The objective of the current study was to examine sleep and academic functioning during middle childhood. More specifically, the twin design was used to determine the heritability of academic competence and sleep. Phenotypic analyses using multi-level mixed model regressions were performed to predict academic functioning from sleep. Lastly, socioeconomic status was tested as a moderator in the associations between sleep and academic functioning. Participants included twins (N = 191 families; Mage = 8.47 years) recruited from Arizona birth records at 12 months of age. Sleep duration, latency, onset, efficiency, variability, and sleep problems were assessed using actigraph watches and the Child Sleep Habits Questionnaire. Academic functioning was assessed using subtests of the Woodcock Johnson Cognitive Test of Achievement-IV (picture vocabulary, passage comprehension, and applied problems) and the MacArthur Health and Behavior Questionnaire. As determined by twin intraclass correlations, the heritability of academic competence ranged from 51-76%. Sleep heritability ranged from 14-80%. In addition, phenotypic analyses only showed a significant association between sleep latency and WJ picture vocabulary scores. More specifically, sleep latency was negatively associated with the picture vocabulary subtest. Additional models were run to examine if any interactive effects were present between early SES and the various sleep parameters. Several significant associations were observed with applied problems scores and parent-reported academic competence. Specifically, for children of low SES, a significant positive association was observed for sleep duration and WJ applied problems scores, as well as for sleep efficiency and WJ applied problems scores. No significant associations were observed for sleep efficiency and HBQ scores with children of any SES. Also, no significant relationships were observed with children of high SES for any of the academic measures.