In Census Bureau classifications, married-couple households only consist of opposite-sex couples, while unmarried partner households may consist of either opposite- or same-sex couples. This classification relies not only on the accuracy of the responses to the household relationship item—either as a spouse or unmarried partner—but also to those on gender. Although gender is usually the most accurately reported item on a survey, minor errors in gender could have a substantial impact on the estimates of same-sex unmarried partner households. This paper outlines some of the issues in estimating the number of same-sex unmarried partner households and the potential effects of these errors on the estimated population.