^c

Reference group = blue-collar/lower-white-collar