Do market researchers over-rely on sample size as proof of survey data quality?
Do we under-invest in other data quality contributors, such as instrument design and scale validity? In this conversation, Kathryn and special guest Ken Faro discuss current thinking on survey research data quality. One key topic: the use of derived measures as a more accurate way to understand customer attitudes and emotions.
That is, sometimes it makes sense to ask people about their attitudes directly (say, about a brand or product), but in some cases that approach is risky; people may overly-rationalize or otherwise have biased responses. More indirect, derived approaches can mitigate these risks, resulting in better data.
Here’s a helpful example from a recent article Ken authored in the Sloan Management Review; “…many market researchers still ask questions the old way, descriptive of what’s being measured (“I like the ad I just saw”) rather than descriptive of derivative behaviors (“I showed the ad to friends”). Of course, doing this well also requires thinking about use of single-item versus multi-item scales. For more details, please check out the episode!
About special guest Ken Faro, Ph.D.: Ken is the VP of Research at Hill Holiday. Be sure to check out Ken’s related (and important!) article in the October 2018 Sloan Management Review.
This video podcast episode is a great follow-up to episode 69, on 8 ways to improve survey response rates.
Please share this video with any colleagues who might want to learn more about survey data quality. And if you think the video has value, please do like and subscribe on YouTube! Prefer podcasts? Also available on iTunes.