I am the furthest thing from a mathematician, but I am always curious where percentages and statistics that we all come to accept as truth, come from. You know what I’m talking about. Fifty percent of all marriages are destined to end in divorce. Stuff like that.
I was reading the story in today’s paper about obesity projections for 2030. Apparently, half of us will be grossly overweight here in Pennsylvania–and it’s worse elsewhere. Like you, I read a lot of these types of projections on a weekly basis. This morning, I found myself asking, “How do they know this?” Furthermore, how do we know for certain that our obesity rate is what it is right now in this country? Where do those numbers come from? No one has weighed me lately.
I know that weight is recorded whenever I go to my doctor’s office. But let’s face it. Not everyone goes to the doctor’s office for regular checkups. And I’m sure there’s a statistic on that, too. Again, based on what?
These are all rhetorical questions, because even I understand it all comes from applied statistics. But I was curious enough to look into the obesity stats that we all accept as reality, namely, according to the CDC, over one third of the American population is obese. To make that assertion, the data had to come from somewhere.
According to its website that data came from “National Health and Nutrition Examination Surveys (NHANES) conducted from 1999 through 2010…NHANES is a cross-sectional survey designed to monitor the health and nutritional status of the civilian noninstitutionalized U.S. population (8). The survey consists of interviews conducted in participants’ homes, standardized physical examinations conducted in mobile examination centers, and laboratory tests utilizing blood and urine specimens provided by participants during the physical examination.”
So it’s a survey. What was the sample?
“NHANES sample is selected through a complex multistage probability design that includes selection of primary sampling units (counties), household segments within the counties, households
within household segments and, finally, sample persons from selected households.”
How big a sample? According to an addendum to the report, it looks like the sample is about 10,000. And the data is weighted and all kinds of analysis and complicated interpretation goes on, of which I know nothing. But at the end of all of this, comes the end point: We’re fat.
And I am not saying we’re not. I’m just saying it’s interesting.