The Scottish Government

« Previous | Contents | Next »

Listen

CHAPTER TWELVE WHAT IMPACT HAVE COMMUNITY WARDENS HAD UPON RESIDENTS' PERCEPTIONS?

Introduction

12.1 There is evidence that shows that, whilst crime rates in the country as a whole are declining, public perceptions may not always reflect this. Thus, the recent Audit Commission Report (2006) highlighted the fact that over 60% of respondents to the British Crime Survey in 2004/05 said that they believed that the incidence of crime in the United Kingdom continued to rise. In addition, over 40% felt that crime in their local area had increased over the same period (Ibid, 13). Similarly, Tilley (2005, 269) notes that the British Crime Survey has consistently found the perceived risk of victimisation to exceed the actual risk and that there are high levels of public anxiety surrounding crime.

12.2 In spite of well documented difficulties in definition, Moore and Shepherd (2006, 293) contend that fear of crime (FoC) is a powerful motivator of economic and social behaviour. This is supported by the findings of "liveability" surveys carried out in 2001 and 2005 which revealed low crime rates, or perceptions of low levels of crime, to be the single most important factor in determining where people wanted to live ( MORI (2001,2005) cited in the Audit Commission 2006, 12). Furthermore, Moore and Shepherd (2006) contend that the effects of FoC extend far beyond any tangible losses imposed, or potentially imposed, by criminal behaviour. For example, fear can result in the alteration of daily habits and negative psychological effects with a corresponding detrimental impact on the fabric of society. It is therefore unsurprising that perceptions of crime and fear of crime are now viewed with nearly equal importance as crime reduction itself in the formulation of government policy (Home Office Annual Report 2001-2002 citied in Moore and Shepherd 2006, Tilley 2005).

12.3 Given the significance of perceptions, the purpose of this Chapter is to look at changes in the views of residents in the case study areas over time. The Chapter starts by looking at the aggregate changes over all 9 areas, goes on to look at the impact in the case study areas individually and finishes by comparing selected changes in the case study areas with changes in the wider area, drawing on the results of the Scottish Household Survey ( SHS).

12.4 The methodology is outlined in Chapter Two. In summary it involved base and endline surveys in the 9 case study and 2 control areas. To enable comparisons to be made, a number of the questions in these surveys were identical to those used in the SHS. Some of the information from these surveys, for example awareness of the wardens and perceptions of their impact, is used elsewhere (Chapter Fourteen). As such it is not drawn on here.

12.5 The remainder of this Chapter looks at the results of the 2 surveys. To make the Chapter more readable the majority of the Tables on which the analysis is based are presented in Appendix 2, which is cross referenced as appropriate ³⁵. The initial analyses look at the changes in the case study areas between the 2 survey dates.

Neighbourhood Quality of Life - Case Study Areas

12.6 Interviewees were asked to rate their Neighbourhood as a place to live on a 5 point scale ranging from "Very Good" to "Very Poor" (Table A2.1). The picture that emerges in the case study areas is one of gradual improvement between the 2 survey dates. For example:-

• Those perceiving their neighbourhood to be a "Very Good" place to live increased from 35% to 42%; whilst
• Those perceiving it to be "Very Poor" decreased slightly from 7% to 5%.

12.7 This finding mirrors that of other scheme evaluations from across the country. For example, the Neighbourhood Warden Scheme Evaluation (Office of the Deputy Prime Minister, 2004, 20) identified a 27% increase in satisfaction in areas patrolled by wardens between January 2001 and May 2003. Similarly, the National Evaluation of the Street Wardens Programme (Department for Communities and Local Government, 2006, 10) found residents in street warden areas to be more satisfied with their local area as a place to live than they were 2 years earlier.

12.8 In terms of community involvement (Table A2.2) interviewees were asked how involved they were in their local community. The picture that emerges is mixed:-

• There has been a slight increase in those who claim a "Great Deal" of involvement (from 4% to 6%); however
• Those who claimed to be "Not At All" involved increased from 34% to 51%.

The interpretation of these statistics must be speculative. However, generally there has been a decrease in involvement in community based activities ³⁶. Contrary to this are the efforts that have been made in some areas, particularly areas of deprivation, to stimulate greater community activity, through such initiatives as Community Voices. What the statistics may therefore be showing is this polarisation, with most residents having little wider involvement (in line with society generally), whilst a few are very intensively involved.

