Fisher et al. (2006) developed a hierarchical Bayes model to estimate the number of people without health insurance within demographic groups for states. The Centers for Disease Control and Prevention are interested in estimates of women without health insurance by demographic groups in families that earn less than 200% of the poverty line. Our approach jointly models direct estimates from the Annual Social and Economic Supplement to the Current Population Survey (CPS ASEC), and Census 2000 Sample Data, tax, food stamp, and Medicaid data, using a multivariate, hierarchical approach. We have improved the preliminary model in Fisher et al. (2006) by adding census data, improving the mean and variance models for the direct estimates and the administrative records data, and developing a raking procedure. In addition, for variance estimation, we have developed a method that takes into account the variance of the direct estimates that are used in the raking procedure.