Session: 1234567891011121314151617         Page 12345678910111213141516171819202122232425262728293031323334353637383940414243444546 of 755

stratified -- we didn't call it that until years later -- of stratified interview samples where we made sure we got people from the low income, low education and right across the board.

One thing led me to another and that was you can't be sure what people tell you they do. I learned that in some of my course work. I would ask the kids what magazines they read. And if you took their answers and put them up against circulation -- Atlantic Monthly was way out ahead and you knew damn well that they weren't reading it. Number one. And number -- two that wasn't a sample that represented real sample outside. And, again, that sort of pushed me in the direction of what Starch and some of the people working at that time talked about as “adequate samples.”

Because I didn't trust what people said they did with respect to radio, the first place I thought even with some of the faculty people that looked down their nose and made fun of radio, there were things that would sneak out in their conversation that suggested to me that they knew more about what was on the air than they were telling. And that certainly was true in the early days of television as well. I remember interviewing somebody one time -- or not interviewing, but at a dinner party somebody said, “Oh, of course, we never look at television.” And then later in the conversation having the guy say to me, “What kind of a woman is Kitty on Gunsmoke?” So you know that there was some viewing going on and that was true also in radio.

So I had this idea that there ought to be a little recorder that you could put on the radio that would record when the set was on.

Where did you get that idea? I mean, how did that come up?