Abstract
This paper presents the results of a survey evaluating the views of 1322 residents of York, UK around their perceptions and attitudes towards crime, their local area, and the police with the aim of understanding the main factors affecting public confidence within the city. The study uses Structural Equation Modelling to develop several models which are evaluated to examine whether differences in public confidence in the police exist across disparate local communities with varying degrees of neighbourhood perceptions. The results indicate that even in a relatively small city such as York, the factors which most affect an individual’s views of the police can vary wildly depending on an individual’s perceptions regarding their local area. The results suggest that policing strategies aimed at improving public confidence must be altered depending on the views residents hold regarding their local communities.
Acknowledgements
The author is grateful for the support of the North Yorkshire Police who provided funding for this study.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1. The likely cause of this high response rate was a survey incentive of two, £50 Amazon voucher available in a prize-draw to all fully completed survey responses.
2. In order to ensure that the results of the analysis were representative of the population of York, weighting variable were created for age, gender, ethnicity, and ward location using the most recently available figures. A total weighting value was created by multiplying these individual weights together, which had the effect of altering significance levels by altering the sample size. In order to counter this, a relative total weight was created by dividing the total weight by its mean. This technique retained the distribution patterns of the weighted sample, but maintained the original sample size. This relative total weight was applied to the data-set before further analysis was carried out.
3. ***shows the path estimate is significant at the P ≥ .001 level.
4. CMIN: 2492.53, DF: 641, CMIN/DF: 3.89, PNFI: .854, P Value Sig.: 0.00, CFI: .952. NFI:.936, RMSEA: .047.
5. A moderating effect occurs when a third variable (in this case, perceptions of local area) changes the relationship between two related factors in an SEM model (Baron & Kenny, Citation1986).
6. Because we are only interested in the equivalence of regression weights between latent factors, we do not test for either factorial equivalence (Jöreskog, Citation1971) or latent mean structure equivalence (Byrne & Stewart, Citation2006).
62% of HPWs residents had high levels of overall confidence compared to 64% of PPWs (mean item values for all ‘strongly/agree’ responses in the PCON factor).
7. 62% of HPWs residents had high levels of overall confidence compared to 64% of PPWs (mean item values for all ‘strongly/agree’ responses in the PCON factor).
8. 99% of HPWs residents agreed that they felt safe in their local area compared to 79% of PPWs residents (mean item values for all ‘strongly/agree’ responses in the LASAFE factor.
9. 87% of HPWs residents agreed that their local area exhibited cohesive qualities compared to 46% of PWs residents (mean item values for all ‘strongly/agree’ responses in the LAC factor).
10. 66% of PPWs residents agreed they were not concerned about specific crimes occurring in their area, compared to 90% of HPWs residents (mean item values for all ‘strongly/agree’ responses in the FOC factor).
11. 43% of PPWs residents agreed that the police were effective in dealing with the problems in York, compared to 54% of HPWs residents (mean item values for all ‘strongly/agree’ responses in the PDEAL factor).
12. 42% of PPWs residents agreed that the police were interacting with their communities, compared to 54% of HPWs residents (mean item values for all ‘strongly/agree’ responses in the PCOM factor).
13. 99% of HPWs residents agreed that they felt safe in their local area compared to 79% of PPWs residents (mean item values for all ‘strongly/agree’ responses in the LASAFE factor.