Neighbourhood Safety - Case Study Areas

12.9 The SHS asks questions about how safe people feel when alone in their homes, and when walking alone in their neighbourhoods at night. In order to allow some comparisons with Scotland as a whole the base and endline surveys asked similar questions.

12.10 The picture that emerges from these 2 questions in the case study areas (Tables A2.3 and A2.4) is again one of gradual improvements. Thus:-

• The percentage of those feeling "Very Safe" when walking alone at night increased from 27% to 32% whilst those feeling "Very Unsafe" fell from 19% to 15%; and
• The percentage of those feeling "Very Safe" in their homes increased from 52% to 62% between the 2 survey dates, whilst the percentage feeling "Very Unsafe" declined from 5% to 3%.

12.11 This conclusion is in keeping with the findings of the Neighbourhood Warden Scheme Evaluation (Office of the Deputy Prime Minister, 2004) in England and Wales which concluded that wardens have had greatest success in addressing quality of life issues and fear of crime. Within those areas patrolled by the wardens, resident concern over crime was seen to have reduced relative to comparator areas. In addition, the percentage of residents in scheme areas who believed that crime had increased in the past year was below the national average of 53%. In Northern Ireland conclusions differed slightly. Opinion among tenants was divided as to whether there had been any change in the levels of crime and fear of crime in the community since the warden initiative started. However, amongst those who believed change had occurred, the majority believed that crime and fear of crime had decreased either slightly or significantly (Northern Ireland Housing Executive, 2003, 66).

12.12 Given that perceptions of fear of crime can influence attitudes and opinions about ones neighbourhood, these results would seem to indicate that the quality of life in the case study areas has been gradually improving. This is in keeping with the findings from other evaluations. Why this should be, might become clearer when we examine the extent to which perceptions of antisocial behaviour in the case study areas have changed between 2005 and 2006.

Perceptions of the Occurrence of Antisocial Behaviour - Case Study Areas

12.13 Both the base and endline surveys asked a number of questions about the perceptions of the prevalence of particular neighbourhood problems and antisocial behaviours. The types of antisocial behaviour about which questions were asked were:-

• Noisy neighbours or regular loud parties;
• Vandalism, graffiti or deliberate damage to property;
• Intimidation and harassment;
• Drug misuse or dealing;
• Rubbish or litter lying about; and
• Rowdy behaviour.

12.14 Selection of these types of antisocial behaviours seems appropriate given that a 2006 survey commissioned by ADT Europe ( ADT Europe, 2006) found disrespectful behaviour, rowdy behaviour and vandalism to be forms of antisocial behaviour highlighted most frequently by respondents from Great Britain. Likewise the Audit Commission (2006) found that one in 3 people living in deprived areas believed that antisocial behaviour was damaging their quality of life, whilst the Neighbourhood Warden Scheme Evaluation (Office of the Deputy Prime Minister, 2004) found that antisocial behaviour continued to be the principal cause of dissatisfaction for 46% of residents in warden scheme areas. Accordingly, whilst it is acknowledged that "quality of life" is affected by multiple and complex components, perceptions of antisocial behaviour seem to be a significant factor.

12.15 For each type of antisocial behaviour interviewees were asked to rate their perceptions on a sliding scale, from the occurrence being "Very Common" to "Not at all Common" in their neighbourhood. For the endline survey this question was repeated, along with a further 2 questions that asked:-

• If the antisocial behaviour occurrence had become "More Common", "Less Common" or had "Stayed the Same" over the last 12 months; and
• Those who claimed that the occurrence had become "Less Common" were asked if they felt that wardens were responsible for this reduction.

The data for each of these questions for each type of antisocial behaviour is shown in Tables A2.5 to A2.22 in Appendix 2.

12.16 Interpretation of the results is not straightforward, mainly as no simple or consistent, pattern emerges from the data. In part this may reflect the fact that some types of antisocial behaviour may be more easily influenced by the wardens' actions than are others, something highlighted in Chapter 11 (Paragraph 11.3).

