Timesplitters: playing video games before (but not after) school on weekdays is associated with poorer adolescent academic performance. A test of competing theoretical accounts

Video games are a common pastime for adolescents. There has been a relatively enduringconcern that time spent playing video games might undermine students’ academicperformance. Hartanto, Toh, and Yang (2018) suggested that frequent gameplay,particularly frequent weekday video gameplay, might displace students’ homework;reducing academic performance, albeit by a small amount. Although some evidence hasbeen presented supporting this view, the emerging evidence is mixed. Significant theoreticalflaws have also limited our understanding of the relationship between video game play andadolescent academic performance. Here we show that, across approximately 219,000students, the frequency of video gameplay does not appear to have a systematicrelationship with academic performance, confirming the results of earlier research(Drummond & Sauer, 2014). Moreover, although there is a small-moderate reduction inacademic performance for some weekday players, this reduction only occurs for playerswho play in the mornings before school. Players who play in the evenings after school showno meaningful difference in academic performance to non-users. As no existing theoreticalaccounts of the relationship between gameplay and academic performance adequatelyexplain this finding, we propose that the results most likely support a third variableexplanation. That is, video game play does not appear to affect academic results per se. Theresults further suggest that media psychologists and educational researchers analysing largedatasets must be especially diligent when specifying and testing theory, especially withregards to what evidence would effectively falsify such theory. Failing to do so increases therisk of false discovery