November 2014 RSS Journal Club

News

  • Author: Statistics Views
  • Date: 28 November 2014

The most recent Royal Statistical Society Journal Club took place last week.

Journal Club is an online webinar where authors in the Society's prestigious journals can present their work to a potentially worldwide audience via a phone-in webinar.

The event featured two papers, both looking at methods used to model criminal behaviour. The authors introduced their papers and took questions from participants afterwards. The meeting was chaired by Professor Chris Skinner, professor of statistics at the London School of Economics.

Both articles are free to read on Wiley Online Library until 4 December. To read the papers, click on the links below.

thumbnail image: November 2014 RSS Journal Club

The following papers were presented by their respective authors:

1. ‘The item count method for sensitive survey questions: Modelling criminal behaviour’ by Jouni Kuha and Jonathan Jackson, published in the Journal of the Royal Statistical Society: Series C (Applied Statistics), Vol 63 (2), 321-341, February 2014.

The item count method is a way of asking sensitive survey questions which protects the anonymity of the respondents by randomization before the interview. It can be used to estimate the probability of sensitive behaviour and to model how it depends on explanatory variables. The results of the author’s analysis of criminal behaviour highlight the fact that careful design of the questions is crucial for the success of the item count method.

2. 'Which method predicts recidivism best? A comparison of statistical, machine learning and data mining prediction models' by Nikolaj Tollenaar and Peter van der Heijden, published in the Journal of the Royal Statistical Society: Series A (Statistics in Society), Vol 176 (2), 565–584, February 2013.

Risk assessment instruments are widely used in criminal justice settings all over the world. However, in recent times, different approaches to prediction have been developed. This paper investigates whether modern techniques in data mining and machine learning provide an improvement in predictive performance over classical statistical methods such as logistic regression and linear discriminant analysis. Using data from criminal conviction histories of offenders, these models are compared. Results indicate that in these data, classical methods tend to do equally well as or better than their modern counterparts.

For more information on the RSS Journal Club, visit the Society's Journal Club webpage.

Related Topics

Related Publications

Related Content

Site Footer

Address:

This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on StatisticsViews.com are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and StatisticsViews.com express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.