Recent studies have sparked lively debates about White working class voting in the U.S. On one hand, there is evidence that the White working class has shifted to vote Republican, and on the other hand, there is evidence that the White working class continues to vote Democratic. Using the American National Elections Studies (ANES), we examine 32 years of presidential election results (1972–2004). We utilize theoretically justifiable measures of routine white collar, skilled and unskilled workers, which we then combine into broad and restricted measures of the White working class. Also, we distinguish between the male and female White working class and consider a variety of other predictors of vote choice. We scrutinize the entire period as a whole, each election year individually, and then concentrate on the 1996–2004 elections. In the entire period, the odds that White working class men voted Republican were greater by a factor of about 1.35. By contrast, White working class women were not significantly different from the electorate. Yet, this conceals temporal variation. Although the White working class was more Republican than the general electorate in 1972 and 1984, and more Democratic in 1976, the White working class generally was consistent with the electorate through 1992. Since 1996, there has been a sharp divergence between the male and female White working class. Men very strongly supported Republicans 1996–2004, while there is some evidence women supported Democrats in 1996 and 2004 (but not 2000). From 1996–2004, the odds White working class men voted Republican were greater by a factor of about 2.0. These patterns are not confined to the South and hold despite controls for a variety of predictors of vote choice.