Dr Ian Brunton-Smith

Biography

I joined the Department of Sociology as a full time lecturer in criminology in 2008, having previously completed my PhD research here. Alongside involvement in a range of research projects, I contribute to the substantive teaching in areas of penology and criminal statistics, as well as offering statistical methods training across our programmes. Prior to this, I worked in the Research Development and Statistics division of the Home Office, where I was responsible for exploring the potentials for improving the measurement of crime and public opinions at small spatial scales, the development of new survey weights to improve the estimation of annual crime figures, and also contributed to the annual publication of Crime in England and Wales.

The main areas of research that I am currently involved in are:

NEIGHBOURHOOD EFFECTS AND PUBLIC PERCEPTIONS OF CRIME
I am particularly interested in the potential impact that neighbourhood context has in shaping local residents perceptions of crime. This has involved the application of multilevel models to crime survey data in order to identify the contribution of neighbourhood context, and combining this with contextual information from the census of England and Wales. Key findings from this work have recently been published in Criminology, and two book chapters have further extended this work to discuss the impact of adjustments for spatial autocorrelation. I have also been looking more closely at the nature of the link between fear and perceptions of disorder, drawing on longitudinal data to assess how disorder and fear co-relate (recently published in The British Journal of Criminology). Additional publications are currently in preparation focusing specifically on the experiences of residents of London, and linking perceptions of crime more directly to collective neighbourhood sentiment.

Key findings from this work are described in:
Brunton-Smith, I., and Sturgis, P. (2011) ‘Do Neighborhoods Generate Fear of Crime?: An Empirical Test Using the British Crime Survey’. Criminology, 49(2): 331-369.

Brunton-Smith, I. (2011) ‘Untangling the relationship between fear of crime and perceived disorder: Evidence from a longitudinal study of young people in England and Wales’. British Journal of Criminology, 51(6): 885-899.

THE EFFECT OF PRISON ON REOFFENDING AND EMPLOYMENT 
I am currently exploring the impact of prison experience on reoffending and employment amongst a cohort of nearly 4,000 prisoners using survey data from the Surveying Prisoner Crime Reduction (SPCR) survey linked to the Police National Computer. This includes the application of multilevel models to adjust for prison context, and longitudinal models to examine changes in prisoner experience and attitudes over time. This work is currently ongoing, with a number of publications in preparation.  

SURVEY METHODOLOGY
I have undertaken a range of research within the field of survey methodology. Currently I am exploring the potential contribution that interviewers make to estimates of measurement error in face to face surveys, and whether this is linked to their success at gaining cooperation from households. This is examined with the application of cross-classified multilevel models with a complex error structure to face to face survey data (forthcoming in Public Opinion Quarterly).

I have also been involved in work looking at the potential for interviewer observation data collected during the interview to adjust survey estimates for nonresponse bias, as well as the potential for panel conditioning effects in longitudinal surveys. More recently I have been applying multiple imputation models to survey data with high attrition, including data with a multilevel structure.

Key findings from this work are described in:
Brunton-Smith, I., Sturgis, P., and Williams, J. (online first) ‘Is success on the doorstep correlated with the magnitude of the interviewer design effect?’ Public Opinion Quarterly.

Sturgis, P., Allum, N., and Brunton-Smith, I. (2008) ‘Chapter 7 - Attitudes over Time: The Psychology of Panel Conditioning’ In P. Lynn (ed) Methodology of Longitudinal Surveys, Wiley.

Brunton-Smith, I., Carpenter, J., Kenward, M., and Tarling, R. (2011) ‘Surveying Prisoner Crime Reduction (SPCR) Phase 2: Using Multiple Imputation to Recover Missing Data Across Waves 2, 3 and 4 of the SPCR’. Report for Ministry of Justice.

Previous areas of work that I have been involved in include:

PROCESSES INFLUENCING CONVENTIONAL AND NON-CONVENTIONAL FORMS OF POLITICAL PARTICIPATION
This explored the varied processes which influence democratic ownership and participation across Europe. This used multilevel and structural equation models to examine the influence of macro-level contextual factors, proximal social factors, and psychological factors which serve to facilitate or inhibit engagement with, and participation in, politics. Initial findings from this research have been included in a report and briefing paper supplied to the European Commission as part of an FP7 funded research project.

Key findings from this work are described in:
Brunton-Smith, I. (2011) ‘Process Influencing Democratic Ownership and Participation work package 5: Modelling existing survey data’. Work Package 5 Final Report for the European Commission.

