Gap between perception and reality not such big news

Written by Web Manager on . Posted in Features

Nobody can have missed recent ’England, a nation of secret binge drinkers?’ headlines spawned by new research ‘How is alcohol consumption affected if we account for under-reporting? A hypothetical scenario’.  Everyone, it seems, was shocked that 40-60% of the alcohol we buy is not included in the amount we say we drink.

Whilst here it’s the size of the gap between perception and reality which has proven so startling to the media, finding a difference between what people think/say they do and what they actually do is not, in itself, big news. Think about it – when you ask family, friends, colleagues a direct question about any of their habits, would you expect them to answer honestly, accurately and to all reply in a uniform way?

Bias -  a skew in the availability of data which makes observations of a certain kind more likely to be reported and used – can and does occur in surveys. Statisticians have ways of adjusting for this. People may not respond to questions or respond unhelpfully. They may respond inaccurately to hide their true behaviour.  They may not be able to assess their own behaviour or maybe simply they don’t remember.

The ’How is alcohol consumption affected ….’ study was undertaken by colleagues in the Department of Epidemiology & Public Health, UCL and published in the European Journal of Public Health). It aimed to predict the implications of under-reporting of alcohol consumption in England esp. what this means for levels of alcohol consumption above government drinking thresholds*. Also to show what alcohol consumption would look like when all of what is sold is accounted for, if everyone under-reported equally.

Whilst lobbyists and government officials pounced on the report to support their plans or to defend their existing strategy, the researchers termed their findings ‘putative’ and made clear the need for further study to gather the evidence required before any new policies or education initiatives are developed.

The researchers now aim to investigate the characteristics of under-reporters: who is drinking the ‘unclaimed’ 40% of wine and 60% of spirits? (clearly not everyone does under-report equally). How much do they drink and why do they under-report? and what does it mean for their health? Initial thoughts on who? Known under-reporting groups include the under 16s and people living on the streets, who would not have been captured by the surveys, but there will be others too. Initial thoughts on ‘why’ are that under-reporting more generally may be due to chaotic patterns of drinking. Mixing what, where and when you drink makes it less likely that you will’ self-report’ accurately.

Returning to the ‘startling’ findings of this research…anecdotally, GPs ask patients what their weekly alcohol unit intake is, then take the number and double it to be safe. Given the interim findings of this research, what doctors do instinctively does seem to make good sense.



The researchers looked at the two main surveys offering up alcohol consumption data:

a) the General Lifestyle Survey (GLF) in 2008, this annual longitidinal survey analysed weekly alchool consumption in 12,490 adults  b) the Health Survey for England (HSE), this annual survey is designed to be representative of the adult population in England. It looked at consumption on the heaviest drinking day in the previous week among 9,608 adults. To compare consumption with the total amount of alcohol sold, the researchers looked at Revenue and Customs data. Both the GLF’s and HSE’s alcohol consumption figures are self-reported i.e. based on respondents’ own feedback. The researchers used logistic regression (a useful technique for evaluating the relative importance of a large number of independent variables to survey response, the dependent variable), to estimate how widespread levels of drinking over the Government guideline levels actually are.

* 21 weekly/4 daily units for men and 14 weekly/ 3 daily units for women (weekly limits are set by the Royal College of Physicians  and daily) (not to be regularly exceeded) limits are set by the UK Chief Medical Officers

Join the RSS

Join the RSS

Become part of an organisation which works to advance statistics and support statisticians

Copyright 2019 Royal Statistical Society. All Rights Reserved.
12 Errol Street, London, EC1Y 8LX. UK registered charity in England and Wales. No.306096

Twitter Facebook YouTube RSS feed RSS feed RSS newsletter

We use cookies to understand how you use our site and to improve your experience. By continuing to use our site, you accept our use of cookies and Terms of Use.