POLICE PRESENCE and PUBLIC CONFIDENCE in LOCAL POLICING: an Analysis of the British Crime Survey

POLICE PRESENCE AND PUBLIC CONFIDENCE IN LOCAL POLICING: An Analysis of the British Crime Survey

Dr. Helen Innes & Prof. Martin Innes

Universities’ Police Science Institute

Cardiff University

March 2011

EXECUTIVE SUMMARY

This report examines the relationship between police presence and public confidence in the police. The concept of police presence is constructed along three key dimensions: ‘local police visibility’ measured by the number of times police are observed on patrol; ‘familiarity’ assessed on the basis of whether an officer is known by name or sight; and levels of ‘contact’ with the police

Analysis of the British Crime Survey identifies an association between police visibility and confidence. The more people see the police the more confidence they have. This is a stepped association. For every additional sighting there is an increase in reported confidence.

Visibility is more important than familiarity (measured by knowing police by name or sight). Familiarity does increase confidence, but not as strongly as visibility.

Actual contact with the police (either police or public initiated) decreases confidence.

Trend analysis identified that over recent years there has been a modest but distinct increase in levels of public confidence in the police. This is in part a reflection of a significant increase in the numbers of people reporting seeing police on patrol in the local area. This is probably attributable to the roll-out of Neighbourhood Policing. The figure below summarises the trend data for male respondents to the British Crime Survey.

Introduction

This report examines the relationship between ‘police presence’ and public confidence in policing. It is based upon available British Crime Survey data for 2006/7 through to 2008/9. Police presence is an umbrella concept capturing the relative prominence of local police in day-to-day life, as perceived by members of the public. The concept has three key dimensions that are constructed as follows using the available survey data:

POLICE PRESENCE CONCEPT

1. Visibility – ‘how often do you see police officers or PCSOs on foot patrol in the local area?’ Responses to this question distinguish the frequency of sightings, ranging from every day, once a day, once a week, once a month or never.
2. Familiarity – ‘do you know any police officers or PCSOs?’ Responses to this question distinguish different levels of familiarity, whether a police officer of PCSO is known by name and sight, by name only, by sight only or by neither.
3. Contact – if the respondent has contacted the police in the last 12 months (yes/no) OR if police contacted respondent in last year (yes/no). These questions allow respondent-initiated contact to be distinguished from police-initiated contact and for investigation of any combination of these contacts.

Police presence therefore incorporates different modes of interaction that might be had at any one time between members of the public and the police, as well as gradations in how visible the police are, and the extent to which the public feel a more interpersonal connection with a representative of the police force. The analysis examines the domains of familiarity, visibility and contact singly and then in combination in order to ascertain whether one has a particular value in understanding public confidence in local policing.

Specifically, the analysis in this report addresses the following questions:

1. What are the key demographic and social patterns in relation to police presence and confidence?
2. To what extent do the different dimensions of police presence account for changes in levels of public confidence in the police, compared with other potentially salient factors?
3. Have trends in the relationship between police presence and public confidence shifted at all over the past five years?

1. What are the key demographic and social patterns in relation to police presence and public confidence?

This section begins by investigating how perceptions of police presence are variable for different population sub-groups. Using the most recently available BCS data for 2008/9, each domain of police presence is analysed in turn by the key demographic indicators of age, gender, social class and ethnic group. (Further details about the data in this section are noted in the appendix).

Familiarity

Table 1 shows that nearly three-quarters of all adults do not report any familiarity with a Police Officer or PCSO, that is, they feel unable to recognise by sight or recall a name of any police worker in their local area. Zero familiarity is greatest in the oldest age group, especially for women where approximately 8 out of 10 aged 65+ report no familiarity with the police. Young men aged 16 to 24 are least likely to report having no familiarity with the police (67 percent) whereas for women it is those in the middle age groups (35-44 and 45-54). Overall, this most ‘personal’ dimension of police presence shows high levels of unfamiliarity with the police across the age spectrum.

Table 1: Percentage of respondents who report no familiarity with the police by age and gender

16-24 / 24-34 / 35-44 / 45-54 / 55-64 / 65+ / All
Men / 67 / 76 / 74 / 68 / 71 / 76 / 72
Women / 73 / 75 / 67 / 65 / 75 / 79 / 72
All / 70 / 75 / 70 / 66 / 73 / 78 / 72

Source: British Crime Survey, 2008/9

Figures 1 and 2 examine variations among those men and women who did report familiarity with the police. Young men are by far the most likely group to report maximum familiarity with the police – approximately 2 in 10 report knowing someone by name and sight. This level of familiarity then halves for men in the 23-35 age group before reaching a lower peak of 16 percent for men aged 55-64 years. After aged 45, there is a sharp increase in the percentage of men who know a police officer by sight, but the percentage that have familiarity by sight or by sight and name drops markedly for men in the oldest age group (65+). Reports of familiarity by name are much lower for all age groups at only 2 percent for younger adults, increasing to 5 percent for the middle age group but once again decreasing in older age.

