Person:

Winship, Christopher

Research Summary This article reviews the causal turn in the social sciences and accompanying efforts by criminologists to make policy claims more credible. Although there has been much progress in techniques for the estimation of causal effects, we find that the link between evidence and valid policy implications remains elusive. Drawing on criminological theory and research insights from disciplines such as sociology, economics, and statistics, we assess principles and strategies for informing policy in a causally uncertain world. We identify three distinct domains of inquiry that form a part of the translational process from evidence to policy and that complicate the straightforward exportation of causal effects to policy recommendations: (a) mechanisms and causal pathways, (b) effect heterogeneity, and (c) contextualization. We elaborate these three concepts by examining research on broken windows theory, policing, video games and violence, the Moving to Opportunity voucher experiment, incarceration, and especially the rich set of experimental studies on domestic violence that originated in Minneapolis, MN in the early 1980s. We also articulate a set of conceptual tools for advancing the goal of policy translation and offer recommendations for how what we call “policy graphs”—causal graphs used to analyze the policy implications of a system of causal relations—can potentially integrate the theoretical and policy arms of criminology.

Policy Implications Evidence, even if causal, does not necessarily inform policy. In fact, the question of “what works,” the focus of the growing evidence-based movement in criminology, turns out to be a different question than, “what will work?” Evidence-based policy research must therefore be concerned with much more than providing policymakers with research on causal effects, however precisely measured. The implication is that we must separate criminology’s increasing focus on causality from its policy turn and formally recognize that the latter requires a different standard of theory and evidence than does the former. In particular, criminologists interested in making policy claims must ask hard questions about the potential mechanisms through which a treatment influences an outcome, heterogeneous effects across people and time, contextual variations, and all of the real-world phenomena to which these challenges give rise—such as unintended consequences, policies that change incentive and opportunity structures, and the scale at which policies change in meaning. Theoretically guided causal graphs enhance this goal and help inform policy in a causally uncertain world. Translational criminology is ultimately a process that entails the constant interplay of theory, research, and practice.

Among the nation’s elite colleges and universities, black graduation rates vary dramatically from institution to institution. Many sociologists have suggested that this is due not to differences in the student bodies but to institutional factors; however, this “institutional hypothesis” has not been recently examined empirically. We test the institutional hypothesis for a set of elite institutions using College and Beyond, a restricted dataset containing data for the entire 1989 cohort of 27 elite institutions, matched to institution-level data, employing HLM techniques. We ask three questions: Do institutional factors affect black students’ probability of graduation? Do they account for between-institution differences in black graduation? And are institutions where blacks have a high probability of graduation the same as or different from those where whites do? Testing for the effect of eight major institutional factors, we find, surprisingly, that only selectivity has a statistically significant effect. Contrary to common belief, selectivity improves black probabilities of graduation, and helps blacks more than it helps whites. It also accounts for roughly 38% of the between-institution variance in black graduation. Finally, we find that after controls, black and white probabilities of graduation across institutions are highly correlated (.909), such that institutions in which blacks are likely to graduate are those in which whites are likely to graduate, too. Findings suggest that researchers should examine other institutional factors in greater depth, as well as the role of pre-college preparation more seriously.

We provide a new statistical model developed from the alumni database at Northwestern University for identifying potential major gift donors. Our logit model with interactions predicts which individuals will give $10,000 or more over three years using CHAID (Chi-square Automatic Interaction Detection) combined with logit analysis. Our work goes beyond our earlier research (Lindahl and Winship, 1992) by providing a more effective means of selecting fund-raising prospects. The critical new component in our model is the inclusion of interactions between past giving and other variables. Specifically, we find that for individuals with different past giving records different variables are important. Our results show that prospects with a low past giving level will rarely give gifts over $10,000.

This article offers a new approach to the identification of age-period-cohort (APC) models that builds on Pearl's work on nonparametric causal models, in particular his front-door criterion for the identification of causal effects. The goal is to specify the mechanisms through which the age, period, and cohort variables affect the outcome and in doing so identify the model. This approach allows for a broader set of identification strategies than has typically been considered in the literature and, in many circumstances, goodness of fit tests are possible. The authors illustrate the utility of the approach by developing an APC model for political alienation.

When experimental designs are infeasible, researchers must resort to the use of observational data from surveys, censuses, and administrative records. Because assignment to the independent variables of observational data is usually nonrandom, the challenge of estimating causal effects with observational data can be formidable. In this chapter, we review the large literature produced primarily by statisticians and econometricians in the past two decades on the estimation of causal effects from observational data. We first review the now widely accepted counterfactual framework for the modeling of causal effects. After examining estimators, both old and new, that can be used to estimate causal effects from cross-sectional data, we present estimators that exploit the additional information furnished by longitudinal data. Because of the size and technical nature of the literature, we cannot offer a fully detailed and comprehensive presentation. Instead, we present only the main features of methods that are accessible and potentially of use to quantitatively oriented sociologists.

Surface agreements about the social definition of a situation, or what Erving Goffman calls veneers of consensus, are necessary for social interaction to be coherent But why and when do social definitions change? In this article the author examines critical moments as points at which change may potentially take place. The author suggests that change is possible when a breach has occurred - an event, action, statement which is inconsistent with the current social definition. However, change depends on whether individuals ignore the breach, oppose it, or legitimize it. The author introduces the notion of an under-layment the attitudes, that is, the beliefs, knowledge, preferences, and normative commitments individuals have about a particular social situation. He argues that whether a particular veneer of consensus will change in the face of a breach is determined, in part, by the under-layment that supports that veneer.

The collection of large-scale administrative records in electronic form by many cities provides a new opportunity for the measurement and longitudinal tracking of neighborhood characteristics, but one that requires novel methodologies that convert such data into research-relevant measures. The authors illustrate these challenges by developing measures of “broken windows” from Boston’s constituent relationship management (CRM) system (aka 311 hotline). A 16-month archive of the CRM database contains more than 300,000 address-based requests for city services, many of which reference physical incivilities (e.g., graffiti removal). The authors carry out three ecometric analyses, each building on the previous one. Analysis 1 examines the content of the measure, identifying 28 items that constitute two independent constructs, private neglect and public denigration. Analysis 2 assesses the validity of the measure by using investigator-initiated neighborhood audits to examine the “civic response rate” across neighborhoods. Indicators of civic response were then extracted from the CRM database so that measurement adjustments could be automated. These adjustments were calibrated against measures of litter from the objective audits. Analysis 3 examines the reliability of the composite measure of physical disorder at different spatiotemporal windows, finding that census tracts can be measured at two-month intervals and census block groups at six-month intervals. The final measures are highly detailed, can be tracked longitudinally, and are virtually costless. This framework thus provides an example of how new forms of large-scale administrative data can yield ecometric measurement for urban science while illustrating the methodological challenges that must be addressed.