PEOPLE who say that you can't learn anything from popular television have obviously never watched Family Feud. The lesson I just learned from that game show is that we continue to live in a sexist society.

In a fierce battle of wits between the DeWisses and the Ebsendons earlier this week, the host asked the family leaders to give the most popular responses to a very simple question: "Name something a husband expects his wife to do." From the 100 men surveyed by the game show, only five answers made it up on the board: 64 men answered "cook;" nine said "clean the house;" eight said "do the laundry;" six said "make love" and four said "do the dishes."

This information presents certain problems. First of all, would ten times more men really expect their wives to cook as to make love? Well, a group of 100 men is small enough that it may be subject to sampling errors. A larger group would certainly have said "make love" more frequently and would likely have included such answers as "replace Mom," "iron my shirts" and "have large litters."

All joking aside, though, the survey results reflect a very serious problem in our society. Since it requires two people to give the same response for an answer to go up on the board, it is frightening to consider the answers that were not listed. "Get a job" or "Help support the family" were said by fewer than two men out of the 100.

IF the sample was accurate, it means that at least 99 percent of the men in our society think they will marry June Cleaver. Or more accurately, 93 percent think they'll marry Mrs. Cleaver, while another six percent think they'll marry Ms. December.

Of course, the problem may lie in the way that Family Feud adds. Those surveyed were counted according to their first response. A pretty substantial number of men might have said, "Cook. Well, half the time--the other half I'll do it," or "Make love. Then go to the office and help support the family while I stay at home with the kids." Another possible explanation is a sampling error. It could be that the 100 men who were surveyed by Family Feud all happened to be chauvinists.

The most plausible explanation, however, is that the survey is a pretty accurate representation of what American men think--at least in their initial thoughts about gender relations. The survey is similar to word-association games in that it records only the first response. With unconscious sexism still so pervasive in our society, even among the well-intentioned, it is possible that a large majority of men would say something sexist when they respond before thinking.

Sexism is clearly a problem, but things may not be so bad as Family Feud makes them seem. The response that most men I know had when I told them what happened on the show was to ask "Who the hell did they survey?" Far more men at Harvard would give a reasonable response, simply because many of us have been made aware of unconscious sexism by our peers.

Still, it is far too easy to imagine someone watching the show who doesn't realize that it is all right for men and women to share domestic responsibilities and have equal job status. If that person learns that a huge majority of men think women should be relegated to a lower caste, the result is a subtle, but effective way of reinforcing inequality. But knowledge of how people think is the key to changing their views, even if it is just knowledge we get from game shows.