About the research
Criminologists and policymakers have been seeking to understand the sources of public confidence in policing for some time. It is argued that if citizens do not have confidence in the police, they are less likely to defer to police authority, report crimes, provide witness information or obey the law themselves .
Previous empirical analyses of the causes of public confidence in policing have been based almost entirely on cross-sectional survey data, with a consequent focus on between-group differences in levels of confidence at a single point in time. This study draws on the British Crime Survey to examine how confidence in policing changes over time for the population of England and Wales as a whole.
Counter to cross-sectional findings, the time-series regression analyses conducted in this study revealed that the level of confidence in the police is not related to monthly fluctuations in the public’s worry about crime and perceptions of social cohesion, nor to perceptions of informal social control, but only to perceptions of crime and the property crime rate. In addition, victimisation rates and perceptions of crime rates are found to be predictive of monthly levels of public confidence in the police. On the other hand, worries about crime and perceptions of disorder, social cohesion and informal social control are found to be statistically unrelated to the level of confidence in the police over time.
In short, this study illustrates that when the public believe that crime is increasing and when rates of property crime rise, public confidence in policing is lower.
Methodology
The analysis is based on applying time-series regression models to aggregate trends in repeated cross-sectional survey data. This is a new approach to the analysis of public confidence in policing, using a method associated most closely with macroeconomics, but which is increasingly being applied by analysts in other social science disciplines, as time-series of sufficient length become increasingly available. Time-series regression methods is an approach that has not, to date, been applied to public opinion data about the police or the criminal justice system.
As the fieldwork for the British Crime Survey is carried out on a continuous basis, it is possible to take monthly observations for each variable of interest. Using survey data collected between April 2001 and March 2008, population aggregate means were taken at each month to create a time series of a number of variables regarding public confidence in the police, perceptions of crime and disorder, worry about crime, perceptions of social cohesion and informal social control and victimisation. Time-series regression analysis was employed to assess the dynamic relationships between these variables.
Publications
This research was featured in the following academic journal:
Sindall, K., Sturgis, P. and Jennings, W. (2012) ‘Public confidence in the police: a time-series analysis’, British Journal of Criminology 52(4), pp. 744-764. doi:10.1093/bjc/azs010 Retrieved 2 September 2013 from http://bjc.oxfordjournals.org/content/52/4/744.full.pdf+html