Reliability and validity of the 2002 and 2006 Chapel Hill expert surveys on party positioning

This research note reports on the 2002 and 2006 Chapel Hill expert surveys (CHES), which measure national party positioning on European integration, ideology, and several European Union (EU) and non‐EU policies. The reliability of expert judgments is examined and the CHES data are cross‐validated with data from the Comparative Manifesto Project, the 2003 Benoit‐Laver expert survey and the 2002 Rohrschneider‐Whitefield survey. The dataset is available on the CHES website.     

How Parties Create Electoral Democracy,Chapter 2

Parties neither cease to exist nor cease to compete for office when the general election is over. Instead, a new round of competition begins, with legislators as voters and party leaders as candidates. The offices at stake are what we call “mega‐seats.” We consider the selection of three different types of mega‐seats—cabinet portfolios, seats on directing boards, and permanent committee chairs—in 57 democratic assemblies. If winning parties select the rules by which mega‐seats are chosen and those rules affect which parties can attain mega‐seats (one important payoff of “winning”), then parties and rules should coevolve in the long run. We find two main patterns relating to legislative party systems and a country's length of experience with democratic governance.     

THE MISSING LINK IN GENERAL DETERRENCE RESEARCH*

Research on the deterrent effects of punishment falls into two categories: macro‐level studies of the impact of aggregate punishment levels on crime rates, and individual‐level studies of the impact of perceived punishment levels on self‐reported criminal behavior. For policy purposes, however, the missing link—ignored in previous research—is that between aggregate punishment levels and individual perceptions of punishment. This paper addresses whether higher actual punishment levels increase the perceived certainty, severity, or swiftness of punishment. Telephone interviews with 1,500 residents of fifty‐four large urban counties were used to measure perceptions of punishment levels, which were then linked to actual punishment levels as measured in official statistics. Hierarchical linear model estimates of multivariate models generally found no detectable impact of actual punishment levels on perceptions of punishment. The findings raise serious questions about deterrence‐based rationales for more punitive crime control policies.     

RACE,RACIAL THREAT,AND SENTENCING OF HABITUAL OFFENDERS*

As legislatures proliferate novel “enhancements” to criminal sentencing, such as “three-strikes” and related provisions, and as criminologists debate their effects, the role of existing enhancements, such as habitual offender statutes, has received little empirical attention. This article explores the effect of race in the decision to prosecute and sentence eligible defendants as “habitual” offenders. During FY 1992–93, 9,690 males admitted to prison in Florida were statutorily eligible (two prior felony convictions or one prior violent felony conviction) for sentencing as “habitual” offenders. Approximately 20% received that disposition. They will serve at least 75% of their enhanced sentence as compared with the state average of about 40%. Logistic regression, controlling for prior record, crime seriousness, and other relevant factors, shows a significant and substantial race effect. The disadvantage of black defendants is particularly strong for drug offenses and for property crimes that have relatively high victimization rates for whites (larceny, burglary). Race is less consequential for violent and weapons-related crimes. Race effects are more often significant in sentencing contexts that are low in terms of percent black, racial income inequality, drug arrest rates, and violent crime rates. The relevance of these findings for a “racial threat” interpretation of sentencing outcomes is discussed.     

PARENTS AND DRUGS REVISITED: SOME FURTHER EVIDENCE IN SUPPORT OF SOCIAL LEARNING THEORY*

Social control and social learning theories suggest divergent relationships between attachment to parents and children's drug use when level of parent drug use is considered. Social control theory proposes a uniformly negative relationship between children's drug use and attachment to parents whereas social learning theory proposes that the relationship is affected by parental drug use. The relationship between attachment to parents and children's drug use was investigated for each of three groups of low, moderate, and high parental drug use through estimation of a latent variable structural model of attachment to family on children's tobacco, alcohol, and marijuana use. Results indicate that attachment to parents related inversely, but with different magnitudes, to children's drug use for youths whose parents use drugs at low or moderate levels. No significant relationship exists between attachment to family and children's drug use for youths whose parents are relatively high-level users. Neither ethnicity nor sex affected these findings. The implications of these results supporting social learning theory are discussed.