Ira Glass

It does not get corrected until much, much later, when childhood is long over.

Alex Blumberg explains.

I can reconstruct the events that led me to one of the most embarrassing conversations of my adult life.

The chain starts back when I was 11 or 12, and I first heard the term Nielsen family.

I was probably listening to some adults talk, and from their conversation I gathered that networks consulted Nielsen families to find out how popular a television show was.

But that didn't make sense.

Why would they only ask people named Nielsen which shows they liked?

I started thinking.

I knew that when they figured things like this out, they didn't ask everybody, they just asked a small percentage of people and then extrapolated.

I think I figured they'd done some research and found that the name Nielsen, because it was a common name maybe, and it seemed to cut across class and economic lines, actually came pretty close to a representative sample.

I knew this wasn't the way they measured public opinion now, but it seemed like the Nielsen surveys had been around for a while, and I figured they were just a holdover from a more primitive, less statistically rigorous time.

After that, I really didn't think about it again.

Or if I did, it was only with a mild curiosity.

I wonder why TV still does it that way.

Fast forward 20 years.

I was talking with a friend of mine who was telling me about her friend who'd been selected to be a Nielsen family.

And I said to her, isn't it weird that they're all named Nielsen?

My friend looked at me for what seemed like a long time.

Somewhere during her very long pause, because of the very long pause, in fact, I realized, of course they're not all named Nielsen.

That makes no sense at all.