Ah, the pollsters! They tell you how you're going to vote weeks before you've made up your mind. Then, on election night, they tell you how you voted even before your vote was tabulated. And now, of course, they're explaining to you why you voted the way you did.
Of all these highhanded acts, the postelection analysis may be the hardest to take. With absolutely no respect for your individual quirks, you're broken down into a demographic statistic - a nonhuman entity defined by the intersection of your geographical locale, age, sex, education, income, and so on.
Did the wing-tip-cordovan types with 1.8 children vote Democrat or Republican? Were the Japanese-car-owners under 37 for or against the nuclear freeze?
Somewhere there is someone who can tell you - and probably will. Fast! For pollsters practice the rudeness of people who finish your sentences for you before you can spit out the second word.
Yet elections polls and their I-told-you-so analyses are exercises in humility compared to polls on the issues. Everybody knows that a poll predicted Alf Landon would defeat Franklin D. Roosevelt in 1936, and many a poll called for another landslide in 1948 - Thomas E. Dewey over Harry Truman. It has remained for a new study from the Charles F. Kettering Foundation, significantly titled ''Polls Apart,'' to emphasize how contradictory - if not downright wrong - polls can be on policy questions.
The authors, John P. Robinson and Robert Meadow, note that two network polls managed a spread of 50 percentage points in estimating public support for SALT II.
Polls counting the number of Americans who favored sending arms to Israel came up with answers that ran all the way from 31 percent to 66 percent. Take your pick.
Do Americans believe in the United Nations? One poll reckoned that only 32 percent of us do, while another poll submitted a figure two and one-third times larger.
Still, a lot of us keep on accepting the stats of a poll as scientific evidence - hard facts. Give us a number, and we fantasize precision: a carpenter with a rule, a chef with a teaspoon, a chemist with a thermometer.
Outside of politics things can get even softer. A poll in Psychology Today announces the percentage of men and women who ''took vitamins in the last 24 hours'' and on the basis of this and other data - for instance, how many respondents roller-skated during the past year - presumes to generalize that Americans ''now spend more time thinking about our health than about love or money.''
Then there are the ''mood'' polls, which seem to be rivaling those grim polls on sexual behavior in popularity and recklessness. ''Are you blissfully happy? Are you inconsolably wretched?'' - check one. This sort of question is getting asked more and more. Even U. S. News & World Report featured on its cover ''A Nationwide Mood Survey.'' The results: 54.5 percent of us aren't too happy with things now, but 56.2 percent of us expect to cheer up soon.
If only these were harmless parlor games - the sociologist's equivalent of Chinese fortune cookies. But polls have a way of manufacturing a synthetic world of their own. They create what Walter Lippmann called a ''phantom public.'' In place of human beings they substitute ''mean'' man and ''average'' woman.
If we accept these dubious abstractions as us, we let them express what only an individual human being can express: a preference, a conviction, an opinion.
If somebody asks us what we think, some day will we have to run to the nearest poll to find out? That's the ultimate nightmare. Not to worry. People just aren't robots with multiple-choice levers waiting to be yanked. We're all whimsical folks, praise be, and as likely as not to turn mischievous when somebody with a clipboard asks us a dumb question that leaves no room for a nice , rambling answer.
We're not saying no to all polls. But we're certainly not saying yes either. Mark us down for ''Undecided.''