November 25, 2019

Data Quality at its Best


Recently there has been a lot of discussion about panel quality in the Nordics, which was mainly focused on recruiting methods. We welcome this discussion because we are convinced that it is key in our industry to be able to distinguish between a good panel and a, well, not so good panel.

But in our mind this discussion doesn’t have to stop with panel recruitment. We think that the quality of maintaining the panel is as important for the outcome. It’s never too late to mess it up, as the famous saying goes.

A couple of weeks ago, we were informed by Mindshare from Norway that we are on the right path. They compared our panel to one of our larger competitors and I am happy to share some facts.

Our client used a sample size of 500 respondents per panel to carry out the exact same questionnaire with two panels who with exactly the same quotas. After the data was collected, speeders and straight-liners were checked in each panel.

Out of the 500 respondents from our panel, the client identified four respondents who qualified as cheaters (either speeders or straight-liners), compared to 76 respondents in our competitor’s sample.

In other words, at least 15% of all answers in our competitor’s panel were useless and could lead to wrong insights. Christian Norby Boe at Mindshare states: “There is no doubt that the Norstatpanel compared to the competitor has significantly higher value”.

In the market research business it is too often about getting the answers to the cheapest price and very rarely buyers of sample question the quality of the data collected. We’d like to lead this discussion more often. Therefore, regardless of which panel you choose, feel free to contact us and have a discussion regarding how to measure quality and how to improve quality in your data. We’re looking forward to hearing from you.