September 2, 2020

expert reaction to study looking at electronic media use in late childhood and measures of academic performance

Research, published in PLOS ONE, looked at electronic media use in late childhood and measures of academic performance.

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This piece of research adds to the very large research literature on possible associations between children and young people’s use of electronic media, of many kinds, and aspects of their health (physical and mental), academic performance, and other related matters. In complicated situations, more data generally helps, but I don’t think that this new research makes the position much clearer.

“One issue with the great majority of research on this kind of topic, including this research, is that it’s observational – children and their parents aren’t asked by the researchers  to do anything different from what they’d do anyway. Because of that, there will generally be many differences between the children who make different uses of electronic media, other than simply how much they use these media. Those other differences might be the real cause of any differences in academic performance, and not the electronic media use itself. The differences in media use may be a side effect of the other factors, or even just a coincidence, and if that’s the case, any concern or action that’s specifically about the media use may not improve academic performance. This new research has an advantage over some other research studies, because it is (partly) longitudinal. That is, it looked at associations between media use at one age (8-9 years) and academic performance of the same children at a later age (10-11 years). That at least rules out the possibility that the academic performance is causing a difference in media use (as is usually a possibility in cross-sectional studies that measure media use and academic performance at the same time). But it doesn’t rule out the possibility that something else, that happens to be related to media use at the earlier age, is the real cause of differences in academic performance later. The researchers did adjust their findings to allow for several factors that might have been involved in this way, including the child’s age and sex, their family’s socio-economic status, previous academic test scores, and also a score on a standard scale of behavioural and emotional symptoms in the child (based on report from the parents) and the child’s BMI. Generally these adjustments did not make a huge difference to the measured associations between media use and academic performance, but that does not mean that all possible factors were taken into account. The researchers themselves mention as a limitation of their study that they could not make and adjustment for parenting style, for instance, because they had no data to do so.

“Previous research on these topics has produced a wide range of different, sometimes contradictory, findings. As just one example, there does seem to have been a previous piece of research* on similar numbers of schoolchildren of a similar age, partly using standard academic scores in the country where it took place (Germany), and which was also partly longitudinal (that is, it investigated an association between media use at one age and academic performance later). This found quite a marked negative association between some uses of video games, particularly by boys, and academic performance. The new research by Mundy and colleagues found no evidence of association between video game use and academic performance. I am not trying to claim that the German research findings are preferable to the new research findings – just pointing out that they look considerably different. One possible reason for such differences, which are widespread in this kind of field, is that there are a huge number of choices that researchers have to make in terms of what they decide to measure, what adjustments are made for other factors, and so on. An important study that was published last year** investigated this particular issue by analysing the same (large) sets of data in thousands of different ways, any one of which might have been chosen by researchers. They found that different choices could lead to very different findings of associations from the same datasets, ranging from quite strong negative to quite strong positive associations between media use and the well-being of adolescents. This work involved older ages than the new research, and different types of outcome measure (measures of well-being rather than academic performance), but my intuition would be that the same kind of thing may very well be going on in relation to research like this new study – what is found can depend very strongly on how you choose to measure and analyse the data. That’s a very clear reason to avoid drawing too many conclusions from any single study.”

* Mößle et al (2010) ‘Media use and school achievement – boys at risk?’ British Journal of Developmental Psychology, https://doi.org/10.1348/026151009X475307, see also Mößle et al (2014), ‘Longitudinal Effects of Violent Media Usage on Aggressive Behavior—The Significance of Empathy’, Societies, https://doi.org/10.3390/soc4010105

** Orben and Pryzbylski (2019) ‘The association between adolescent well-being and digital technology use’, Nature Human Behaviour, https://doi.org/10.1038/s41562-018-0506-1

Dr Sam Wass, Senior Lecturer in the Department of Psychology, University of East London, said: 

“This paper builds on the large number of previous studies that looked at correlations between screen time and academic outcomes.  It takes a slightly different approach – by looking at how screen time at 8-9 years affects reading performance at 10-11 years after controlling for (taking into account) reading and media use at 8-9 years, in order to focus specifically on how media use during this critical period affects reading development.  But this study still has the same fundamental problem of all correlational studies – which is that there are lots of differences between families where children spend lots of time viewing screens, and it’s impossible not to be sure that it’s these other differences, rather then the screen time itself, that are affecting academic outcomes. 

“Interestingly, though, many of the explanations that the authors suggest for why TV viewing might associate negatively with academic outcomes – such that it takes away from homework time, or time spent engaging in shared activities with parents – apply equally to video gaming.  But in fact, whereas strong associations were found between TV viewing and academic performance, no negative effects of video gaming were found.  This contributes to the emerging body of evidence suggesting that it’s not screen time on its own that’s good or bad – but that it’s essential to consider how the screens are being used.” 

Prof Dame Til Wykes, Vice Dean Psychology and Systems Sciences, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, said: 

“The authors followed up a large group of children in grade 3 to grade 5 and had their parents report how long they spent using different types of screen – TV, games and a computer.  The most interesting results are the longitudinal ones and the authors only detected two significant differences – a relationship of TV watching at Grade 3 with future reading and Grade 3 computer use on numeracy.  Interestingly the associations with reading disappear at Grade 5.  After all if you want to play a game or google you do need to be able to spell and read at a certain level. 

“So what does this study tell us?  Does it mean that cutting down on TV use will have a positive effect on reading?  Probably not, and it certainly doesn’t tell us that if we cut down on computer use then numeracy will increase.  If cutting down means parents will interact more with their children, that their children would read more books or play constructive games or activities then we are likely to see some benefits, but that would be because of what is added rather than because screen time is taken away. 

“Computer use, TV use and playing online games may actually be beneficial too.  We know that games sharpen your reactions, some of their social interactivity can also be useful for developing co-operation and strategy.  Watching programmes about the lives of people around the world, wildlife or science is going to help improve our children’s knowledge and we certainly wouldn’t want to ban Sesame Street! 

“Screen time is much more complex than measuring the hours online, even more so when it is your parents who are counting.  The study does have tantalising data that merits further investigation, but it should not be read as meaning parents need to take more control and cut down media use.  It all depends on what children do with their time and a mix of activities is what most children need.” 

‘Electronic media use and academic performance in late childhood: A longitudinal study’ by Lisa K. Mundy et al. was published in PLOS ONE at 19:00 UK time on Wednesday 2 September 2020. 

DOI: 10.1371/journal.pone.0237908 

Declared interests 

Prof Kevin McConway: “I am a member of the SMC Advisory Committee, but my quote above is in my capacity as a professional statistician.”

Dr Sam Wass: “I have worked as a media spokesperson for companies including Lego, Fisher Price, Aardman Animations and Nikleodeon.” 

Prof Dame Til Wykes: “I don’t have any conflicts.”