If only I were a hard-working, hard-drinking Jewish agnostic

Unexplained correlations

A sex survey published in the United States this week sheds important light on the welfare debate currently raging in this country. With the swapping of statistics on benefit fraud, the amount of abuse of disability payments and the existence or otherwise of a dependency culture, the all- important question is raised yet again about whether we are seeing causation or simply correlation. This week, fortuitously, assistance in the interpreting of such figures has come from Penn State University, under whose aegis a major study of sexual activity involving 10,000 Americans has just been completed. And the results of this very large survey have sent academics and sociologists into a frenzy of evaluation, as they compete to explain the results.

The first important figure tells us that there is less of this sex stuff going on than we thought. (Although a warning here: as far as I can tell onanism is not counted as sexual activity by the Penn State researchers.) Twenty per cent of respondents had not "had" sex in the previous year, while 5 per cent did it at least three times a week. Overall this led to a statistic revealing that, while 15 per cent of the people had 50 per cent of the sex, the remaining 85 per cent had to share the rest between them - the sort of figure one associates with grossly unequal and unfair societies.

So, if you are an American who is not having a lot of sex, someone somewhere is getting your share. One day, an observer might speculate, the sexually dispossessed will rise up and occupy the boudoirs of the over-privileged. The Peter Stringfellows of Muncie and Boise will be found, suspended from the lamp-posts by their ...

But hold on to that rope, for there is much more interesting material to come. A series of figures breaks down the types of people and groups within society who are most and least likely to have more than their fair ration of sex. And there are some surprises here. Those who work long hours report having sex more often, as do those who watch the largest amount of television. At a rate of 111 times per year, married couples under 29 do it more than singles of the same age (72), or older married couples (figure not given, but I suspect around the six mark).

Those with doctorates manage less houghmagandy than those with first degrees only, jazz enthusiasts more than rock music lovers, Catholics a weeny bit more than Protestants but 20 per cent less than agnostics or Jews. Smokers do it 10 per cent more than the average, drinkers 20 per cent more, and those who both drink and smoke claim to have twice as much. In a final twist, everyone claims to have more sex during periods when confidence in the presidency is generally low.

Now, we are entitled to ask some questions about these results. It may well be, for instance, that drinkers simply think they've had more sex than they actually have, or that heavy smokers are forced - by their shortness of breath - to count things as being sex which the rest of us don't. Baths and suchlike.

But once we have dealt with any obvious anomalies, we must move on to explanations. And this is where we are forced to consider the difference between correlation and cause. Is it the fact of one's occupation or religion or musical preferences that determines one's sexual behaviour? Or is there just a correlation, which is actually caused by the general characteristics that led to some of these choices in the first place?

The significance of this question cannot be overstated. Theoretically, the person most likely to have the most sex is a twentysomething, married, Republican Jew with religious doubts, who works long hours, failed his or her doctorate and sits in front of the TV watching the Jazz station all evening, chain-smoking and boozing.

Now suppose that you are a teenage American who wishes, during his or her lifetime, to experience the optimum level of carnality. Were you to believe that these statistics indicated clear causal (as opposed to correlative) relationships, you would now begin to take action. If you were a goy you would go through the process of conversion to Judaism as soon as was practicable. And once converted, you would stand outside the synagogue on Saturday muttering, "I'm just not sure, Lord!" You would take up smoking at once, and force yourself to drink copious amounts of alcohol until you were fully acclimatised.

As soon as you came home from work or studies - which would be very late - you would flop down in front of the box, taking this opportunity to catch up on the cigarettes and drinks denied you during the day. In the car you would, at all times, play contemporary jazz tapes while complaining loudly about how Bill Clinton is sending the country to hell in a handcart.

When we scrutinise such an itinerary, it does not seem - on the face of it - to be one that holds out the promise of many long nights of writhing physical pleasure. Far from it. Personally, I would not want to sleep with such a person.

So I'd draw one of two very different conclusions from the above scenario. The first is that two surveys have become mixed up and misattributed in the Penn State computer - one on the amount of sexual activity undertaken, and the other on the incidence of early death from heart failure.

The other conclusion is that what we have here is merely a set of random correlations, and that we still have no idea what it all means. Which brings us back, I'm sure you will agree, to the Welfare debate.