Back to News List


3 ways to combat under-reporting of “bad” behaviours in market research

May 12th, 2021

Particularly in quantitative research, respondents are often asked to report the frequency with which they perform certain behaviours. Depending on the research topic, social desirability bias (1) can have an unintended, yet significant distorting impact on respondent reporting. This either takes the form of under-reporting of “bad”, or unreasonable behaviour – or over-reporting of “good” behaviour.

Below are 3 ways researchers can successfully combat the first of these – under-reporting of “bad” behaviour:

1. MAKE THE BEHAVIOUR MORE ACCEPTABLE: If, for example, we want to understand average alcohol units consumed per week, we may well be concerned some respondents might under-report their behaviour. By ensuring higher levels (e.g. 30 units / week) appear more towards the middle of answer group options presented to respondents (rather than towards the upper end), we help to “normalise” this level of consumption, making it seen as more natural and accepted.

2. ASK ABOUT OTHERS: An alternative approach, which draws on social psychology, doesn’t attempt to counter social desirability bias as regards a respondent’s own responses. Instead, by asking respondents to report on, for example, the alcohol consumption habits of “other people like them”, we are able to get more accurate and honest answers.

3. DISGUISE THE QUESTION: Finally, an advanced quantitative method called the unmatched count technique has recently been used successfully for this exact purpose (2), and involves asking respondents to admit – at a total level – to how many, of a number of different behaviours (one of which is the behaviour of interest), they have performed. While this approach does not allow us to find out which individual respondents are under-reporting “bad” behaviour, critically it helps us deduce a more accurate incidence level for this behaviour at the total sample level.

By using one – or preferably more than one – of the above methods, we can make sure we get a much clearer picture when it comes to understanding true incidences of “bad” behaviours.

Chris Harvey
Founder, Activate Research

1. Krumpal, I. (2013). Determinants of social desirability bias in sensitive surveys: a literature review. Quality & Quantity, 47, 2025–2047.