Statistically Significant Nonsense
Posted on Mar 23 in Issues in Customer Research, VoC by robert gerst
If you don’t believe it, why are you using it?
Almost all customer survey research, probably the research you are using now, assumes that what is important to people can be determined by the formula for standard error – significance testing.
If your research draws attention to what is statistically significant, then you are being sold snake oil and it is costing you. It is costing you in deteriorating employee satisfaction and declining customer loyalty.
The formula for standard error works well enough for polling, where the problems concern how best to spin political messages. But for those of us in the real world, concerned less with spin and more with bottom-line results, confusing statistical significance with practical importance is a disastrous error. It means most of what you think your research is telling you is dead wrong.
How can statistical significance determine what is important to customers?
It can’t.
When you are ready to leave The Cult of Statistical Significance, and begin identifying the real issues and problems in the employer–employee relationship, give us a call or drop us an e-mail.
Improved customer relations is just around the corner.
