ROANOKE TIMES 
                      Copyright (c) 1996, Roanoke Times

DATE: Sunday, December 1, 1996               TAG: 9612030121
SECTION: EDITORIAL                PAGE: 3    EDITION: METRO 
SOURCE: GEOFF SEAMANS EDITORIAL PAGE ASSOCIATE EDITOR 


MILLIONS WHO VOTED, MILLIONS MORE WHO DIDN'T

WHAT IF they gave an election and nobody came? The question gets less hypothetical by the year.

The upward spike in the 1992 presidential election, when 55 percent of adult Americans voted, proved a temporary aberration. Last month, the steady decline since World War II in voter participation resumed, to less than half the country's adult population.

Chicago Tribune reporters Charles Leroux and Ron Grossman have noted that the causes of voter apathy run deeper than dissatisfaction with less-than-inspiring candidates, or the public's unhappiness with the huge amounts of money spent on campaigns.

Voter apathy, they suggest, is simply part of the way we live nowadays; re-engaging the electorate, they argue, would require returning to the days "[b]efore the domination of the automobile and the suburbanizing and franchising of America ....:"

If they overstate their case, I suspect it isn't by much. At the least, blaming the low turnouts of today solely on unsatisfying candidates and disgust with campaign overspending fails to explain why the many uninspiring candidates and extensive political corruption of yesteryear did not produce the same low-turnout results.

For all that, voting isn't entirely a lost art. Some 93 million Americans did cast ballots in the 1996 presidential election, and some 2.3 million Virginians voted in the U.S. Senate race between the two Warners, John and Mark.

Two news headlines in this paper the day after the election drew criticism in a letter to the editor published the next week, and the writer raised an interesting question.

One headline said, "Clinton re-elected easily." The second, over another story, said "John Warner wins narrowly." Clinton's victory margin of eight percentage points, however, wasn't much bigger than Warner's of six.

"What constitutes the difference between an easy victory and a narrow one?" the letter writer asked sarcastically. "Is it 7 percentage points?"

Well, yes, in a way - if the eights points is the margin in a presidential contest and the six is the edge in a U.S.Senate race. There are two reasons, one structural and one statistical, for looking at presidential-vote margins in a way different from those in elections for lesser office.

Structurally, the poll that counts in presidential races isn't the total popular vote but the Electoral College vote. In that, Clinton outpolled Dole 379-159 - which is to say, easily. How easily? He could give up the electoral votes of all the states he carried with less than 55 percent of the two-party vote, and he still would be re-elected.

Statistically, the key point is that the bigger the universe of data, the more significant is each percentage point of variation from the average. Win a presidential election by 20 points - that is, 10 points above the average candidate (in a two-person race) - and it's considered an avalanche. Win by 10 points, it's a landslide. Win by eight, as Clinton did? No landslide, but plenty of breathing room.

Yet win a seat in, say, the U.S. House by just 10 points, and you're on the list of potentially endangered incumbents. The difference: The presidential electorate consists of tens of millions of voters; a House electorate, only tens of thousands.

But what about John Warner, who won by six points among an electorate of hundreds of thousands? For that, "narrowly" might be a little strong, though "wins by a moderate margin" would sound strange. Still, John Warner didn't win "easily," as Clinton did.

In true 1990s fashion, the letter writer saw political bias in the contrasting headlines. That's silly. (For one thing, this newspaper editorially endorsed both Democrat Clinton and Republican John Warner, though I should add that news and editorial are separate departments.)

But did the headline writer(s) succumb to a sort of pollster bias? That's possible.

Clinton's actual percentage margin turned out to be a bit less than what most of the late-campaign public polls were showing, but not astoundingly so. By contrast, the relative tightness of Warner's race, not polled nearly so intensively as the presidential contest, came out of the blue.

(To be sure, Mark Warner in the campaign's closing days claimed the race had tightened to eight points. But that was discounted, as such things usually should be, as coming from a self-interested source.)

To many folks on Election Night, John Warner's six-point margin looked small, smaller than it really was, because earlier polls had shown him so much farther ahead.

The fallacy, of course, is treating public-opinion polls as if they were genuine elections. Come to think of it, the preponderance of the former - giving people the impression, however false, that election outcomes are predestined - may be another reason for declining participation in the latter.


LENGTH: Medium:   88 lines






































by CNB