September 4, 2015

expert reaction to screen time, sedentary behaviour and GCSE results

Screen time (watching TV, playing computer games or browsing the internet) and sedentary behaviour have both been at the centre of recent controversy, causing particular concern amongst parents worried that their children are being harmed through long periods sitting and looking at a screen. Publishing in the International Journal of Behavioral Nutrition and Physical Activity, researchers have conducted a study to determine if looking at screens, compared to reading or doing homework, or being sedentary appears to have any impact on GCSE results. These Roundup comments accompanied a briefing.

Dr Andrew Przybylski, Experimental Psychologist and Research Fellow at the Oxford Internet Institute, University of Oxford, said:

“The study presents an intriguing thesis, that time spent on some forms of sedentary behaviour, namely homework and reading, might have different relations to academic performance than other forms, such as TV viewing, Internet use, and computer game use. The strengths of the design include a focus on a relatively large sample and the use of physiological assessments of activity in addition to self-report metrics. Understanding these features, it would be unwise to interpret these findings as evidence that screen time has an effect on academic performance for two central reasons.

“First, although the study does collect data at different points of time it is not, strictly speaking, a prospective study. We do not know anything about the academic inclinations or aptitudes at the start of the study. As a result, we do not have a baseline to evaluate the potential enduring effects of different forms of screen time on academic performance. It is entirely possible that there are important factors that we have no information on that are driving the effects. It is possible that those who play computer games take on difficult subjects or that high TV-watchers also happen to be impulsive and make errors on tests. Without knowing the baseline of the participants it is not possible to draw these inferences.

“Second, the analytic approach used in the study is mainly concerned with whether or not the p-values of the statistical tests are “statistically significant” that is, report levels below the p < 0.05 level.  It does not focus necessary attention on the effect size, or practical significance linking the factors under study. Given that this study uses a very large population and conducts a number of statistical tests the p-values that range from 0.052 (not significant; for computer game time) to 0.011 (significant; for TV viewing time) the fact that these values fall below the threshold say more about the study design than the phenomena under study. As a result of this approach, and the omission of necessary statistics in reporting, i.e. a Proportion Reduction in Error (or PRE) statistics such as an adjusted R^2 it is not possible to judge the size of the effect. This is crucially important because it appears (from Table 3) that every hour of TV viewing may be linearly negatively related to 9.6 GCSE units (this is statistically significant), whereas every hour of gaming is negatively related to 9.8 GCSE units (but is not statistically significant). For some reason, 0.2 GCSE units are making the difference between whether or not a claim can be made. The lower p value does not mean it’s more detrimental per se. Given the numbers available I would estimate that less than an average of 2% of the variance is shared between these screen time measures and academic performance. But, because this information is missing from the paper, it is not possible to know.

“Taking the two points together, I think it would inadvisable for readers of this work to take it as evidence that the: “implementation of greater priority by parents to reduce screen based behaviors involving TV, Internet, or Computer Games” is needed. As it stands there is an extensive literature dealing with parental strategies concerned with mediating child engagement with media such as TV, games, and the Internet. This work suggests that time-based limits and restrictions are far from effective when compared to strategies that involve active parent involvement. As it is, this work presents very modest correlational evidence that there is a link between test scores and screen time.”

‘Revising on the run or studying on the sofa: Prospective associations between physical activity, sedentary behaviour, and exam results in British adolescents’ by Corder et al. published in International Journal of Behavioral Nutrition and Physical Activity on Friday 4^th September. 

All our previous output on this subject can be seen at this weblink: http://www.sciencemediacentre.org/?s=%22screen%20time%22&cat

Declared interests

Dr Przybylski: “I do not have any interests which might be regarded by a reasonable and objective third party as giving rise to a conflict with your role as an SMC expert in this story.”