Public confidence in the police: A time-series analysis

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. Our aim here is to introduce a time dimension to this area of investigation. Employing repeated cross-sectional survey data from the British Crime Survey, we apply time-series regression methods to show how confidence in policing changes over time for the aggregate population. Counter to cross-sectional findings, time-series analyses reveal that confidence in the police is not related to aggregate worry about crime and perceptions of social cohesion, nor informal social control, but only to perceptions of crime and the property crime rate.