Search for a command to run...
Abstract Public opinions of the police have been a fixture at the top of the policy agenda in England and Wales in recent years, with successive governments stating they wish to see improvements in 'trust and confidence'. But significant doubts remain as to how this might be done, and even if it is possible for police to enhance public confidence in any straightforward way. Indeed, it often seems that it is much easier for police to damage public opinion than to improve it. This paper reports findings from two surveys on contact between the public and the police conducted in England and Wales. First, panel data are used to examine the issue of 'asymmetry' in the relationship between satisfaction with police contacts and wider public confidence in the police. Negative pre-existing opinions of the police are found to be predictive of negatively received contact, while positive views do not predict well-received contact. Yet, single contacts, both negative and positive, are predictive of subsequent confidence in the police. Second, British Crime Survey data are used to investigate what 'drives' satisfaction among crime victims. Personal treatment appears to be valued over criminal justice outcomes, providing support for process-based policing models. It appears that fears about an absolute asymmetry in the effect of contact on confidence may be overstated, and that improving the way officers handle encounters might lead to enhanced trust and confidence. Keywords: public attitudespoliceprocedural justicevictims Acknowledgements The authors would like to thank Katrin Hohl of the London School of Economics for valuable advice on data analysis. Notes 1. Because interviewing the same people at two points in time allows pre-existing attitudes to be taken into account, panel designs provide a stronger methodology for assessing the causal direction of associations between contact and confidence. 2. Calculation of response rates was complicated by the need to ensure respondents were from very specific geographical areas. Response rates were higher for contacts known to be in the required areas (see BMRB Social Research Citation2005). 3. Respondents were asked if they had contacted the police in the last year, by telephone, letter, email, or in person, for any of the following: To report any type of disturbance, noise, or nuisance (including car/house alarms); to report any other suspicious circumstances or persons; to report a missing person or lost or found property; to report a traffic accident or other emergency; to provide other information; to ask for information or advice; or for 'any other reason'. 4. Full methodological information is available at: http://webarchive.nationalarchives.gov.uk/20110218135832/http://rds.homeoffice.gov.uk/rds/bcs-methodological.html. 5. Out of scope incidents include offences that are subsequently found to have occurred outside of England and Wales. 6. Though response times may not always be seen as reflecting police treatment, we feel that the time spent waiting forms part of the overall process of police response. Qualitative work has suggested that at least some citizens with negative perceptions of the police regard slow perceived response times as a function of the inferior treatment they feel they are allotted by the police (see for example Communities and Local Government Citation2008). 7. We again decided against an ordinal logistic model as Brant's test of the proportional odds assumption suggested a significant breach of this assumption. 8. To check the results shown we estimated a second model in which the six possible combinations of police interest and outcome were represented by five separate dummy variables. The results of this model (available from the lead author) were entirely consistent with those presented here. 9. There is evidence that officers value the ability to handle encounters well, but not necessarily in the sense of acting in the interests of the citizen. Quinton et al. (Citation2000) found, in relation to foot stops, that officers were primarily concerned with maintaining order in encounters: procedurally fair treatment acted as a mechanism to ensure that officers did not 'lose face' by allowing an encounter to get out of their control.