The Summit on Measurement, Part 4: Don Stacks and The Loneliness of the Measurement Mathematician

Mathematics is at the heart of a fascinating and frustrating contradiction inherent in public relations measurement. To do proper measurement, you need to use solid mathematics to reveal the patterns and trends in the data. But many -- probably almost all -- of those who provide or use or buy PR measurement services have little interest or inclination in understanding the complex mathematics required to answer their simple question: "Did the PR work?"

To those of us who enjoy mathematics, it is the beautiful and elegant bridge between messy nature and tidy logic, between a million human minds and a single line that reveals how they think. But to those who do not enjoy math, or who have no time to build the careful framework of knowledge required to understand the practical statistical methods of measurement, it is an opaque and confusing mystery. A mystery whose secrets are revealed only to Don Stacks.

Now, Dr. Don Stacks is not the only person in measurement who knows a chi-square distribution when he sees it. But he is one of the very, very few people in measurement with both the mad skills and the balls to announce to a Summit full of the best and brightest minds in measurement that, "If your measurement provider doesn't know what R-squared is, then tell them to take a hike." And of course most of those best and brightest don't themselves know what R-squared is, and don't really want to know what R-squared is, but are afraid to let anyone know it because Dr. Stacks is looking very severely at them all like they are back in Stats 101 and they forgot to do their homework.

"A good consumer of research," says Dr. Stacks, "needs to understand the concepts."

I sat in on the Measurement 301 pre-Summit workshop, a review of best practices and statistical methods. There were only three or four of us attending, while meanwhile next door about 40 people packed a session on social media. Oh the irony of it all: There are Don Stacks and David Michaelson (another major measurement math guy) reminding us that samples should be random and that research objectives should be well defined and I'm thinking, "Well duhh, this is such basic stuff why doesn't everybody already know it?" And then I'm thinking, "But this is the most important and basic stuff in measurement so why isn't this room jammed full of eager students?" Ah, Math: So beautiful, yet so lonely.

And the point here is that so few people can or will ever really understand the math behind proper measurement, yet if we want to do proper measurement, we have to know that the math is solid. And we can't all aspire to the skills of Drs. Stacks and Michaelson, who, after all, are the guys -- the guys who write the books and train the graduate students and in general make sure that the gleaming Temple of Measurement Math stays clean and tidy and available for us all.

But, even the most math-challenged among us can appreciate a graph with a smooth line that explains the data. There has got to be a way to provide the power of math without the intimidation of mathematics. Somebody please invent a Stats Box into which we chuck all the data and out pops the charts and graphs. -- Bill Paarlberg

Comments

[Bill Paarlberg]

Dr. Stacks has reminded me that the important question is, "What is 1-(R-squared)?" (What you don't know.)