Figure 1 | Scientific Reports

Figure 1

From: The impact of digital media on children’s intelligence while controlling for genetic differences in cognition and socioeconomic background

Figure 1

Path diagram of a strict measurement invariant Latent Change Score model with the change in intelligence from ages 9–10 to 11–12. Screen time Watching, Screen time Socializing, Screen time Gaming, cogPGS (Polygenic scores for cognitive performance), and SES (socioeconomic status) are exogenous variables, each already accounting for the effect of the others on baseline intelligence and on the change in intelligence after two years. All variables are standardized. (Which is why the loadings differ between time points—the constraining of loadings and intercepts must be done for the unstandardized estimates.) Non-significant values are marked with “n.s.”. Following convention, rectangles represent observed or exogenous variables and circles represent latent variables (screen time types are shown with only one rectangle for aesthetic reasons—they are actually three distinct rectangles). Single-headed arrows denote regression weights, while double-headed arrows represent variances, covariances, or errors. Standardized betas.

Back to article page