PUBLIC UNDERSTANDING OF SCIENCE
Prior to starting my PhD research I was employed as a research officer on an ESRC funded project examining public attitudes towards science and technology. This work included the completion of a multilevel meta-analysis exploring the association between knowledge about science and attitudes towards science, and a longitudinal examination of the impact of information provision on subsequent attitudes.

Key findings from this work are described in:
Sturgis, P., Brunton-Smith, I., and Fife-Schaw, C. (2010) ‘Public Attitudes to Genomic Science: An Experiment in Information Provision’. Public Understanding of Science, 19(2): 166-180.

Allum, N., Sturgis, P., Tabourazi, D., and Brunton-Smith, I. (2008) ‘Science Knowledge and Attitudes across Cultures: A Meta-analysis’. Public Understanding of Science, 17(1): 35-54.

Research Interests

My research interests are broad, and include criminal statistics; survey methodology; advanced quantitative methods; cross-national data analysis; missing data problems; and public opinion research. 

A list of my funded research: 

Curriculum Innovation: Integrating Quantitative Methods and Substantive Teaching for HE Level One Sociology Students. January 2012 – December 2013, £70,000, Economic and Social Research Council: Co-investigator

I am co-investigator on a ESRC funded project producing new teaching materials to improve the uptake of quantitative methods amongst undergraduate social science students. Specifically, this project emphasises the linkage of substantive teaching with Quantitative Methods teaching during the first year of undergraduate study, and will involve the creation of a new online resource designed to reinforce these links.

Surveying Prisoner Crime Reduction (SPCR) Missing Data Project. October 2010 – March 2011, £59,000, Ministry of Justice: Principal Investigator
SPCR Follow-up Analytic project. November 2011 – May 2012, Ministry of Justice: Principal Investigator

The Surveying Prisoner Crime Reduction (SPCR) programme was a four wave longitudinal survey fielded to nearly 4,000 prisoners across the entire prison estate, with interviews conducted with prisoners twice during their sentence (immediately on entry and prior to release), and again after release. This project involved a detailed examination of attrition across the survey to uncover potential missing data mechanisms. Multiple Imputation (MI) approaches were then outlined to adjust survey estimates for missing data, including multilevel MI procedures enabling the incorporation of the prison structure.

A follow up project implementing the MI procedures to produce descriptive results from the survey is currently in progress. Longitudinal models are also being developed to examine prisoner outcomes on release from prison, and how these are related to experiences in prison and prior offending histories.  

Processes Influencing Democratic Ownership and Participation. May 2009 – April 2012, €1,495,000, European Commission Seventh Framework Programme: Co-ordinator Work Package 5 – Modelling Existing Survey Data

This project examines the processes which influence democratic ownership and participation in eight European states – Belgium, Czech Republic, Germany, Italy, Portugal, Sweden, Turkey and the UK. Drawing on the disciplines of Psychology, Politics, Sociology, Anthropology, Social Policy and Education it examines macro-level contextual factors (including historical, political, electoral, economic and policy factors), proximal social factors (including familial, educational and media factors) and psychological factors (including motivational, cognitive, attitudinal and identity factors) which facilitate and/or inhibit civic and political engagement and participation.

I am the co-ordinator of workpackage 5, which focuses on modelling existing survey data including the European Social Survey, Eurobarometer, International Social Survey Programme and Comparative Study of Electoral Systems. This includes the application of multilevel and structural equation models to better understand the processes influencing participation within and across countries.

Project website: http://www.fahs.surrey.ac.uk/pidop/

An assessment of the utility of interviewer observation variables for nonresponse adjustment in the National Survey for Wales. September 2011 – November 2011, Welsh Assembly Government: Co-investigator

This study provided an assessment of the extent that observation data collected by interviewers during data collection can be useful as a technique to reduce nonresponse bias. Drawing on data from the Understanding Society innovation panel and the ONS census link study, survey estimates with and without nonresponse weighting adjustments were compared, and the impact on response bias was assessed.

The effect of demographic make-up on perceptions of ASB in London. December 2009 – March 2010, £18,783, Government Office for London: Principal Investigator

This project explored variations in perceptions of Anti-Social Behaviour across London Boroughs. Using multilevel models, contextual information about each London borough was linked to survey data from the British Crime Survey and the London Place Survey.

Publications

Teaching

At Surrey I am responsible for the provision of a number of taught modules at both undergraduate and postgraduate levels. This includes training in statistical modelling, as well as substantive modules in Criminology.

