Utilizing data from the 2000 Census, this study examines the impact of family composition, education, and labor force factors on the difference between female and male poverty rates in the 70 largest U.S. cities. A stepwise regression analysis indicates that 41 % of the difference between female and male poverty rates can be explained by the percent of women in the three US Bureau of Labor Statistic's lowest wage occupations. There was no evidence of a unique impact from the percentage of female headed families in each city, or the study's other independent variables, on the gender poverty gap, with the exception of their contribution through the proportion of females in the lowest wage occupations. This study provides empirical support for the likely ineffectiveness of TANF initiatives promoting employment and marriage for alleviating female poverty.

In addition, the study found important non-random geographic variations in the difference between cities with the highest and lowest gender disparity in poverty rates. Only one of the ten US cities with the highest rankings in gender poverty disparity is located west of the Mississippi River.