Published by De Gruyter

Volume 11 Issue 2

Issue of Statistics, Politics and Policy

Submit manuscript

Contents
Journal Overview

Publicly Available December 4, 2020

Frontmatter

Page range: i-iii

Download PDF

Articles

Open Access September 30, 2020

Healthcare Expenditure Prediction with Neighbourhood Variables – A Random Forest Model

Sigrid M. Mohnen, Adriënne H. Rotteveel, Gerda Doornbos, Johan J. Polder

Page range: 111-138

More Download PDF

Abstract

We investigated the additional predictive value of an individual’s neighbourhood (quality and location), and of changes therein on his/her healthcare costs. To this end, we combined several Dutch nationwide data sources from 2003 to 2014, and selected inhabitants who moved in 2010. We used random forest models to predict the area under the curve of the regular healthcare costs of individuals in the years 2011–2014. In our analyses, the quality of the neighbourhood before the move appeared to be quite important in predicting healthcare costs (i.e. importance rank 11 out of 126 socio-demographic and neighbourhood variables; rank 73 out of 261 in the full model with prior expenditure and medication). The predictive performance of the models was evaluated in terms of R 2 (or proportion of explained variance) and MAE (mean absolute (prediction) error). The model containing only socio-demographic information improved marginally when neighbourhood was added ( R 2 +0.8%, MAE −€5). The full model remained the same for the study population ( R 2 = 48.8%, MAE of €1556) and for subpopulations. These results indicate that only in prediction models in which prior expenditure and utilization cannot or ought not to be used neighbourhood might be an interesting source of information to improve predictive performance.

Open Access August 24, 2020

Southeast Asia after the Caliphate: Identifying Spatial Trends in Terrorism and Radicalization in Malaysia

Samuel Henkin, Marcus A. Boyd, Adam Martin

Page range: 139-165

More Download PDF

Abstract

Recent scholarship on militant Islamist radicalization in Southeast Asia highlights the significance of local context in understanding support and recruitment into militancy. While research on terrorism in Southeast Asia engenders a dynamism of epistemic inquiry in diverse areas of research, an empirical mapping of radicalization is generally absent. Research on militant Islamic radicalization in Southeast Asia needs more robust consideration of geospatial relations and data to fill this lacuna. A geospatial analysis is, above all else, synthesis. It bridges spatial statistical analytics and qualitative socio-spatial investigation. We argue that a geospatial analytical approach to understanding radicalization offers a way in which to begin empirically mapping radicalization in the region. Correspondingly, our work considers militant Islamic radicalization in Malaysia by employing geospatial analysis to build a more nuanced layering and comprehensive understanding of the spatial arrangements of radicalization. Viewed from this perspective, radicalization can be understood through spatial logics and practices aimed at facilitating better understandings of socio-political relations of political violence and terrorism more broadly.

October 9, 2020

A Regression Model for Predicting the Likelihood of Reporting a Crime Based on the Victim’s Demographic Variables and Their Perceptions Towards the Police

Malebogo Pulenyane, Tlhalitshi Volition Montshiwa

Page range: 167-193

Abstract

Despite the growing criminal activities in South Africa, many victims still do not report the crimes, therefore there was a need to understand the determinants of the likelihood of reporting a crime in the country. Binary logistic regression is a supervised machine learning algorithm that can assist in predicting the likelihood of reporting a crime but the selection of relevant variables to add in the model varies from one author to the other. Selection of theoretically sound and statistically relevant independent variables is key to achieving parsimonious multivariate models. This study sought to test the efficiency of some commonly used variable selection methods for logistic regression models in order to identify the most relevant determinants of the likelihood of reporting a crime of housebreaking. The study used 17 candidate variables such as the victims’ demographic variables and their perceptions on the police. The multivariate model fitted using stepwise selection was found to be a best fit for the data based on the lowest AIC, the highest classification accuracy rate and the highest Area under the Receiver Operating Characteristic curve. The model fitted using the Hosmer-Lemeshow (H-L) algorithm was the worst fit for the data. The study revealed a limitation of the stepwise selection method which is that this method may select different independent variables for each unique set of randomly selected observations of the same dataset. The study established a multivariate logistic regression model to predict the likelihood of a victim reporting a crime of housebreaking and the determinants thereof.

Publicly Available July 20, 2020

Authoritarianism, Prejudice, and Support for Welfare Chauvinism in the United States

Jason Kehrberg

Page range: 195-212

More Download PDF

Abstract

This article examines the role of authoritarianism in shaping individual policy support for creating a waiting period before legal immigrants can access welfare programs. I use logistic regression models to analyse the opinions of non-Latino whites from the 1992 American National Election Study (ANES). I find that authoritarianism is related to a waiting period before immigrants can access welfare benefits, despite a significant number of non-authoritarians also preferring a waiting period. This effect is robust with the inclusion of several traditional predictors of welfare attitudes.

About this journal

Objective
Statistics, Politics, and Policy studies the ways that statistical analysis drives public policy decisions, and publishes significant research on the application of statistical ideas to problems that relate to policy implementation.

The increasing amount and complexity of available data is constantly creating new challenges for statistical thinking in policy problems. While many academic statisticians tend to share among themselves their latest methods and models, less attention has been paid to the usefulness of those statistical methods and models to inform public policy decisions, and what statistical approaches might be most effective in designing how policies are implemented. In the policy sphere, statistical methods are sometimes taken as a given, with less attention to all the variations, assumptions, and effects of different methods in differing contexts. But it is in the policy sphere that statistical debates can have the great value and impact, and the intersection of statistics and public policy is a fertile ground for statistical research and analysis to address important policy issues that may have widespread ramifications.

As an electronic journal, Statistics, Politics, and Policy will use a mix of voices and approaches to reach a broad audience. The journal aims to open avenues of communication between statisticians and policy makers on questions that pique the interest of the public. The primary objective of the journal will be to highlight the use of innovative statistical methodology in order to elucidate and resolve important public policy issues.

Statistics, Politics, and Policy will publish applied research articles that explore the implications of statistical thinking and methods applied to public policy issues. The journal will also publish engaging commentary pieces and innovative policy ideas on the public issues of the day where statistics plays, or ought to play, a role.

Papers for the commentary and ideas section are short, ideally 600-2000 words, and are intended to be of general interest and readability. That is, they should contain deeper analysis than is found on the Op-Ed page of The Wall Street Journal or The New York Times, but to be of comparable interest the Statistics, Politics, and Policy readership.

Topics

Applied statistical research
Policy issues
Educational testing and policy
Energy and environmental policy
Demography
Climate change
Public finance
History and review of statistical ideas applied to public policy controversies
Taxation and business policy
Justice, crime and forensic analyses
Health policy including health care finance
Drug development
Approval and monitoring
Politics
Sociological policy analyses
Statistical methodology including study design and causal inference, and survey methods

Article formats
Research articles, commentaries