the results remain unchanged when including additional controls. We then also include additional childhood variables related to communication and behavior at age 10, such as the Rutter behavioral score, being sullen/sulky, and three interpersonal skills (communication, problem-solving, and teamwork). The results show that reading skills no longer significantly relate to adult life satisfaction, suggesting that part of the relationship between reading and life satisfaction is driven by interpersonal and behavior skills. For math, the results remain unaffected by these controls, albeit with a slight reduction in the magnitude of the estimated relationship. 2.4.2 Income and daily functioning as potential mediators In the appendix, Table A.13 explores income and daily functioning as potential mediators in the relationship between early skills and adult life satisfaction at age 46 (controlled for sex and intelligence). For reading, the appendix shows that income and daily functioning together (almost) fully mediate the baseline relationship between early reading skills and adult life satisfaction. For math, it shows that income and daily functioning together mediate around 64 percent of the baseline relationship between early math skills and adult life satisfaction, where the mediating role of daily functioning is stronger than the mediating role of income.6 Table A.13 in the appendix also shows that all daily functioning variables (at least) partly mediate the relationship when including them 6The coefficient for math in the last column of Table 2.2 is .087. When adding income and daily functioning in our model, this coefficient drops to .031. This means that the estimate has decreased by 64 percent. 29
RkJQdWJsaXNoZXIy MjY0ODMw