The Scottish Government

« Previous | Contents | Next »

Listen

CHAPTER TEN THE IMPACTS OF THE WARDENS ON CRIME IN THE CASE STUDY AREAS

Introduction

10.1 To measure the impact of the warden schemes on crime and antisocial behaviour, each scheme was required to produce a plan for monitoring and evaluating their service and required to collect survey and statistical indicators at 6, 12 and 24 month intervals. These were then to be used to monitor progress. As Chapter 8 makes clear, when the survey based indicators were assessed, following the submission of the 6 month reports, there was a degree of concern about the methodologies that some of the schemes were using. To remedy these deficiencies the decision was made to undertake independent attitudinal base and endline surveys in the case study and 2 control areas. The results of these are reported in Chapter Twelve. As such the focus of this Chapter is upon the statistical indicators.

10.2 As Chapter 2 makes clear, the schemes were to select a set of core statistical indicators from a list provided by the Executive. Schemes were, however, at liberty to select other indicators in addition to those prescribed by the Executive. The schemes were therefore undertaking self-assessment, with progress being measured against indicators that they had selected themselves: indicators that were to reflect objective attainment. However, as Chapter Eight (Paragraph 8.10) makes clear, in retrospect some scheme managers now feel that some of the indicators originally selected are no longer appropriate, in part as experience has shown that wardens are unable to have an impact upon some aspects of antisocial behaviour and crime.

10.3 The statistical indicators suggested by the Scottish Executive for use in the monitoring and evaluation of the schemes covered:-

• The impact on crime (based on levels of recorded crime, focusing on crimes such as vandalism, littering, housebreaking and harassment);
• The impact on the number of recorded incidents of antisocial behaviour; and
• The impact on the number of empty properties and rental income generated.

As the above list shows, the Executive was generally not so prescriptive as to specify the exact indicators to be used. For example, antisocial behaviour covers a multitude of different crimes and offences. As such schemes had considerable discretion as to the indicators they selected.

10.4 Ideally impact assessment would be undertaken by looking at the changes in a range of crime and antisocial behaviour statistics over time across all of the Scottish warden schemes. These changes would then be compared to changes in the wider areas in which the schemes were based and the differential impact of the schemes could therefore be isolated. However, each of the schemes selected different combinations of indicators, reflecting the local priorities of par tners and differing objectives. This means that there are relatively few indicators (if any) that are common to all schemes. It is also the case (as Chapter Eight makes clear) that the quality of the monitoring data provided varies considerably, with relatively few schemes providing complete time series statistical data from base line through to the 24 month report.

10.5 Given these methodological issues, the approach taken in this Chapter is to analyse the data that relates to the selected indicators for the 9 case study areas. Each area is looked at alphabetically, starting with Aberdeen. The approach is to look at the changes in the statistical indicators and to compare these changes with the wider trends in the local authority area as a means of isolating the impact of the warden schemes. When possible the changes are related to the various theories outlined in Chapter Nine. The case studies and the survey data (that is explored more thoroughly in Chapter Twelve) are also drawn upon to provide additional evidence to explain the observed changes.

10.6 The analysis is complex and often does not result in clear conclusions being reached as to the impact that the wardens are having upon crime and antisocial behaviour. The analysis is further complicated by:-

• The failure of all schemes to provide full data sets;
• Variations across schemes as to whether they collected, and provided, data for all of the sub-areas in which the wardens operated. Generally the larger schemes did this, whilst the smaller ones tended to provide aggregated data only; and
• The existence of other initiatives targeted at antisocial behaviour, both within the areas patrolled by the wardens and the wider local authority area.

Aberdeen

10.7 In the Aberdeen case study 6 indicators were collected as part of the monitoring of the scheme. They covered:-

• Vandalism;
• Breach of the Peace;
• Drunk and Incapable;
• Racially Aggravated Conduct;
• Wilful Fire Raising; and
• Urinating in Public.

These indicators have been collected for 6 warden areas in the city. These cover Heathryfold/Middlefield (though there is no baseline data for this scheme which means it has been excluded from this analysis), Northfield and Cummings Park, Tillydrone and Woodside, Torry, Mastrick and Seaton and Old Aberdeen/Powmis. Where data for the whole of the local authority area are available this is used to draw comparisons.

