Combatting Fraud in Market Research

Discussion within our industry about levels of fraud and data quality have been increasing, due to a range of factors including advances in technology and a seeming proliferation of bad actors. Data reliability is an important principle at the heart of our industry, that what we produce is fit for purpose is critical for confidence, not only for decision makers but as a way of positioning the industry as a source with integrity, produced using scientific principles.

There are many component parts to the discussion about fraud; the research eco system itself is extremely complex, ways in which to describe the eco system are varied and there are no universal quality signals. Laid on top of this, we have the amplification potential of AI.

There is minimal discussion about the difference between poor research practice and genuine fraud. This is a critical element as some estimates note that poor practice can contribute up to half of ‘fraudulent data.’ It is important not to downplay the effects of genuine fraud within the discussion, but it is essential that the contributing factors are broken down into their component parts so that they can each be addressed.

To tackle these issues industry associations have set up a global initiative called GDQ (Global Data Quality). One area that ESOMAR has focussed on within this initiative is the way in which researchers’ interface with participants. There is much discussion about over long surveys acting as a deterrent to research participants, but this is an over-simplification of the problem. If the participant can get through the screening process, in itself a barrier, there are issues of tone and topic, repetitive question sets, an arguable overuse of likert scales, and lengthy lists of questions.

‘How to improve research participants experience and enhance data quality’, provides a systematic approach to reviewing all aspects of the way in which researchers interact with research participants covering skills, training and education of the researchers working on studies, survey design, survey length, screen out procedures, quality assurance procedures, survey satisfaction, tackling drop out, evidence of good survey design and regular review. The working group hopes that all researchers will consider reviewing the quality of their work as a matter of business practice against this framework.

Focussing on this material does not seek to downplay the importance of tackling genuinely fraudulent behaviour, but it does tackle issues associated with an inappropriate labelling of poor research practice and by doing so, it has the potential to substantially reduce the amount of problematic research data.

The Global Data Quality initiative (GDQ) is an initiative involving ESOMAR and its partners including MRSI, the Insights Association, SampleCON, CRIC, the MRS, the Research Association, QRCA, AQR, VMO