Genetic Approaches to Investigating the Contribution of Noncognitive Skills to Academic Achievement over Development

• In Event: Academic Achievement: What Matters? Insights from Longitudinal, Multi-Cohort, Genetically-Informative Investigations

Abstract

Twin studies have shown that academic achievement remains substantially heritable over compulsory education, even after accounting for individual differences in cognitive ability (Rimfeld, Malanchini et al., 2018) In addition, several other, partly heritable, factors (e.g. personality, motivation and socioemotional competencies) have been found to account for a portion of the heritability of academic achievement beyond cognitive skills (Krapohl, Rimfeld et al., 2014; Malanchini, Engelhardt, Grotzinger, Harden, & Tucker-drob, 2018). These characteristics have been broadly described in the literature as noncognitive skills (Heckman, Stixrud, & Urzua, 2006). The predictive ability of genome-wide polygenic scores (GPS) constructed from the latest genome-wide association (GWA) study of educational attainment (EA3; Lee et al., 2018) is also likely to reflect most of those heritable characteristics that propel individuals to different educational trajectories, not just cognitive ability.
The current study examined the relative contributions of educationally relevant cognitive and noncognitive characteristics to variation in academic achievement at several stages during compulsory education. In addition, the current project investigated the extent to which genetic and environmental factors are implicated in the development of these associations by triangulating multiple methods, including classical twin design and DNA-based methods. Adopting this multi-method approach, the current project provides a detailed account of how cognitive and noncognitive characteristics are linked to the development of individual differences in academic performance from age 9 to 16 in the Twins Development Study (TEDS). We applied multivariate twin design to investigate the extent to which cognitive and noncognitive skills contribute to variation in academic achievement at key points during compulsory education (ages 9, 12 and 16). Results of multivariate twin analyses showed that, after accounting for general cognitive ability, noncognitive skills significantly predict variation in achievement at all developmental stages. This prediction was found to be largely genetic in origin and to increase over development. However, sensitivity analyses showed that the etiology of the prediction from noncognitive skills to academic achievement at age 9 varied substantially based on raters (children, teachers, and parents).
In addition to twin design, we applied DNA-based methods –Genomic structural equation modelling (SEM, Grotzinger et al., 2019) and polygenic scoring– to investigate how the genetics of cognitive and noncognitive skills account for individual differences in academic achievement throughout the school years. Specifically, we obtained summary statistics for cognitive and noncognitive skills from recent work that has applied a GWAS-by-subtraction approach (Demange et al., 2020) leveraging GWA studies of educational attainment (EA; Lee et al., 2018) and cognitive performance (CP; Lee et al., 2018; Savage et al., 2018). Results of these genomic analyses showed that noncognitive genetics significantly accounted for variation in noncognitive skills at every developmental stage and, also in this instance, the prediction increased over development.
By examining their contribution to academic achievement at several developmental stages, the current research provides new crucial insights into the role of cognitive and noncognitive characteristics, starting from how each construct plays a critical role at a different stage during childhood and adolescence.

Authors

• Margherita Malanchini, Queen Mary University of London
  Presenting Author

• Andrea Giuseppe Allegrini, King's College London
  Non-Presenting Author

• Kaili Rimfeld, King’s College London
  Non-Presenting Author

• Rosa Cheesman, King's College London
  Non-Presenting Author

• Sophie von Stumm, University of York
  Non-Presenting Author

• Perline A. Demange, Vrije Universiteit Amsterdam
  Non-Presenting Author

• Michel Nivard, Vrije Universiteit Amsterdam
  Non-Presenting Author

• Andrew Grotzinger, University of Texas at Austin
  Non-Presenting Author

• Laurel Raffington, University of Texas at Austin
  Non-Presenting Author

• Javier de la Fuente, University of Texas at Austin
  Non-Presenting Author

• Paige K. Harden, University of Texas at Austin
  Non-Presenting Author

• Elliot M Tucker-Drob, University of Texas at Austin
  Non-Presenting Author

• Robert Plomin, King's College London
  Non-Presenting Author