January 14, 2019

expert reaction to adolescent well-being and digital technology use

Research published in Nature Human Behaviour demonstrates that the association between digital technology use and adolescent well-being is negative but small.

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

“What’s important about this study is that it puts findings from large data sets about relationships between technology use and adolescent well-being firmly into their context. Much (though certainly not all) of the evidence for these relationships comes from studies that re-used data from very large groups of young people that were originally collected for a different or less specific purpose. Findings from that approach have some major difficulties of interpretation. It’s pretty well impossible to establish what causes what, so if the data shows a link between technology use and well-being, that could indeed be because the technology use actually causes well-being to be worse, it could be because being depressed causes young people to use technology more, it could be because some other factor is causing greater technology use in some young people and independently causing those people to have lower measures of well-being, or (most likely) the whole picture of cause is much more complicated than any of these. Then, in findings from such large data sets, a relationship can be highly statistically significant, that is, it’s very likely not just a random consequence of the choice of which people to study, but can still be very small in size, so that it’s of little or no practical importance. That’s just how statistical significance works. Finally, in this kind of data there are many different ways to measure technology use and well-being, and to look for possible relationships between them, and in previous studies, the researchers have chosen just one, or a few, ways of doing these things, and choosing different ways could have given different results.

“The new study deals with the issue of researchers choosing different ways to investigate the same data set by looking at a very large set of choices of analysis for each of the three data sets involved, by investigating each of them by a wide range of different methods (up to 41,000 different ways) and then presenting and summarizing the results from all these different approaches. This does lead to a quite wide range of different estimates of the strength of the relationship between technology use and well-being in adolescents, and not just the median figure that only 0.4% of well-being is explained by technology use. But, in my view, the new research does establish that the relationship is very likely to be weak, and that technology use very probably explains much less about well-being than do things like smoking cannabis, eating breakfast, or getting enough sleep. The relationship is statistically significant, but that’s not the point.

“The researchers also give reasons for doubting that the relationship between technology use and adolescent well-being is one of cause and effect. Some of these reasons are based on technical statistical points, but they are also based on the findings that the relationship is similar in size to those between riding a bicycle or eating potatoes and well-being, where a causal link seems unlikely. I agree with their conclusions here. The new research can’t establish that the relationship between technology use and well-being is definitely not causal, any more than any research based on this kind of data can establish that it is definitely causal. But, in any case, should we really be spending so much time and effort on discussing how technology use should be restricted or changed to improve the lives of our teenagers, given that the impact of technology use on well-being is likely to be small? Surely we should spend more discussion time and policy effort on tackling other things whose effects on well-being are bigger and based on better evidence.”

Dr Max Davie, Royal College of Paediatrics and Child Health Officer for Health Promotion said:

“This study supports the findings of our recent review of the evidence base on screen time, that while there are some negative associations between screen time and poor mental health, we cannot be sure that these links are causal, or if other factors are causing both negative health outcomes and higher screen time.

Despite the fact that technology is an increasingly essential part of daily life, the current evidence base for its health impacts is limited. It is therefore encouraging to see much-needed new research in this area, as we know that technology use is an issue that causes anxiety for parents. Our recently-published guidance encourages parents to approach screen time based on their child’s developmental age, the individual need, and the value the family place on positive activities such as socialising, exercise and sleep.  Parents should feel empowered to adjust the amount of time spent on screens by them and their children, depending on what’s important to their family life”.

Dr Ben Carter, Senior Lecturer in Biostatistics, Institute of Psychiatry Psychology & Neuroscience, King’s College London (IoPPN), said:

“This is an interesting paper that further develops the link between digital technology use and mental health, specifically linked to adolescent wellbeing. The study strengths include using three existing established cohorts to explore population level data. The findings report weak associations between digital technology use and adolescent wellbeing and conclude they are not large enough to impact on public health policy changes.

“It’s really important to note that the definition ‘digital technology use’ in this paper is not the same as social media use. It encompasses the use of the TV, computer, electronic games and mobile phones as well, and then calculates the average use of all of these (although some surveys did not specify social media use at all). Also, as lots of the data is from a time before adolescents had smart phones and tablets etc. (some of the studies collected data from as far back as 2007), this makes it hard to relate these findings to adolescents in 2019.

“This study was only able to look at the duration of use and not the way in which the technology was being used (the content and time it was being used) which could possibly provide a better indicator of well-being. For example, is the technology use of a twelve-year-old spending a few hours on a computer doing school work during the day the same as if it was say, 10pm at night and scrolling through social media posts for the same length of time? Therefore, the way in which technology is used by young people and the impact of social media specifically is something we need to look at in future studies.”

‘The association between adolescent well-being and digital technology use’ by Amy Orben et al. was published in Nature Human Behaviour at 16:00 UK time on Monday 14th January.

All our previous output on this subject can be seen at this weblink: http://www.sciencemediacentre.org/tag/screen-time/

Declared interests

Prof Kevin McConway: “Kevin McConway is a Trustee of the Science Media Centre.”

Dr Ben Carter: No conflicts

None others received.