Skip to main content
. Author manuscript; available in PMC: 2018 Nov 20.
Published in final edited form as: Demogr Res. 2017 Mar 21;36:863–892. doi: 10.4054/DemRes.2017.36.30

Table 6:

Multinomial logistic regression predicting group membership showing log odds

Model 1 Model 2
Recovery vs Robust Early onset vs Robust Late onset vs Robust Recovery vs Robust Early onset vs Robust Late onset vs Robust
Demographic characteristics
 Married at Wave 5 −0.402*** −0.547* −0.289** −0.337** −0.159 −0.285*
 Nuclear household −0.110* −0.215 −0.347** −0.123 −0.529* −0.322**
Childbirth characteristics
 Number live births 0.037* 0.114* 0.046 0.023 0.113* 0.046
 Successful births ratio −0.240 −2.108** −0.448 0.268 −2.009* −0.351
 Age at birth of first child 0.001 0.045 0.073** −0.015 0.044 0.070***
Socioeconomic status characteristics
 Education at Wave 1
 Primary --- --- --- --- --- ---
 High school −0.117 −0.213 −0.196 0.093 −0.168 −0.202
 College −0.008 −1.466*** −0.748 −0.084 −1.509*** −0.783***
 Wealth score at baseline 0.430*** 0.538*** 0.250*** 0.397*** 0.452*** 0.222***
 Urbanicity score at baseline 0.042*** 0.052*** 0.032*** 0.043*** 0.050*** 0.031***
Health characteristics
 BMI 0.581*** 1.059*** 0.419***
 Chronic condition 1.905*** 1.116*** 0.639***
 Excellent self-rated health −0.517*** −0.126 −0.151
Constant −2.714 −3.845 −3.410 −3.756 −4.660 −3.687
Log-likelihood -1635.3 -1511.1
Δ −2 × Log-likehood1 193.0*** 248.4***
*

.05 < p < .10

**

.01 < p < .05

***

p < .01

1

model 1 compared to intercept-only, model 2 compared to model 1.