A recent paper by Louis-Philippe Beland (Louisiana State), Richard Murphy (University of Texas at Austin), published in the journal Labour Economics (ungated earlier version here), provides some suggestive evidence about mobile phones, but in the context of secondary school. Specifically, the authors looked at data from students at secondary schools in four cities in the UK (Birmingham, London, Leicester, and Manchester) and at school mobile phone policies. They looked at how the implementation of a mobile phone ban at a school affected students' performance in the GCSE exams (which are standardised national exams).
Of course, the problem here is that schools will be more likely to implement a mobile phone ban if they believe that it will positively affect their students (including by increasing their learning), but Beland and Murphy rightly identify that this means that their results offer an upper bound of the effects of banning mobile phones from schools. It may not be a big issue though - only one of the 91 schools in their sample didn't implement a mobile phone ban at some point over the period they look at (2001-2011).
They find that:
...following a ban on phone use, student test scores improve by 6.41% of a standard deviation. This effect is driven by the most disadvantaged and underachieving pupils. Students in the lowest quintile of prior achievement gain 14.23% of a standard deviation, while, students in the top quintile are neither positively nor negatively affected by a phone ban.The stated effects are only marginally statistically significant, and the effect (0.06 standard deviations) is not huge (which is why I labelled this as 'suggestive evidence' above). Beland and Murphy undertake a battery of robustness checks though, which suggest the effect is real (if possibly over-stated, given the self-selection of schools into banning mobile phones). Some of the supplementary results are interesting in their own right though, such as:
The interaction of the ban with prior achievement is negative... implying that it is predominantly low-ability students who gain from a ban.The high-ability students (as measured by their results in testing prior to starting secondary school) were not affected by the mobile phone bans, but the low-ability students were made better off. This is similar to the effects of laptops in university classes I blogged about last week. Also:
...the impact of the ban is mostly on language based subject English and other subjects with no impact on mathematics.Are mobile phones more of a distraction in English classes than in maths? I'm not convinced. This result is also a bit strange:
We find that less strict bans of mobile phones are more effective in raising student test scores than bans that prohibit phones to be on school premises. Exploring the impact of strict and less strict bans by prior student achievement, we find that both are effective in raising the test scores of the lowest performing students, but again that less strict bans are more effective.It's hard to see how a weak ban on mobile phones (e.g. students may bring them to school, but they must be on silent mode) would have a greater positive impact on student outcomes than a stronger ban (e.g. students cannot bring mobile phones to school at all). Beland and Murphy suggest that a weak ban may require more teacher time to enforce. Maybe. Perhaps the English teachers are more vigilant at enforcing bans than maths teachers too?
So, this is clearly not the last word on this topic. The negative effect may be small and statistically significant, but some of the other results need further exploration. Laptops may be bad for learning, but for mobile phones the evidence is not so clear.
[Update]: Peter Lyons says something closely related in this Herald opinion piece.
- Strong evidence that laptop use in lectures is bad for learning
- Is it time to ban laptops from lectures?