|One of the major limitations of self-report data about crime or punishment is that few longitudinal studies have asked respondents the same questions over extended periods of time. In response to the problems of incomplete data series, political scientists have used algorithms to create a single measure of public mood by incorporating several different data sources that ask respondents similar questions about a policy issue over time. This process focuses on what we don't know - missing values - to what we do know, by combining similar series from different sources over time. This study examines several indicators of public mood created using algorithms, including political disaffection, racial policy preferences and public policy mood. Implications for the study of criminology and public opinion over time using algorithms are discussed.
(Return to Program Resources)