Figure 1: Police familiarity for men by age

Source: British Crime Survey, 2008/9

For women, maximum familiarity with the police is not among the young but for those aged 35-44 and 45-54 at 16 percent. After age 54, the percentage of women who know a police officer by sight and name halves to only 8 percent. Familiarity by sight only increases for women after age 34 and then falls to approximately 11 percent for women after the age of 55. However, unlike the other indicators of familiarity, it does not decrease any further for women in the oldest age group. Familiarity by name follows a similar trajectory for women as for men, with a slight peak at age 45-54 and low levels among those aged 65 and above.

Figure 2: Police familiarity for women by age

Source: British Crime Survey, 2008/9

Figures 3 and 4 show that police familarity varies markedly with ethnic group. The vast majority of men and women who are Indian or Pakistani/ Bangladeshi have zero familarity with the police. This equates to approximately 9 out of 10 Pakistani and Bangladeshi women. White men and women are much more likely than any minority ethnic group to report knowing a police officer by name and sight. Black African Caribbean men and women are more likely than other ethnic groups to report familarity by sight only, whereas Pakistani/ Bangladeshi women are more likely to know a police officer by name only.

Figure 3: Police familiarity by ethnic group for men only

Source: British Crime Survey, 2008/9

Figure 4: Police familiarity by ethnic group for women only

Source: British Crime Survey, 2008/9

Visibility

Having so far seen that peoples’ sense of ‘knowing’ a police officer is generally very low, particularly in old age and for Asian ethnic groups, we now examine levels of police visibility among the same population groups. This measure focuses on the frequency of sightings of police on foot patrol, in other words ‘doing policing’ on local streets.

Table 2 shows men and women who reported that they ‘never’ see police on foot patrol, for whom the police have no visibility. The first point to note is that the percentages for zero visibility are much lower than for zero familiarity, meaning that people are overall much more likely to report ‘seeing’ a police officer than ‘knowing’ one. The second point is that there is an age related increase in the percentage reporting no visibility. One quarter of adults aged 16-24 report ‘never’ seeing an officer on foot patrol and this increases to 31 percent by age 45 to 54 and is greatest at 38 percent for adults aged 65 and above. This lack of visibility for the oldest age group is most marked for women; where approximately 4 out of 10 report no sightings. By contrast, at the other end of the age spectrum, visibility is high among young men as only 2 out of 10 aged between 16 and 24 report never seeing a police officer on foot patrol. This pattern is consistent with what is known about the available street population and use of public spaces.

Table 2: Percentage of respondents reporting no police visibility

No Visibility / 16-24 / 24-34 / 35-44 / 45-54 / 55-64 / 65+ / All
Men / 20 / 26 / 27 / 31 / 36 / 34 / 29
Women / 29 / 28 / 29 / 31 / 38 / 41 / 33
All / 25 / 27 / 28 / 31 / 37 / 38 / 31

Source: British Crime Survey, 2008/9

Young men (16-24 years) are most likely to report seeing police on foot patrol and Figure 5 shows that 15 percent of these sightings are ‘at least once a day’ and over one-third are ‘once a week’. High frequency sightings fall markedly for men with increasing age, with reports of ‘less than monthly’ sightings increasing through middle age groups. Interestingly, despite low police familiarity at age 65+, there is a slight rise in police visibility at this age for high frequency or monthly sightings, although the majority see the police ‘less than monthly’.

Figure 5: Frequency of police visibility for men by age

Source: British Crime Survey, 2008/9

For women, it is the same picture of high police visibility in the youngest age group, with over a quarter of 16 to 24 year olds seeing the police on foot patrol ‘once a week’. However, sightings at high frequency decline markedly with older age, such that only 15 percent of older women see police ‘once a week’ and 6 percent ‘at least once a day’. Conversely, the percentage of women sighting police ‘less than monthly’ rises steeply from 15 percent at age 16-24 to 25 percent by 45 to 54.

Figure 6: Frequency of police visibility for women by age

Source: British Crime Survey, 2008/9

The profile of police sightings differs markedly by ethnic group and presents a different picture to that seen using an indicator of police familiarity (Figures 3 and 4).

