Send email for updates
Taylor M.J., Vlaev I., Maltby J.J. et al.
Health Psychology: 2015, pre-print.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Taylor at firstname.lastname@example.org. You could also try this alternative source.
‘Social norm’ interventions which aim to reduce consumption by telling heavy drinkers how their drinking compares to their peer-group norm have a patchy record, but this British study suggests for students they might be improved by ranking against peers (eg, ‘You drink more than 80% of students’) rather than comparing how many units of alcohol they consume.
Summary Excessive drinkers typically underestimate their consumption relative to that of others. ‘Social norm’ interventions aim to reduce consumption by correcting these misperceptions, telling people how their drinking actually compares to the general population or relevant subgroups of the population, such as those of the same sex and age.
‘Social norm’ interventions aim to reduce consumption by telling heavy drinkers how their drinking compares to the norm, correcting overestimations of how much others drink.
A randomised British trial among college students made this comparison in two ways: comparing against average drinking of same-sex students, or ranking (eg, ‘More than 80% of other students’) against the same benchmark.
Though relative to other types of messages the ranking comparison did not reduce drinking over the following month, it did stimulate requests to find out more about alcohol and how to get help.
For the first time the featured study compared two types of ‘frames’ for social norm messages, in this case for excessive drinkers among UK university students. One compared their drinking against the average drinking of people of the same sex in the same sample of students (“You drink X units per week, the average male participant drinks Y units per week”). The second ranked their drinking against the same set of students (“You drink more units per week than X% of (or X out of 10) male participants”). Messages to remaining students made no comparison with other students. They were either just told the official guidelines for maximum alcohol consumption, or their drinking was compared to these guidelines.
Students at two British universities were invited to join the study through email and university social media. Of the 146 who responded and completed baseline assessments, 101 were drinking excessively according to their answers to the three questions of the AUDIT-C screening questionnaire. Students were then randomly allocated to be sent one of four sets of four weekly messages containing one of the types of information described above. When comparisons were made, it was against the drinking of all 146 students, including those who did not screen as risky drinkers.
A month later 78 of the 101 heavy-drinking students responded to a follow-up assessment including questions about their drinking in the previous week, and were offered the opportunity to seek further information in the form of expert recommendations on alcohol consumption, links to web sites about alcohol consumption, or contact details of services for people worried about their own or someone else’s drinking.
All four sets of students had reduced their drinking and risk levels as measured by AUDIT-C, but these reductions did not significantly differ depending on the type of messages they had been sent. However, compared to the other three sets of students, those sent the ranking messages were significantly more likely to seek further information (55% did so compared to at most 14% in the other groups), requested significantly more types of information, and in particular were significantly more likely to ask for contact details for alcohol services. The other three types of information did not significantly differ in their impacts on information-seeking.
The study showed that a minor reframing of how comparisons are made in a social norms intervention can greatly increase its effectiveness at promoting information- and help-seeking. Social norms interventions have not been found consistently effective, but most involve telling recipients how they compare to the average person. As in other studies, in this study that type of comparison was no more effective than simply informing students about safer drinking recommendations. However, being told where they ranked among other students stimulated requests for more information on drinking and sources of help.
Findings suggest that social norms interventions might benefit from focusing on telling people how their behaviour ranks amongst others, and that more generally, consideration should be given to how interventions can be improved through presenting information in ways more naturally processed by the recipients. Future work could also investigate whether social norm messages with a ranking frame reduce drinking as a downstream consequence of encouraging excessive drinkers to seek appropriate support.
commentary This innovative British study may have found a way to improve the patchy record of interventions based on messages comparing the recipient’s drinking with the norm for the population as a whole or for more closely related subgroups. Brief face-to-face interventions based on this tactic appear to have a better record at reducing drinking than automated procedures of the kind tested in the featured study. From the featured study it seems possible that this is partly because the framing of these messages was sub-optimal, but information-seeking does not necessarily translate into reduced drinking. Some students will not be displeased to hear that they rank among the heaviest drinkers, and unless selected out, others will find themselves being told they drink relatively little, and might see this as an invitation to ‘catch up’.
Thanks for their comments on this entry in draft to research author Michael Taylor of Imperial College in London, England. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 06 October 2015. First uploaded 25 September 2015
Comment/query to editor
Give us your feedback on the site (one-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates