Six experiments were conducted with primary age children (years 2 to 6) to investigate whether they learned better in the morning (9am) or afternoon (3pm). All the experiments were conducted in an independent school in the North East of England; all used random assignment to allocate the children to a morning or afternoon group. Four of the experiments were concerned with a listening comprehension activity, one with mathematical skills, and one with a test of non-verbal reasoning. An average of 38 children took part in each experiment.
In all the experiments, the procedures were standardised (eg identical instructions recorded on a tape) to ensure that the only difference was the time of day. Five of the experiments also used a delayed test, identical to the immediate test, but carried out one week later at the same time of day as the original exercise. One experiment also had a 4-week delayed follow up test. Altogether, 12 comparisons between morning and afternoon learning were generated.
Of the 12 effect sizes, 11 favoured the afternoon group, although only two of these showed statistically significant differences (the twelfth was also not significant). Effect sizes ranged from 0.84 to –0.16, with an overall average of 0.26 (p = 0.01).
The table shows a summary of the 12 effects. Click on each experiment for a link to the full TERSE Report.
These results suggest that, contrary to popular belief, children seem to do better when they are taught and tested in the afternoon than in the morning.