Monday, September 24, 2012

The Precision of Words

Data is blunted by words
that are imprecise.
I recently reviewed a report and reacted a little bit like @angry_mr_client.  Not out loud of course but deep inside I rumbled like a freight train over a lack of precision.   Interestingly, it was not a mathematical error or lack of base size, which is a common cause of precision issues.  It was a word choice.

Mathematics is a nearly universal language originally used to describe phenomena observed in the natural world.  Over time maths have become abstractions - stretching their utility to explain things we can only imagine.  My children, in fact, have not seen a number in a math class for a couple of years.   In relative terms, probability and modeling, arguably our most mathematical domains in market research, are simple.  What has become more complex and therefore less precise is the other language used to describe observable and imagined phenomena:  words.

Interpretation of market research is a hairy business.   Most practitioners have no problem regurgitating the numbers and calling out what is significant.  This is what I affectionately call "monkey math."  I would expect to be able to train a reasonably smart ape to do this and even he would be insulted if called a monkey.  I also reserve the term for times when I stretch the numbers to do things you wouldn't even learn in Pilates.  Math that makes a monkey of me is also monkey math.  But sometimes I do it.

I lose my cool when words are used incorrectly - unintentionally so - to interpret the data.   A recent report I received said that an event occurred "rarely" because half of those asked about frequency said the event "rarely" happened.  The problem with this is that more than half of respondents experienced the event and those were the people asked the follow-up question.  So the event happens OFTEN - very often and not at all rarely.  Among those who experience it, the event occurs rather infrequently or rarely, perhaps only the first time they use the product but we don't know.  What we do know is that it's not RARE!  This makes me crazy.

It's time for more precision with our words so that our numbers aren't blunted in their impact.

** ** **
Today's guest post is from Kelley Styring. Styring is principal of InsightFarm Inc. a market research and consumer strategy consulting firm. She has led insights for Procter & Gamble, Pepsico, Black & Decker and NASA prior to founding her own firm in 2003. Kelley is a published author and has been featured in USA Today, ABC News, Good Morning America, Brandweek, Fortune, Quirk's Marketing Research and The Market Research Daily Report from RFL Online.  She will be live blogging from The Market Research Event 2012 this November 12-14 in Boca Raton, Florida.  If you'd like to join her, register today and mention code TMRE12BLOG to save 15% off the standard rate!

No comments: