IN A LEAD STORY LAST MONTH in the Wall Street Journal, Gerald Seib argued that President Clinton’s electoral fate now hangs on the gender gap. ” There have, of course, been gender gaps in past elections,” wrote Seib, “but today the gap appears to be widening into a chasm.” He cited a December 1-5 $ IWall Street Journal/NBC News poll that found 54 percent of women saying they would vote for Clinton “if the presidential election were held today,” but only 45 percent of men — a “gender chasm” of 9 points.
This kind of hype is just plain silly. There are indeed gender differences in partisan preferences, following a pattern that began taking shape in the 1970s. Nothing much is happening to this pattern — which finds women regularly giving more support to Democratic candidates and causes than men do; it shows no signs of expanding or shrinking. And it is far from clear how the partisan consequenc es of the gender gap should be read. Exit polls found Ronald Reagan backed by 5 5 percent of men in the 1980 three-way contest and by 47 percent of women — an 8 point difference. The gap was 4 points in 1984, 7 points in 1988, 4 points in 1992. The composite of he Dole-Clinton trial heats taken over the past two months falls within this range.
Seib argued that the president’s support among women voters “has increased amid fierce debate over the budget,” but the preponderance of available data says otherwise. I’ve reviewed the presidential approval scores, for example, from 18 surveys taken in November, December, and January. The proportion of women saying they approve of Clinton’s handling of the presidency ranges in these surveys from the low 40s to the high 50s with no consistent change over time. Overall, Clinton’s approval percentage averaged 5 points higher among women than among men — almost exactly the average gender gap throughout his presidency.
These men-women differences in presidential job approval reflect well- established gender differences in party identification. In 24 surveys taken from November through January where party data were available, the proportion of women identifying themselves as Democrats exceeded the proportion of men calling themselves Democrats by an average of 5.5 points. Conversely, the percentage of men identifying themselves as Republicans surpassed the percentage of women by an average of 6 points. Again, these recent differences match what polls have been finding for 15 years.
Finally, partisan differences by gender mirror differences about the role government should play. When questions of government’s responsibilities are raised in surveys or in state ballot questions, women typically declare themselves more supportive of government intervention than men do — by proportions similar to what we see on party preference. The rising incidence of economically needy single-parent, female-headed households since 1960 probably goes far to explain the emergence and persistence of the gender gap.
So it’s real — but hardly a chasm. And there’s no indication that it’s widening. Researchers plumbing divisions in American politics need to keep gender in their analytic arsenal, along with other key demographics. Yet if gender belongs in this company, the divide it locates is by no means the most prominent. Differences between blacks and whites are obviously much larger; but so are those among educational and income groups, among Americans by their level of religious participation, and between union members and non-unionists.
Like many students of electoral behavior, I’ve long been especially interested in the party preferences of the young. Young people are hardly blank slates politically — parental influences are clearly evident, for one thing — but they are less bound than older people by experience and thus usually more reflective of today’s trends. Well-designed political surveys of teenagers are few and far between, but one such was taken by ABC News January 24-28. It gives us a glimpse of the gender gap among the next generation of voters.
Asked which is more important, balancing the budget or maintaining the current level of domestic programs, the teens (ages 12-17) divided roughly two to one (60-33 percent) in favor of budget balancing. President Clinton fared roughly the same with males and females in the sample: 52 percent of the former and 54 percent of the latter said, for example, that the congressional Republicans’ position on the budget is closer to their own than is the president’s. And 45 percent of the males, along with 49 percent of the females, said they approved of the way Clinton is handling his job as president. No hint here of a widening gender gap — and no hint of Republicans’ being in trouble with the next generation.
The party identification data on the teens are striking. Like earlier generations when they came of age, today’s teens have a large proportion (50 percent) of self-described independents. But the 46 percent of the sample who declare a party preference break two to one Republican (30-16 percent). Republicans do 10 points better among the boys than among the girls (36 percent to 26 percent), but the Democrats do only 4 points better among the girls (18 percent to 14 percent). Democrats trail Republicans in both groups, but gender differences in party preferences among these young people are substantial.
On the key political question — which party, if any, has drawn advantage sin ce 1980 from the relatively modest but persisting gender gap — there can be no conclusive answer. Obviously the GOP should be especially interested in strengt hening its position among women, the Democrats theirs among men. It is worth no ting that the proportion of women identifying themselves as Democrats now is si milar to what it was 25 years ago. If the electorate comprised women alone, its partisan make-up in 1996 would resemble that in 1968 or 1976. Men, however, hav e shifted a lot over this span. The Republican party’s rise nati onally from minority to parity status has occurred as the party has gone from being way behind among male voters to being modestly ahead.
Everett Carll Ladd is president of the Roper Center for Public Opinion Research.