12.17 If we try to interpret the data, by looking at the perceptions of behaviours that are said to be "Very" or "Fairly Common" (Table 12.1) then it seems that between the base and endline surveys:-

• There has been no or little change in the perceptions of the prevalence of incidents of noisy neighbours and vandalism; and
• The other 4 types of behaviours have all seen reductions in the numbers stating they were "Very" or "Fairly Common", with 3 showing differences of over 20% between the base and endline survey dates (Harassment, Rubbish/Litter and Rowdy Behaviour).

Given this, it could be argued that, between 2005 and 2006, the quality of life in the 9 case study areas has improved, in part because occurrences of a range of antisocial behaviours have decreased. It is particularly worth noting that no incidences of antisocial behaviour are believed to have become more common.

TABLE 12.1 Comparisons of Baseline and Endline Surveys on Perceptions of the Occurrence of Particular Types of Antisocial Behaviour

Note:-

1. This information is derived from Tables A2.5, A2.8, A2.11, A2.14, A2.17 and A2.20.

12.18 That the wardened areas have seen improvements in the incidence of a number of types of antisocial behaviour mirrors the findings from other research. For example:-

• The Neighbourhood Wardens Scheme Evaluation (Office of the Deputy Prime Minister, 2004) found evidence of changing perceptions in relation to antisocial behaviour, with there being a 0.7% decline within wardened areas amongst those stating that young people loitering was a problem compared to a 5.4% increase in the areas without wardens;
• In Northern Ireland two fifths of tenants who participated in the survey stated that there had been fewer incidences of antisocial behaviour since the introduction of the wardens (Northern Ireland Housing Executive, 2003); and
• In England and Wales the wardens were felt to have had a positive impact on environmental problems such as litter and dog fouling (Department for Communities and Local Government, 2006).

12.19 Overall the survey results indicate that the quality of life in the case study areas has generally improved between the 2 survey dates. For example:-

• Ratings of the neighbourhoods as places to live have improved;
• Fear of crime has decreased; and
• A range of antisocial behaviours are felt to have become less common.

12.20 Given this, when the changes in the case study areas are considered in isolation, there is evidence of a positive impact on the quality of life. However, we now need to consider the extent to which these positive changes are mirroring changes in the control areas.

Changes in the Control Areas

12.21 The control areas were intended to represent the "policy off" scenario. The aim was to look at the changes within the wardened case study areas, compare these to changes in the control areas and then ascribe any differences to the wardens' interventions. The control areas (Oxgangs and Firhill in Edinburgh and Rattray in Perth and Kinross) were selected so that, as far as possible, they had similar characteristics to the wardened areas, but did not have additional antisocial behaviour initiatives in place (Paragraphs 2.26 to 2.27).

12.22 When the endline surveys are considered what emerges is a picture of improvements in both the case study and control areas. However, the magnitude of the improvements in the control areas has, for some variables, been greater than in the case study areas, in some cases by a considerable margin. For example:-

• The percentage increase in those feeling "Very Safe" walking after dark between the base and endline surveys was 18% for the case study areas but 112% for the controls, whilst the comparable figures for feeling "Very Safe" in the home were 19% and 53% (figures derived from Tables A2.3 and A2.4, Appendix 2); and
• There has generally been a greater decrease in the percentages of those claiming that certain types of antisocial behaviour were "Very Common" in the control than in the case study areas (Table 12.2). Thus it can be seen that:-
• • For 4 of the types of antisocial behaviour the improvements in the control areas were greater than in the case study; and
  • For Rubbish/Litter and Rowdy Behaviour there was a greater comparative change in perceptions in the case study areas. Interestingly Rowdy Behaviour is one of the forms of antisocial behaviour which had a very low level of attribution to the wardens (see Table 14.1) ³⁷.

TABLE 12.2 Comparative Changes for Those Claiming that Certain Types of Antisocial Behaviour were "Very Common" between 2005 and 2006

Note:-

1. The percentages are derived from Tables A2.5, A2.8, A2.11, A2.14, A2.17 and A2.20, in Appendix 2.

12.23 These changes in both types of area are reflected in perceptions of the quality of the areas as places to live (Table 12.3). Both areas have improved but the improvements have been greater in the control areas.

