Early child care and the employment potential of mothers: evidence from semi-parametric difference-in-differences estimation

Journal for Labour Market Research

Table 7 Results of heterogeneity analysis—effect variation.

	Employment	Agreed hours	Preferred hours	Mismatch (hours)	Full time	Part time
Heterogeneities
Education (reference: lower secondary school)
Middle secondary school	0.044	6.216	5.600	− 0.616	0.161	− 0.113
	(0.063)	(6.250)	(6.881)	(1.904)	(0.139)	(0.154)
High school	0.122*	10.457*	9.773	− 0.684	0.182	0.116
	(0.069)	(6.051)	(6.673)	(1.979)	(0.130)	(0.159)
Number of children	− 0.004	1.462	2.550	1.087	0.014	0.055
	(0.025)	(2.629)	(2.912)	(1.059)	(0.053)	(0.074)
Partner (reference: no partner living in household)
Partner living in household	0.052	2.641	− 0.454	− 3.095*	0.027	0.104
	(0.073)	(6.840)	(7.343)	(1.834)	(0.140)	(0.160)
N	11,640	3505	3505	3505	3505	3505

The results represent estimations of \(\gamma _0\) given in Eq. (3). Hence, they give the difference to the reference group for categorical variables or to a one-unit increase in case of continuous variables. Standard errors (in columns) are bootstrapped with 1000 replications considering clusters on the district level. The sample includes 18–45 years old mothers of up to 3-year-olds. Agreed and preferred hours are measured on the weekly basis. The control variable for estimating the propensity score are presented in Table 4
*\(p<0.1\), **\(p<0.05\), ***\(p<0.01\)