Table 2

Estimated relationship between mental health status and log earnings from ordinary least squares (OLS) regressions

Total problemsInternalising problemsAttention problemsConduct problems
Full sampleSibling sampleFull sampleSibling sampleFull sampleSibling sampleFull sampleSibling sample
School FESchool FESibling FESchool FESchool FESibling FESchool FESchool FESibling FESchool FESchool FESibling FE
Mental health problem scale (z-standardised)−0.067***−0.053*−0.046†−0.063***−0.032†−0.025−0.047***−0.044*−0.052*−0.036**−0.040†−0.007
(0.010)(0.021)(0.027)(0.011)(0.018)(0.027)(0.010)(0.017)(0.024)(0.012)(0.021)(0.027)
Female−0.321***−0.332***−0.354***−0.303***−0.325***−0.347***−0.328***−0.336***−0.358***−0.341***−0.347***−0.359***
(0.022)(0.034)(0.043)(0.023)(0.036)(0.044)(0.022)(0.034)−(0.043)(0.022)(0.035)(0.044)
Parental divorce−0.160***−0.158***−0.055−0.168***−0.167***−0.083−0.171***−0.165***−0.049−0.173***−0.166***−0.087
(0.022)(0.040)(0.157)(0.022)(0.041)(0.154)(0.022)(0.041)−(0.158)(0.022)(0.040)(0.154)
First born−0.032**0.0480.0670.0210.0480.0660.0170.0460.0660.0210.0480.064
(0.011)(0.041)(0.062)(0.017)(0.040)(0.063)(0.016)(0.041)−(0.062)(0.017)(0.041)(0.062)
Number of siblings0.016−0.046−0.033**−0.047−0.034**0.048−0.033**−0.045
(0.016)(0.030)(0.011)(0.030)(0.011)(0.030)(0.011)(0.030)
Immigrant background−0.352*−0.546***−0.342*−0.543***−0.357*−0.541***−0.351*−0.550***
(0.136)(0.154)(0.133)(0.155)(0.137)(0.154)(0.138)(0.153)
Parental education (ref. = below upper secondary) 
 Vocational upper secondary0.051*0.097**0.055*0.098**0.053*0.098**0.051*0.096**
(0.025)(0.034)(0.025)(0.035)(0.025)(0.035)(0.025)(0.034)
 Academic upper secondary0.072†−0.0120.070†−0.0160.073†−0.0120.075†−0.009
(0.040)(0.087)(0.040)(0.086)(0.040)(0.086)(0.040)(0.087)
 University, short0.122***0.144**0.124***0.144**0.121***0.142**0.122***0.144**
(0.026)(0.052)(0.026)(0.052)(0.026)(0.052)(0.026)(0.053)
 University, long0.195***0.203**0.203***0.209**0.192***0.199**0.197***0.205**
(0.046)(0.071)(0.046)(0.071)(0.046)(0.071)(0.046)(0.071)
R20.0740.0770.2320.0730.0750.2300.0710.0760.2330.0700.0760.230
  • SEs in parentheses. Mental health problem scale refers to the relevant indicator in each panel and is z-standardised (mean=0, SD=1). There are 7885 child observations in the full sample and 2637 child observations in the sibling sample. School fixed-effects models adjust for 63 school IDs. Sibling fixed-effects models adjust for 1281 family IDs. All models control for birth year dummies. Intercepts not reported.

  • †p<0.10; *p<0.05; **p<0.01; ***p<0.001 (two-tailed tests).

  • FE, fixed effects.