Like all the rest of you, I have never won a prize in a lottery or contest (even when the letter comes announcing that I have ALREADY WON!, it turns out that my prize is the opportunity to compete in another contest). I also have never been polled by Gallup or Harris, and have never formed part of that mystical band of families who determine which television shows shall live and which shall die.

Many Americans still believe that there are no such families, and that the ratings are merely concocted out of thin air by people who sense what we watch. And since the replacement shows are almost always imitations of the show which have gone before, it doesn't seem to matter.

Believe it no longer. You are reading the work of the head of an Arbitron family, and because of the preferences we expressed in April, I have no doubt that heads will roll in June.

Arbitron, or Audience Research Bureau as it was known until four years ago, is the nation's largest purveyor of television and radio ratings. It is, sad to say for out prejudices, all quite scientific (at least until you get to the diary-keepers). Arbitron rates programs in all 206 broadcasting "markets" in the nation. These are called, scure jargon, "ADIs," or Areas of Dominant Influence, by which is meant, simply, the place where everybody watches the same stations. Arbitron rates only commercial stations, by the way, thus saving public television considerable embarrassment.

In the Washington ADI, for example, the latest Arbitron ratings survey is based on the viewing habits, self-reported, of 3285 families selected at random from people with listed telephones. In ADIs where it is believed there is a substantial number of people without, or with unlisted, telephones (Appalachia, say, or Beverly Hills), Arbitron has developed an "expanded sample frame" to include them.

There follows, once the families for the month (each serves for one solid week of viewing) are selected, a letter explaining the survey and requesting participation, and then a telephone call to confirm. At this point, two variables (as we social scientists used to say) set in. In the case of Spanish-surname families, a biligual viewing diary is sent, and a biligual interviewer is provided, each day of the survey.

The other variable is astonishing, considering that it involves more than twenty-five per cent of viewers in this ADI: families assumed to be black are asked a series of questions designed to elicit that fact. However, they are not sent diaries, but instead are asked each night on the telephone which programs the family watched. The reason, according to Arbitron spokesmen, is that black families don't return the diaries. One wonders if the data supporting that assumption are as scientifically gathered as the rest.

The diaries themselves are comprehensive, and one is provided for each television set in the household. Families are asked to record, for each fifteen-minute period from 6 a.m. to 2 a.m., every program watched or if the set was not use. Each member of the viewing family is listed by sex, age and race. Income, education and ethnic background are not sought by Arbitron.

The Other rating service, which Arbitron people are reluctant to identify, is Nielsen, and they have far fewer families in the whole nation than Arbitron has in Washington alone.Of course, the Nielsen families are not selected at random, but carefully identified as to their representation in the population. Further, Nielsen puts a mechanical device on the set and thus know what you're watching, not content to rely on either your honesty or you diligence.

Arbitron, by the way, pays each family - first a quarter, then a dollar, then a thank-you letter. It's little enough for such a test of honesty. How many people are like me, I wondered one evening, who are embarrassed to tell even Arbitron that they're watching "The $25,000 Pryamid," when they should be watching "The McNeil/Lehrer Report?" Washington, one suspects, is full of people who tell Arbitron they watched a steady diet of Cronkite, Chancellor, Walters, "60 Minutes," "Meet the Press" and "King Lear" - but who are closest viewers of "Charlie's Angels" just like everybody else.