Table 4

Lags and leads in equality of opportunity and mortality

Attainment of bottom quartileAttainment of top quartileCorrelation of parental and children’s attainment
β (CI 95%)R2β (CI 95%)R2β (CI 95%)R2
Lags of equality of opportunity
 ContemporaneousM1: −0.23* (−0.36 to 0.10)M1: 0.28* (0.11 to 0.46)M1: 0.13* (0.05 to 0.20)
 Five years youngerM1: −0.21* (−0.37 to 0.05)M1: 0.25* (0.08 to 0.43)M1: 0.12* (0.05 to 0.20)
 Ten years youngerM1: −0.28* (−0.45 to 0.12)M1: 0.10 (−0.07 to 0.27)M1: 0.12* (0.04 to 0.20)
 Total effectM1: −0.72* (−0.98 to 0.46)0.16M1: 0.64* (0.35 to 0.93)0.154M1: 0.37* (0.23 to 0.50)0.16
Leads of equality of opportunity
 ContemporaneousM2: −0.20* (−0.34 to 0.06)M2: 0.15 (−0.03 to 0.33)M2: 0.07 (−0.02 to 0.157)
 Five years olderM2: −0.06 (−0.22 to 0.10)M2: 0.22* (0.03 to 0.41)M2: 0.08 (−0.02 to 0.175)
 Ten years olderM2: −0.09 (−0.23 to 0.06)M2: 0.23* (0.06 to 0.40)M2: 0.10* (0.01 to 0.181)
 Total effectM2: −0.35* (−0.61 to 0.09)0.09M2: 0.60* (0.28 to 0.92)0.108M2: 0.24* (0.09 to 0.401)0.10
  • Total number of observations is 1000 (N is lower than in previous tables due to inclusion of lags). CIs are computed using heteroskedasticity-robust SEs corrected for clustering at the age group-year level. Regressions are population weighted. Regressors from ESS are additionally design-weighted. All models include year, age group and country fixed effects and all controls described in Methods section and in table 1. Reported R2 are 'within' R2. Significant associations (uncorrected, at 5% level) are shown in bold.

  • Analyses were conducted using the REGHDFE Stata package42 .

  • *Significant at table-wide false discovery rate of 5% (using Benjamini-Hochberg method).

  • ESS, European Social Survey.