Candidates' policy platforms and election outcomes: The three faces of policy representation

Abstract.   Recent empirical studies by Alvarez and Nagler, Erikson and Romero, and others conclude that candidates' and parties' policy platforms only modestly affect their electoral support. This suggests that candidates/parties can win elections even when their policies differ sharply from the policy beliefs of the constituencies that elect them. This raises the question: How can voters exercise control over government policies via elections? We report applications to American and French presidential election data that suggest three paradoxical conclusions. On the one hand, we find that presidential candidates can moderate their policies with at most a modest change in vote share, but if they move by the same amount to a more extreme position, they face severe vote losses that could cripple their election prospects. Alternatively, movement by both candidates in the same direction or a policy shift by the voters may have a major effect on the outcome.     

THEORIZING SANCTIONING IN A CRIMINALIZED JUVENILE COURT*

Recent juvenile justice reforms have produced increasingly complex and criminal‐like approaches to sanctioning youths, yet research to date has not examined the full range of newly available sentencing options nor systematically drawn on theories of adult sentencing. The present study addresses these issues by developing competing hypotheses about the effects of legal, extralegal, and processing factors, as well as sentencing options, in a highly proceduralized and criminalized juvenile court in Texas. These hypotheses are then tested using quantitative and qualitative data. The results are largely consistent with derived expectations and do not support arguments that increased proceduralization and criminalization of juvenile courts will eliminate consideration of age, gender, or race/ethnicity in sentencing decisions.