In randomized controlled trials, intention-to-treat analysis is customarily used to estimate the effect of the trial. However, in the presence of noncompliance, this can often lead to biased estimates because intention-to-treat analysis completely ignores varying levels of actual treatment received. This is a known issue that can be overcome by adopting the complier average causal effect approach, which estimates the effect the trial had on the individuals who complied with the protocol. When compliance is unobserved in the control group, the complier average causal effect estimate can be obtained via a latent class specification using the gsem command.
Large research efforts have been directed at the exploration of ethnic disparities in the criminal justice system, documenting harsher treatment of minority ethnic defendants, across offence types, criminal justice decisions, and jurisdictions. However, most studies on the topic have relied on observational data, which can only approximate ‘like with like’ comparisons. As a result, researchers, practitioners and policy-makers have often been wary of interpreting such disparities as evidence of discrimination. We use causal diagrams to lay out explicitly the different ways estimates of ethnic discrimination derived from observational data could be biased. Beyond the commonly acknowledged problem of unobserved case characteristics, we also discuss other less well-known, yet likely more consequential problems: measurement error in the form of racially-determined case characteristics or as a result of high heterogeneity within the ‘Whites’ reference group, and selection bias from non-response and missing offender’s ethnicity data. We apply such causal framework to review findings from two recent studies showing ethnic disparities in custodial sentences imposed at the Crown Court (England and Wales), questioning whether the reported disparities should be interpreted as evidence of discrimination. We also use simulations to recreate the most comprehensive of those studies, and demonstrate how the reported ethnic disparities appear robust to a problem of unobserved case characteristics. We conclude that ethnic disparities observed in the Crown Court are likely reflecting evidence of direct discrimination in sentencing.
In the study of sentencing disparities, class related hypotheses have
received considerably less attention than explanations based on
offenders’ ethnicity. This is unfortunate since the two mechanisms are
likely interrelated, at the very least as a result of their overlap in
the population, with ethnic minorities being generally more deprived
than the White majority. In this registered report we propose exploring
the mediating and moderating effects between offenders’ area deprivation
and their ethnic background using a novel administrative dataset
capturing all offences processed through the England and Wales Crown
Court. Specifically, we seek to test two key hypotheses: i) the reported
ethnic disparities in sentencing are mediated and explained away by area
deprivation; and ii) ethnic disparities are moderated by area
deprivation, with ethnic disparities being narrower in the more deprived
areas. Results from this empirical analysis will shed new light on the
underlying causes of sentencing disparities, but crucially - if
deprivation is shown to play a major role in the generation of ethnic
disparities - they will also help inform the adequate policy responses
to redress this problem.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.