How Do You Calculate the Gender Gap?

Despite the extensive news coverage of the gender gap for the past two decades, more recent years have witnessed some confusion in its methodological calculation. For example, in 1996 election return coverage, television networks calculated a gender gap of 16%, while the New York Times described an 11% gender gap.

1996 Election

Networks compared the percentage of women voting for Clinton with women voting for Dole, while the New York Times calculated the difference between levels of support for Clinton between male and female voters. Still other political analysts calculated the gender gap by measuring the difference between women's support for Clinton and Dole and comparing that figure with the difference between men's support for the two candidates, which produced a 17% gap.

So, what was the gender gap in 1996 and how do you measure it? The gender gap represents the difference in voting patterns between men and women. The correct way to calculate the gender gap is to assess gender differences in preferences for a particular candidate, party, or issue. In 1996, the New York Times methodology was right. President Clinton's victory was the result of an 11 point gender gap.

Comparing women's level of support for one candidate with women's support for another candidate does not accurately capture differences in voting between women and men, which is the clear objective of "gender gap" methodology. The impact of the gap on election outcome cannot be assessed by adding the difference between women's level of support for two candidates to the difference between men's level of support for the two candidates.