TABLE 12.3 Comparative Changes in the Ratings of Neighbourhoods as Places to Live between 2005 and 2006

Note:-

1. The percentages are derived from Table A2.1, Appendix 2.

12.24 Why have the control areas seemingly improved far more than the wardened case study areas? Both areas were initially similar in both socio-economic and perceptual terms, yet the wardens' interventions do not seem to be reflected in a greater differential improvement in the case study areas. One can put forward 2 overlapping hypotheses to explain this:-

• The introduction of the wardens, as the evidence in Chapter 14 shows (Paragraph 14.5), resulted in residents thinking that they could make a difference. This might result in residents becoming less tolerant and their perceptions of the intensity and severity of antisocial behaviours may have increased as they now expect that something will be done about it. This interpretation of the statistics shows the wardens in a very positive light; and/or
• It may be that the control areas have experienced additional interventions, or other changes, that have resulted in perceptions of antisocial behaviour changing significantly.

12.25 There is evidence from the case study areas that perceptions of wardens amongst some are positive (see, for example, Paragraph 14.5), This might lend credence to the first interpretation: that residents in wardened areas have become less tolerant as they expect that the wardens will make a difference. They therefore become more critical about local conditions (Table A2.25).

12.26 There is also evidence to support the fact that there have been changes in the control areas, despite attempts being made to select stable areas. Thus, the Edinburgh wardens' manager attributes the changes in Oxgangs/Firrhill to the clearance of tower blocks that had been a focus for antisocial behaviour. The demolition of the blocks is said to have changed residents' perceptions of crime considerably. This may explain why the fear of crime in the control areas has decreased so significantly when compared to the case study areas (Paragraph 12.22). The reasons for the changes in Rattray are not as clear cut. The Perth and Kinross scheme manager ³⁸ is not aware of any physical changes in the area, or of any changes to housing management, that might have resulted in antisocial tenants being moved elsewhere. In terms of antisocial behaviour initiatives, the local authority's Youth Services department has set up "Midnight Football" schemes. However, the manager did not feel that they alone could account for the substantial changes in perceptions.

12.27 Difficulty in identifying changes attributable to wardens is not unique to schemes in Scotland. The National Evaluation of the Street Wardens Programme (Department for Communities and Local Government, 2006) concluded that there was no measurable difference in the change in perceptions of livability ³⁹ between the wardened and comparator areas. Two explanations were proposed:-

• Despite being well implemented, schemes had failed to improve livability of their neighbourhoods; and/or
• It is too difficult to measure any change in livability brought about through wardens by relying on residents' surveys as:-
• • Livability is a complex concept which is therefore problematic to measure;
  • The Street Wardens Programme used a number of approaches that had varying degrees of success;
  • The impact of the Street Wardens Programme has been dispersed; and
  • Local authorities may have directed additional resources towards non-wardened areas in compensation.

12.28 This failure to be able to clearly identify a differential impact is disappointing, although not unexpected in the light of other evidence. We will now turn to look at change in the individual case study areas.

Changes in Perceptions in the Individual Case Study Areas

12.29 Table 12.4 looks at the changes across the 9 areas on perceptions of antisocial behaviour for 5 of the indicators used in the base and endline surveys ⁴⁰. The negative changes (that is change that is moving in the correct direction with perceptions of the incidence of antisocial behaviours decreasing) have been highlighted.

12.30 What emerges is a very mixed picture. Of the 9 areas:-

• East Renfrewshire and Perth and Kinross show decreases across all the 5 indicators, often by quite substantial amounts. Residents therefore feel that their areas have improved considerably between the 2 dates. Arguably Dumfries and Galloway can also be placed into this category, given that it shows improvements on 4 of the indicators, whilst the fifth (drug misuse) is static;
• In Inverclyde 4 of the indicators have moved in the correct direction, with the fifth (vandalism) showing an increase;
• In 3 areas (Aberdeen, Dundee and Edinburgh) there is a mixed picture, with different combinations of pairs of indicators showing improvements whilst the other 3 show increases in the prevalence of antisocial behaviour; and
• At the other extreme there have been significant increases in the perceptions of incidences of 4 of the indicators in Orkney, albeit that the base is often quite small. Thus 7% of interviewees felt that noisy neighbours were "Very" or "Fairly Common" in 2005. This increased by almost a third in 2006. Over the same period perceptions of drug misuse increased threefold.

