The people in television news are on a collision course with public opinion on an issue of some importance to our politics, and it would be well for all of us in the press if the collision could be avoided.

The issue is the broadcast of presidential election returns into states where the polls are still open -- especially, the early projection of an election decision while voters are still weighing how, or whether, to vote.

There was an uproar in the West last November, when NBC News led its two rivals in calling the election for Ronald Reagan at 8:15 p.m. eastern time, almost three hours before the voting ended in California and other Pacific Coast states. There were reports of people leaving the lines outside polling places, of drivers on their way to the polls going home instead. There were accusations from losers of close local races that the broadcasts had cost them the election.

My suspicion at the time was that the accusations were exaggerated and that the uproar would soon subside. But on the latter point, at least, I was wrong.

When I spoke recently to civic audiences in Spokane, Wash., and Grants Pass, Ore., the first question each evening was whether the networks would "inflict" their projections again in the 1984 election.

A late January poll of California voters by Mervin D. Field found 74 percent of those interviewed would prohibit such projections before the polls have closed. Bills to ban such broadcasts, introduced in the heat of last fall's anger, have been reintroduced by such senators as James McClure (R-Idaho) and S. I. Hayakawa (R-Calif.). Hearings are planned in the House. The problem is not going away, and the closer we come to 1984 without a resolution, the more anger and controversy there will be.

As a journalist, I have a lot of sympathy for the networks' position on this issue. As Bill Small, the president of NBC News, says: "To me, the thought of suppressing something is awful." And despite the anecdotal evidence, it is very hard to show through statistics that TV projections discourage voting turnout significantly -- let alone to the disadvantage of one party or candidate.

The most careful sifting of the statistical evidence, by Raymond Wolfinger and Peter Linquiti in the February-March issue of Public Opinion magazine, suggests that the impact may be in the range of a 2.7 percent falloff. That's not huge, but it's not insignificant either.

What is clear is that the belief that "my individual vote counts" is important to many, many people, and they are offended by being told that their act is without significance to the outcome.Yet, it is also a fact that an Electoral College majority may be collected in this country before the West has cast or counted its votes for president.

So what is to be done? Some bills would ban the broadcast of any presidential election returns anywhere until the polls have closed everywhere. That not only is offensive to the First Amendment, it is nonsensical. Why should Maine voters wait to learn what they have done until the polls close in Alaska?

Other bills would restrict journalists' access to areas around voting places and thus impede the interviews that are sometimes used for voting projections. Again, the First Amendment would surely be damaged by that precedent, and, in some areas, the absence of press scrutiny would also invite voting frauds.

Other bills provide a uniform poll-closing time across the nation. That is achievable, but only at extra expense and inconvenience for the voting officials in some regions, or the deprivation of some voters of the convenience of voting before or after work.

The simplest solution, it seems to me, is for the American networks to do voluntarily what their Canadian counterparts are already required to do by law when a federal election is held there: activate the broadcast networks by time zone, from east to west, as the polls close.

That means the networks can begin broadcasting in the East at 7 p.m. as the polls close, bring in their Midwest affiliates an hour after that, the Mountain States an hour later, finally, the Pacific Coast.

To be effective, the segmenting and sequencing would also have to apply to the major news services -- Associated Press and United Press International. In that way, no voters would be denied prompt results after the polls had closed in their region, but no one would have them imposed unwillingly while the polls were still open.

I do not think this raises First Amendment questions. Not unless that constitutional guarantee is violated when the networks routinely make West Coast viewers wait three hours later than East Coast viewers to see John Chancellor, Dan Rather or Frank Reynolds on their nightly newscasts.

The newscasts are timed to the convenience of most viewers, and I think those western viewers and listeners deserve to be taken seriously when they say it suits their psyches not to be told how the election came out until they have had a chance to vote.