National Affairs: Public's Opinion of Polls

Every election year Los Angeles' City News Service conducts a telephone poll of Los Angeles residents on a few major ballot choices, supplies the results to local newspaper clients. The polling is carried out mostly by college students, who pick the names at random from metropolitan Los Angeles' five phone books. Over the years, Editor Joseph Quinn has come to expect about 1,500 replies out of 3,000 calls. But this year things went wildly wrong. C.N.S.'s results in last week's poll on Nixon v. Kennedy, plus two local ballot questions:

Total calls made: 3,812.

Hung up without: listening to a single question: 2,107.

Listened to at least one question (usually only one), but refused to answer any because "all polls are rigged," or something to that effect: 1,671.

Willing to answer: 34.

Undecided on Kennedy v. Nixon: 19.

Voting on Kennedy v. Nixon: 15.

And how did the 15 votes divide between Kennedy and Nixon? Nobody at the C.N.S. is sure. "We were so shocked," explains Quinn, "that nobody remembered to tabulate the final answers before we threw the stuff away. Obviously, people have completely lost confidence in polls. Maybe the scientifically conducted ones are still O.K., but I wouldn't want to bet even on that."