12.31 When the indicators are considered individually across the schemes no pattern emerges. For example, it is not the case that some indicators are all moving in the same direction, which might indicate that the wardens were able (or were not able) to have an impact upon some activities but not others.

12.32 To try to explain these changes we can first look at the changes in the wider area, using the SHS in an attempt to isolate the wardens' interventions from changes in the wider area. For example, although the perception of an incidence might be increasing in one of the case study areas, if this is not increasing as fast as within the wider area then it could be argued that the wardens are having a positive impact. However, there are problems in using the SHS data, as highlighted earlier (Paragraph 2.29-2.30). For example, the wording of the questions between the surveys and SHS was not always comparable ⁴¹. Of more significance is the fact that local authority level data is only collated every 2 years for the smaller authorities. For the intervening year the data is only available for geographically larger areas. This means that while the baseline position across the wider area was measured by reference to the SHS results at Local Authority level, the endline had to be measured by reference to a larger geographical area in some cases. For this reason some scheme managers were very sceptical about the validity of any comparisons using SHS data. Our view, however, is that it is better to have some comparative data that none at all. Table 12.5 shows the areas for which the data is available in 2004/05 and 2005/06.

TABLE 12.4 Changes in the Incidence of Antisocial Behaviour across the Whole Local Authority Area in which the Case Study Areas are Located

TABLE 12.5 Areas for Which Scottish Household Survey Data is Available

12.33 Table 12.6 compares the percentage changes in perceptions of the prevalence of various forms of antisocial behaviour in the case study areas (measured using the base and endline surveys) and changes in the wider area (measured by the SHS) between the base and endline surveys. Again the cells in the Table have been highlighted where the decrease in perceptions of the indicator being "Very" or "Fairly Common" are greater in the wardened area than in the wider area. Although the caveats about the different geographies need to be borne in mind, what emerges is that:-

• Again both the East Renfrewshire and Perth and Kinross case studies have experienced changes in perceptions between the 2 survey dates that are greater, in some instances far greater, than the changes in the wider areas;
• One area (Aberdeen) has experienced relative increases in perceptions of the incidence on all 5 indicators, that is antisocial behaviour has got relatively worse;
• A further 3 areas (Dumfries and Galloway, Inverclyde and Orkney) have experienced relative positive changes on 3 indicators and negative on the other 2. Of these, all 3 areas have experienced positive change on perceptions of the prevalence of noisy neighbours and rubbish and litter. All 3 have also seen negative change on the intimidation indicator; and
• The final 3 areas (Dundee, Edinburgh and South Lanarkshire) have seen negative changes on 4 out of the 5 indicators.

TABLE 12.6 Comparative Changes in the Incidence of Antisocial Behaviour Between the Case Study and Wider Areas

12.34 Table 12.7 looks at changes in perceptions of safety across the case study areas. Again the cells in the Table that show increases between the 2 survey dates have been highlighted. Overall a far more positive picture emerges, with only Inverclyde showing a decrease in perceptions of safety on both indicators whilst 5 areas show increases on both. Some of these are quite considerable, for example East Renfrewshire.

TABLE 12.7 Comparative Changes in Perceptions of Safety In the Case Study Areas

12.35 Table 12.8 looks at the comparative changes in perceptions of safety. Again the cells in the Table have been highlighted to show those areas where there has been a comparative increase in the percentage of residents who reported feeling safer. What is apparent is that:-

• In 5 of the case studies (Dumfries and Galloway, Dundee, East Renfrewshire, Orkney and South Lanarkshire) feelings of safety have increased in comparative terms on both indicators;
• In Aberdeen and Inverclyde comparatively residents feel less safe; and
• In Perth and Kinross and Edinburgh there is a mixed picture, with both showing increases on one indicator and decreases on another.

TABLE 12.8 Comparative Changes in Perceptions of Safety Between the Case Study and the Wider Area

What Impact Has There Been in the Case Study Areas?

12.36 The point was made earlier that, when the case study areas are considered in isolation, the quality of life for residents seems to have improved (Paragraph 12.19). There is also evidence that attributes these changes to the wardens. For example:-

• Residents feel that the wardens have the potential to improve the quality of life in their areas (Paragraph 14.5);
• A third felt that the wardens had already had a positive impact; and
• A minority attributed positive changes in the occurrence of various types of antisocial behaviour to interventions by the wardens.