Current teaching
Crime and Offending (Msc, Criminology, Criminal Justice and Social Research)
Statistical Modelling (Msc, Social Research Methods)
Prisons and Prisoners (3rd Year undergraduate, Criminology)
Historical and Theoretical Perspectives on Punishment (2nd Year undergraduate, Criminology)  
Crime and Society (1st Year undergraduate, Criminology and Law)

Previous Teaching
Quantitative Methods (1st Year undergraduate, Sociology, Criminology, and Media)
Quantitative Methods (2nd Year undergraduate, Sociology, Criminology, and Media)

External Teaching
In addition to my teaching at Surrey, I also run a number of external courses in advanced statistical analysis. This includes courses in multilevel modelling and missing data analysis.

Courses in Applied Social Surveys – ‘Applied multilevel modelling’, University of Southampton, January 25-27, 2011
European Survey Research Association pre-conference short course – ‘Multilevel models for survey data’, Lausanne, July 18, 2011
Ministry of Justice training – ‘Multiple Imputation to recover missing data in longitudinal surveys’, Ministry of Justice, May 4, 2011
Survey Skills Workshop (part of ESRC Survey Resources Network project). City University, London, January 6, 2011
Courses in Applied Social Surveys – ‘Applied multilevel modelling’, University of Southampton, June 16-18, 2010
MSc in Official Statistics – ‘Multilevel modelling’, University of Southampton, March 15-19, 2010
Home Office training – ‘An introduction to multilevel modelling in SPSS’, Research Development and Statistics, Home Office, October 2009

Conferences

Invited talks
Brunton-Smith, I. (2011) ‘Neighbourhoods Matter: Spillover Effects in the Fear of Crime’ ESRC Seminar Series on Neighbourhood Effects, Neighbourhood Based Problems and International Policy Solutions at Glasgow University April 7-8, 2011

Brunton-Smith, I. (2011) ‘Do Neighbourhoods Generate Fear of Crime?’, Royal Statistical Society Local Group, Lancaster, January 20, 2011

Brunton-Smith, I. (2009) ‘What is a Country? Accounting for Compositional Differences between Groups in Assessments of Cross-national Measurement Equivalence’, International Workshop on Cross-national Data Analysis QMSS2 Seminar at Bolzano-Bozen, June 11-12, 2009

Brunton-Smith, I. (2008) ‘The use of ACORN Classifications and the Index of Deprivation in British Crime Survey Analysis’, ONS Area Classification User Group meeting, Royal Statistical Society, September 15, 2008

Brunton-Smith, I. (2006) ‘Science Knowledge and Attitudes across Cultures: A Meta-analysis’ Interdisciplinary Workshop on Meta-analysis, Cologne, October 26-27, 2006

Other conference contributions
Brunton-Smith, I. (2011) ‘The Impact of Interviewers on Survey Error’ Session convenor for the 4th Conference of the European Survey Research Association, Lausanne, July 18-22, 2011

Brunton-Smith, I. (2010) ‘Untangling the Relationship between Fear of Crime and Perceptions of Disorder: Evidence from a Longitudinal Study of Young People in England and Wales’, Paper prepared for the Crime Surveys User Group meeting, December 7, 2010

Brunton-Smith, I. (2010) ‘Is Success on the Doorstep Correlated with the Magnitude of the Interviewer Design Effect?’ Paper prepared for the American Association of Public Opinion Research Annual Conference, Chicago, May 14-17, 2010

Brunton-Smith, I. (2009) ‘Assessing the Relative Influence of Interviewer and Area Differences on Reported Levels of Fear of Crime: A Cross-Classified Multilevel Analysis of the British Crime Survey’, Paper prepared for the Crime Surveys User Group meeting, December 7, 2009

Brunton-Smith, I. (2009) ‘A New Approach to Separating Interviewer from Area Variability in Face-to-Face Interview Surveys’, Paper prepared for the American Association of Public Opinion Research Annual Conference, Florida, May 14-17, 2009

Brunton-Smith, I. (2008) ‘The Relative Influence of Interviewers and Areas on Fear of Crime: A Cross-Classified Multilevel analysis of the British Crime Survey’ Paper prepared for the RC33 7th Annual Conference on Social Science Methodology, Naples,  September 1-5, 2008

Brunton-Smith, I. (2007) ‘New and Old Measures of the Fear of Crime: A Multilevel Assessment of Measures of Intensity and Frequency’ Paper prepared for the British Society of Criminology Annual Conference, London School of Economics, September 18-10, 2007

Brunton-Smith, I. (2007) ‘New and Old Measures of the Fear of Crime: A Multilevel Assessment of Measures of Intensity and Frequency’ Paper prepared for the 2nd Conference of the European Survey Research Association, Prague, June 25-29, 2007