Individual Confidence Intervals Do Not Inform Decision-Makers About the Accuracy of Risk Assessment Evaluations

Some recent articles have proposed that the confidence interval for the predicted outcome of a single case can be used to describe the predictive accuracy of risk assessments (Hart et al. Br J Psychiat 190:60–65, 2007b; Cooke and Michie, Law Hum Behav 2009). Given that the confidence intervals for an individual prediction are very large, Cooke and colleagues have questioned the wisdom of applying recidivism rates estimated from group data to single cases. In this article, we argue that the confidence intervals for the recidivism outcome predicted for a single case will range between zero to one (i.e., be uninformative) when the outcome is dichotomous and the predicted probability is between .05 and .95. This is true by definition and limits the utility of using individual confidence intervals to measure predictive accuracy. Consequently, other quality indicators (many of which are non-quantitative) are needed to determine the accuracy and error of risk evaluations.     

Proactive Policing: a Summary of the Report of the National Academies of Sciences,Engineering, and Medicine

This paper provides a summary of our report for the National Academy of Sciences, Engineering, and Medicine on proactive policing. We find that there is sufficient scientific evidence to support the adoption of many proactive policing practices if the primary goal is to reduce crime, though the evidence base generally does not provide long-term or jurisdictional estimates. In turn, we conclude that crime prevention outcomes can often be obtained without producing negative community reactions. However, the most effective proactive policing strategies do not appear to have strong positive impacts on citizen perceptions of the police. At the same time, some community-based strategies have begun to show evidence of improving the relations between the police and public. We conclude that there are likely to be large racial disparities in the volume and nature of police–citizen encounters when police target high-risk people or high-risk places, as is common in many proactive policing programs. We could not conclude whether such disparities are due to statistical prediction, racial animus, implicit bias, or other causes.

     

Singing the same tune? International continuities and discontinuities in how police talk about using force

This article focuses on a research project conducted in six jurisdictions: England, The Netherlands, Germany, Australia, Venezuela, and Brazil. These societies are very different ethnically, socially, politically, economically, historically and have wildly different levels of crime. Their policing arrangements also differ significantly: how they are organised; how their officers are equipped and trained; what routine operating procedures they employ; whether they are armed; and much else besides. Most relevant for this research, they represent policing systems with wildly different levels of police shootings, Police in the two Latin American countries represented here have a justified reputation for the frequency with which they shoot people, whereas at the other extreme the police in England do not routinely carry firearms and rarely shoot anyone. To probe whether these differences are reflected in the way that officers talk about the use of force, police officers in these different jurisdictions were invited to discuss in focus groups a scenario in which police are thwarted in their attempt to arrest two youths (one of whom is a known local criminal) by the youths driving off with the police in pursuit, and concludes with the youths crashing their car and escaping in apparent possession of a gun, It might be expected that focus groups would prove starkly different, and indeed they were, but not in the way that might be expected. There was little difference in affirmation of normative and legal standards regarding the use of force. It was in how officers in different jurisdictions envisaged the circumstances in which the scenario took place that led Latin American officers to anticipate that they would shoot the suspects, whereas officers in the other jurisdictions had little expectation that they would open fire in the conditions as they imagined them to be.

---

What's So Magical about Five Percent? A Nationwide Look at Factors That Influence the Optimal Size of State Rainy Day Funds

State rainy day funds have increased in popularity as countercyclical planning devices over the past 15 years. The view is widely held that all states need a rainy day fund balance of five percent in order to guard themselves against the threat of budgetary dislocation. This article compares the actual balances in state rainy day funds in 1997 to several factors affecting budgetary volatility. Little relationship is found between rainy day fund balances and the actual level of volatility in a given state. The article finds no justification for a "one size fits all" approach; each state should design policies based on its own peculiar needs.     

A Framework for Analyzing Emergency Management with an Application to Federal Budgeting

Emergency management is a complex policy subsystem that involves an intergovernmental, multiphased effort to mitigate, prepare for, respond to, and recover from disasters. This article develops a framework for analyzing the fiscal and functional aspects of disaster policy. It uses established theories of intergovernmental relations to offer a rationale for examining the capabilities required to implement disaster policy and the behavioral incentives that drive policy formulation. In particular, the article identifies the extent to which the capabilities and political objectives characteristic of each level of government are aligned, and illustrates the interplay between incentives and competencies by reviewing the federal disaster funding process. The current rules for federal budgeting may inappropriately promote spending on disaster response and recovery, while de-emphasizing mitigation and preparedness. Various proposals for reform could establish more coherent incentives, making disaster spending more consistent with the relative functional capabilities of the various levels of government.     

Populist possibilities and political constraints in Mexico

This article defines populism (in Latin American context) as a form of politics that is based mainly upon a personalist appeal, attracts significant popular support and, if successful, significantly redesigns social or political institutions. Neopopulism is defined as a form of populism which begins outside the state as opposed to classic populism which originates within the state. The main theoretical argument is that the study of populism can be made significant for comparative politics if we emphasise the issue of institutional redesign and then conceptualise it in structure vs agency terms. The comparative question then becomes whether a particular set of political institutions is robust enough to withstand the challenge of a charismatic individual. The empirical background is that Latin America has seen a significant number of populist movements that have indeed fundamentally altered institutional structures. By the same token, there have been an even larger number of populist attempts which ultimately failed. Having made this point, the discussion then focuses on Mexico. It seeks to understand the different sets of conditions which allowed Lázaro Cárdenas to succeed as classic populist in the 1930s, Cuauhtémoc Cárdenas to come close to success as a neopopulist in the 1980s and which greatly reduce the likelihood of successful neopopulism in the near future.