12.37 However, when the changes in perceptions of various forms of antisocial behaviours are compared to changes in the control areas a less positive picture emerges, with the control areas experiencing more positive change on most of the indicators. This might imply that the wardens are having limited impacts, although changes in the control areas, especially in Edinburgh, mean that such an interpretation is probably erroneous.

12.38 When the individual case study areas are considered a very mixed picture of changes in perceptions of the various antisocial behaviour indicators emerges in that only 2 of the 9 case studies show improvements on all 5 indicators. However, when the indicators are considered across the case studies:-

• Of the 45 cells in Table 12.4 (which looks at changes in incidences of 5 indicators) 27 (60%) show positive change; and
• When the changes are placed in context of the comparative change in the wider area (Table 12.6), then of the 45 cells 22 (49%) show changes that compare favourably with changes in the wider area.

12.39 An even more positive picture emerges when perceptions of safety are considered. For example:-

• Of the 18 cells in Table 12.7, 13 show positive change (72%); and
• Of the 18 cells in Table 12.8, 13 (72%) demonstrate changes that are greater than those in the wider area.

When making these interpretations the deficiencies in the SHS data (Paragraph 12.32) need to be remembered.

12.40 Overall we would argue that, in the areas where wardens operate, there have been comparative positive changes in perceptions of the frequencies of a range of antisocial behaviours and, to a greater extent, in perceptions of safety, both in the neighbourhood and in the home. The data is not "clean", in that by no means all of the areas show positive changes on all indicators. However, given the complicated inter-relationships between interventions and impacts, and the fact that the wardens in different areas are likely to have different priorities, this must be expected. It is however, worth considering if there is any relationship between perceptions and the reality of crime and antisocial behaviour.

Do Perceptions of Crime Bear any Relationship to its Reality?

12.41 Exploring the relationship between the reality of crime (as measured by the statistics of recorded crime) and perceptions of its frequency, is difficult for a number of reasons:-

• As has been highlighted earlier (Paragraph 9.21) recorded crime is not an accurate reflection of the "true" levels of crime within an area; and
• Comparable data across the case study areas are not always available.

12.42 Accepting these problems, it is possible to look at the relationship between the recorded number of incidents of vandalism and residents' perceptions of vandalism across 5 of the case study areas (Table 12.9). What can be seen from the Table is that there does seem to be a relationship in that (with the exception of Dumfries and Galloway) changes in recorded vandalism levels are matched by changes in residents perceptions of the extent to which vandalism is felt to be "Very" or "Fairly Common". Although this is a very small sample, this visual relationship is backed up by statistics, there being a correlation coefficient of +0.949 between the 2 variables. This implies that as reported incidents of vandalism rise or fall then there is an almost exact rise or fall in perceptions of their incidence ⁴².

TABLE 12.9 Comparative Changes in Recorded Crime and Perceptions of its Incidence - Vandalism

Note:-

1. The data for Dumfries and Galloway relates to the change between year 1 and 2 as no baseline was provided.

Conclusions

12.43 On the basis of the evidence presented here it could be argued that perceptions of safety and antisocial behaviour are grounded in reality. Thus, as is shown in the earlier paragraphs, if perceptions are becoming more positive this might imply that crime and antisocial behaviour occurrences are decreasing. There is also evidence that attributes these changes, at least in part, to the wardens. Accordingly we would argue that the surveys indicate that the wardens are having a positive impact on residents' perceptions which, in its turn, reflects their success in reducing crime and antisocial behaviour.

12.44 This impact is not always simple to interpret and seems to vary across the case study areas. Accordingly this may be seen as an unduly optimistic interpretation, given that, in Table 12.6, around half of the Table's cells indicate that the case study areas have experienced changes in perceptions that are not as positive as those experienced in the wider comparator areas. However, against this has to be balanced the fact that, at the time of the endline survey, the wardens had been operating for a maximum of 2 years and often less. They were also working in areas where the antisocial behaviour problems were long standing and often quite severe. Given these factors one could argue that the improvement seems to have been better than one might have expected.

« Previous | Contents | Next »