A meta-analysis of the relationship between socioeconomic status and executive function performance among children

Abstract

The relationship between childhood socioeconomic status (SES) and executive function (EF) has recently attracted attention within psychology, following reports of substantial SES disparities in children's EF. Adding to the importance of this relationship, EF has been proposed as a mediator of socioeconomic disparities in lifelong achievement and health. However, evidence about the relationship between childhood SES and EF is mixed, and there has been no systematic attempt to evaluate this relationship across studies. This meta-analysis systematically reviewed the literature for studies in which samples of children varying in SES were evaluated on EF, including studies with and without primary hypotheses about SES. The analysis included 8760 children between the ages of 2 and 18 gathered from 25 independent samples. Analyses showed a small but statistically significant correlation between SES and EF across all studies (r_random = .16, 95% CI [.12, .21]) without correcting for attenuation owing to range restriction or measurement unreliability. Substantial heterogeneity was observed among studies, and a number of factors, including the amount of SES variability in the sample and the number of EF measures used, emerged as moderators. Using only the 15 studies with meaningful SES variability in the sample, the average correlation between SES and EF was small-to-medium in size (r_random = .22, 95% CI [.17, .27]). Using only the six studies with multiple measures of EF, the relationship was medium in size (r_random = .28, 95% CI [.18, .37]). In sum, this meta-analysis supports the presence of SES disparities in EF and suggests that they are between small and medium in size, depending on the methods used to measure them.

Figure 1

Flow chart illustrating the identification of included studies.

Figure 2

Forest Plot of all studies included in the meta-analysis.

Figure 3

Scatterplot of the relationship between number of EF measures and effect size.

Figure 4

Scatterplot of the relationship between number of SES measures and effect size.

Figure 5

Funnel Plot of Standard Error by Fisher’s Z for all studies included in the meta-analysis.