Figure 7: Police visibility by ethnic group for men

Source: British Crime Survey, 2008/9

Whereas minority ethnic groups – particularly the Asian groups – are much more likely to have zero familiarity with the police, the percentage reporting high frequency sightings of police on patrol is greater among non-white groups. Approximately 3 in 10 Black African Caribbean men and women report seeing police on patrol ‘at least once a day’ compared with 1 in 10 white men and women. When high frequency sightings of ‘at least once a day’ and ‘once a week’ are taken together, police visibility is at 60 percent for Black African Caribbean and Pakistani/Bangladeshi men and 50 percent for Indian men. The same pattern is found for women, with visibility at 50 percent for Black African Caribbean women, 40

Figure 8: Police visibility by ethnic group for women

Source: British Crime Survey, 2008/9

percent for Pakistani/ Bangladeshi women and 35 percent for Indian women. The contrast between familiarity and visibility is particularly striking for Pakistani/Bangladeshi women because 91 percent reported ‘not knowing’ any police offer by either sight or name, yet only 11 percent report ‘never’ seeing a police officer on foot patrol. This example clearly illustrates that zero familiarity with the police does not equate with zero visibility.

Figure 9 shows that police visibility was also linked in a consistent way with occupational social class (NSSEC), based on the working conditions of current or last main job. For both sexes, the percentage reporting zero police visibility – never seeing the police on foot patrol – is greatest for those in professional and managerial occupations. Zero visibility then falls consistently across intermediate, routine and non manual occupations and is lower still for those men and women who are long-term unemployed (12 months or more) or who have never had a paid job. It is important to note that the opposite class gradient is to be found at the other end of the visibility spectrum (analyses not shown). Seeing a police officer ‘at least once a day’ is 28 percent for professional and managerial men, rising to 34 percent and 39 percent for men in intermediate, routine and non manual classes respectively and substantially greater at 51 percent for men outside the labour market.

Figure 9: Percentage reporting no police visibility by gender and occupational class

Source: British Crime Survey, 2008/9

Contact

Having noted marked differences between familiarity and visibility of the police, we now attend to reported contact with the police in the last twelve months. Here we make a distinction between contact that is initiated by the police and contact initiated by the respondent. Out of all our measures of police presence, contact is the only one that guarantees a social interaction has taken place between the individual and at least one member of the police.

Figures 10 and 11 show that respondent led contact with the police is much more common for men and women than police led contact. There are however marked age differences between these contacts. Men are most likely to contact the police between the ages of 25 and 44 – contact is markedly lower for men in other age groups, and is halved from 30 percent at age 45-54 to only 15 percent at age 65+. Contact led by the police is also at its lowest level for older men (7 percent), having reached a peak of 15 percent for young men (16-24) and men aged 45 to 54 years.

Figure 10: Nature of contact with police for men by age

Source: British Crime Survey, 2008/9

The contact that women initiate with the police clearly peaks at age 35 to 44 at 33 percent but thereafter falls substantially in each age group to only 11 percent at age 65 and above. The data suggests that police led contact is also elevated in the 35 to 44 age group for women, who along with young women aged 16-24 are most likely to have been contacted by the police in the last year. The age differences in police contact add to the overall picture of low familiarity and visibility in older age especially where police presence can be judged to be low.

Figure 11: Nature of contact with police for women by age

Source: British Crime Survey, 2008/9

To examine ethnic differences in police contact, a variable was constructed to show the combined nature of contact over a one year period, whether or not an individual had both respondent and police initiated contact, one type or nothing at all.

Figure 12 shows that a small percentage of men had both police and respondent led contact over the reference period and that this was higher for some minority ethnic groups than for whites; 9 percent among Pakistani/Bangladeshi and 7 percent among Indian men. For Black African Caribbean men, the percentage contacted by the police only was greater than the percentage who had self- initiated contact only (10 percent and 8 percent respectively), but

Figure 12: Nature of contact with police for men by ethnic group

Source: British Crime Survey, 2008/9

in general men were more likely to have contact that they themselves initiated. For women, those belonging to minority ethnic groups were slightly more likely than whites to report no contacts at all over the last year, but Indian and Pakistani/Bangladeshi women were most likely to have had a contact led by the police.

Figure 13: Nature of contact with police for women by ethnic group

These ethnic group differences in contact provide a mixed picture to add to the one of low familarity but high visibility. There is some evidence that minority groups are more likely to have ‘no contact’ with the police over a year relative to whites, but equally it can be seen the that members of minority communities are contacting the police or having a two-way interaction with police.

Figure 14 shows how contact varies for men and women by occupational class (NSSEC). Women in professional or managerial occupations are most likely to initiate contact with the police (20 percent), but respondent contact is comparable for all other occupational groups, the long-term unemployed and never worked at 15 percent. Respondent led contact is greatest for men in the professional and managerial class and also the category of never worked and long-term unemployed, the latter also having the lowest contact led by the police.