wru: Who Are You? Bayesian Prediction of Racial Category Using Surname and Geolocation
Software to predict individual-level race/ethnicity using surname, geolocation, and other attributes, such as gender and age. The method utilizes Bayes' Rule to compute the posterior probability of each racial category for any given individual. This method is described fully in Imai and Khanna (2015) "Improving Ecological Inference by Predicting Individual Ethnicity from Voter Registration Records."
Type install.packages(“wru”) to install in R. Please see the ‘wru’ GitHub page for more information.CBS News Poll Database
Proprietary database and application to analyze long-term trends in CBS News surveys, from the 1970s to the present. When I first joined the Election & Survey Unit, trends were compiled manually by going through old poll releases… quite a laborious process! Building on my dissertation research, this database of historical polls and complementary web application quickly combines and weights data relevant to a particular trend analysis, such as presidential approval over time. The application also allows users to filter certain types of respondents.