Skip to main content
. 2023 Aug 30;25(5):743–762. doi: 10.1007/s10198-023-01620-8

Table 9.

Effect of WfH - Heterogeneity analysis

Gender Parenthood Age Commuting
Male Female Childless Parent Age < 35 35 <= Age
< 50
Age >= 50 Commuter No Comm.
Very good health:
 WfH (dummy:1=yes) 0.373** 0.153 0.175 0.601* – 0.908 0.694** 0.155 0.214 0.568
(0.159) (0.325) (0.151) (0.327) (1.161) (0.276) (0.166) (0.157) (1.444)
 Observations 5755 2144 6027 1872 1006 2763 4130 5445 1452
 Effective F-statistic 37.85 5.761 35.08 8.987 0.875 13.21 24.51 32.32 0.477
Good health:
 WfH (dummy:1=yes) 0.104 0.0607 0.125 -0.555 0.784 -0.204 0.0119 -0.0443 1.774
(0.201) (0.498) (0.213) (0.372) (1.232) (0.304) (0.247) (0.209) (2.377)
 Observations 5755 2144 6027 1872 1006 2763 4130 5445 1452
 Effective F-statistic 37.85 5.761 35.08 8.987 0.875 13.21 24.51 32.32 0.477
Satisfactory health:
 WfH (dummy:1=yes) -0.388** − 0.294 -0.199 0.0591 0.419 -0.406 -0.243 -0.144 – 1.616
(0.191) (0.478) (0.200) (0.295) (0.877) (0.272) (0.244) (0.190) (2.303)
 Observations 5755 2144 6027 1872 1006 2763 4130 5445 1452
 Effective F-statistic 37.85 5.761 35.08 8.987 0.875 13.21 24.51 32.32 0.477
Poor health:
 WfH (dummy:1=yes) -0.0247 0.115 -0.0464 0.0708 – 0.142 -0.0243 0.167 0.0421 0.0271
(0.101) (0.309) (0.119) (0.157) (0.458) (0.142) (0.151) (0.112) (0.819)
 Observations 5755 2144 6027 1872 1006 2763 4130 5445 1452
 Effective F-statistic 37.85 5.761 35.08 8.987 0.875 13.21 24.51 32.32 0.477
Bad health:
 WfH (dummy:1=yes) – 0.0637 – 0.0353 – 0.0548 – 0.176* – 0.154 – 0.0598 – 0.0908 – 0.0680 – 0.753
(0.0558) (0.152) (0.0610) (0.106) (0.167) (0.0855) (0.0775) (0.0586) (0.932)
 Observations 5755 2144 6027 1872 1006 2763 4130 5445 1452
 Effective F-statistic 37.85 5.761 35.08 8.987 0.875 13.21 24.51 32.32 0.477
WHO-5 Well-Being Index:
 WfH (dummy:1=yes) 0.672* – 0.102 0.524 0.710 – 1.177 1.316** 0.259 0.530 3.204
(0.389) (0.974) (0.422) (0.688) (2.085) (0.658) (0.492) (0.415) (4.563)
 Observations 5718 2132 5982 1868 1003 2754 4093 5412 1442
 Effective F-statistic 37.58 6.451 35.92 8.451 0.773 13.22 25.00 32.61 0.476

This table shows the estimated coefficients of 2SLS second stage regressions of Eq. 1 by subgroups for all dependent variables but No. of sick days. We refrain from testing heterogeneity for this outcome given the relatively small sample size and the non-significance of the effect of interest, cf. Table 8. Control variables are included as indicated in the respective equations. Standard errors reported in parentheses are clustered at the individual level for all regressions. The Effective F-statistic refers to Olea and Pflueger [50]

* p<0.10, ** p<0.05, *** p<0.01