Every so often, we write an article knowing few people will read it. And let's face it, survey sample selection bias is not a hot topic.
But you should read this article because survey selection bias is a growing problem.
This is because easy-to-use online survey tools allow people who are untrained in survey methods to easily and quickly create and field surveys.
Schwab's Choiceology podcast Survey Says covers what selection bias is and why it's a problem.
The podcast starts with the story of the 1936 presidential election, which Franklin Roosevelt won in a landslide over Republican Alf Landon.
What few know about that election is that the Literary Digest presidential poll, considered the gold standard of presidential polls at the time, predicted Alf Landon would not just win - he would win in a landslide.
This, of course, was a huge miss. Key quote from Social Science History's "President" Landon and the 1936 Literary Digest Poll:
"The disastrous prediction of an Alf Landon victory in the 1936 presidential election by the Literary Digest poll is a landmark event in the history of American survey research in general and polling in particular."
The prediction was so bad that the Literary Digest - a respected and successful magazine - went out of business two years later.
The failure was due to selection bias. Simply put, the Literary Digest's survey sample heavily overweighted Landon supporters. This led to an erroneous prediction that Landon would win.
Because we're industry analysts, we review dozens of surveys yearly. More than a few suffer from selection bias, and we know not to trust their results.
So if you're a user of surveys or someone who fields surveys, you need to know about selection bias - and how to identify and avoid it.
And listening to the Schwab podcast is a quick and easy way to do this.
