Figure 2.
(Panel A). Associations between latent factors of noncognitive skills and academic achievement at ages 7, 9, 12 and 16, after accounting for general cognitive ability at the same age using multiple regression. Each bar indicates the effect size of standardized regression coefficients and the error bars indicate the 95% confidence intervals around the estimates. The left panel shows the associations for latent measures of education-specific noncognitive skills (NCS), while the right panel the associations for latent dimensions of domain-general self-regulation skills. The figure is further divided into self-rated (top panel), parent-rated (middle panel) and teacher-rated (bottom panel) measures. (Panel B). Each bar represents genetic effects on academic achievement over development and includes three shadings. The lighter (yellow) shadings indicate the proportion of genetic variance in academic achievement that can be attributed to genetic variance in cognitive skills (Cog). The orange shadings indicate the proportion of genetic variance in academic achievement that can be attributed to genetic variance in noncognitive skills, independent of the genetics of cognitive skills (Noncog – Cog). The red shadings indicate genetic effects on academic achievement independent of the genetics of cognitive and noncognitive skills (Achievement specific). Results are further divided into self-rated (top panel), parent-rated (middle panel) and teacher-rated (bottom panel) measures. 95% Confidence intervals for all estimates are presented in Supplementary Tables 12 and 13.