10.8 Vandalism accounts for the single largest number of crimes in the Aberdeen warden scheme. Overall, the number of crimes increased from a baseline of 1,544 to 1,933 in year 2, an increase of 25%. This was faster than a 15% increase in Aberdeen as a whole, which suggests that the wardened area has performed worse than the local authority area.

10.9 There was a significant degree of variation across the warden areas in the city. For example:-

• There was a 120% increase in incidents in Mastrick, although most of this took place between the baseline and year 1;
• There was a 52% decline in vandalism in Northfield and Cummings Park overall, with most of the decline occurring between the baseline and year 1, although there was a further, albeit slower, decline between years 1 and 2;
• There was an 87% increase in Torry between the baseline and year 2, although this hides a 4% decline between years 1 and 2;
• There was a 16% increase in Tillydrone and Woodside between the baseline and year 2, although this hides a 5% decline between years 1 and 2; and
• There was a 31% increase in Seaton, Old Aberdeen and Powmis between year 1 and year 2.

The changes across the individual areas and for the whole of the City are shown in Chart 10.1.

CHART 10.1 Changes in Reported Incidents of Vandalism in Aberdeen

10.10 This is a mixed picture showing some apparent successes alongside some significant growth in the number of vandalism crimes. When the results of the perception surveys are considered (Chapter Twelve, Table 12.4) it can be seen that the general view is that vandalism has increased between the base and endline surveys. However, consultations with the police and the scheme manager, suggest that proactive policing (supported by the wardens) has increased detection rates for vandalism under a city-wide crackdown. However, this might also result in perceptions of the scale of vandalism increasing. The opinion of the police and the local authority is that the overall trend is positive and that the increase reflects vandalism having been made a priority for Grampian Police. This has increased the recording of incidents, the detection of perpetrators and the perceptions of it as a crime. Given this it could be argued that the trend conforms to the Rise (Positive) theory. However, it is equally plausible that the Rise (Negative) theory applies. As such it is wise to assume that no conclusive explanation can be put forward.

10.11 The number of recorded Breach of the Peace incidents was also used to monitor the impact of the warden service. Overall, there was an increase in the number of offences: from a baseline of 971 to 1,686 in year 2, an increase of 74%. This increase was significantly faster than the 52% increase in Aberdeen for the same 2 year period.

10.12 There was a significant degree of variation across the wardened areas, further complicating the picture. The key trends were that:-

• There was a 188% increase in the number of Breach of the Peace offences in Mastrick, though most of this occurred between the baseline and year 1;
• There was a 91% increase in Tillydrone and Woodside. Again most of this took place between the baseline and year 1;
• There was a 10% decline in the number of offences in Northfield and Cummings Park. This decline was relatively small but consistent between each of the years; and
• There was a 66% increase in offences in Seaton, Old Aberdeen/ and Powmis and a 50% increase in Torry. However, in each case the overall increase masks a decline (of 3% and 14% respectively) between years 1 and 2.

The changes are illustrated in Chart 10.2.

CHART 10.2 Changes in Reported Breach of the Peace Offences in Aberdeen

10.13 Again, this is a mixed picture, showing some decreases as well as some major increases in the number of offences. Although there is some evidence, from the perception survey (Table 12.4), that perceptions of antisocial behaviours most closely associated with breaches of the peace (noisy neighbours and intimidation) have improved over time, a more cautious interpretation would seem to be that the changes conform to the Rise (Negative) theory.

10.14 Overall, there was a 29% increase in the number of reported drunk and incapable incidents. This is an increase from the baseline position of 209 to a year 2 figure of 269. The overall change though was slower than the 97% increase in the number of incidents across Aberdeen. This change also masks a 6% decline between year 1 and year 2. This fits with the Rise (Positive) theory in that the overall change is actually an improvement relative to the City.

10.15 The number of racially aggravated conduct incidents declined by 8% across the warden areas. This change has largely been driven by the decline between years 1 and 2 after a small increase from the baseline to year 1. Unfortunately an Aberdeen City comparator is not available that is calculated on the same basis. As such no definitive conclusion as to impact can be drawn.

10.16 The number of urinating in public offences increased by 400% across the Aberdeen warden schemes. However, this amounts to just 20 more incidents in year 2 than in the baseline. Across Aberdeen as a whole there were only 43 more incidents over the same period, suggesting that the warden areas accounted for half of the City's increase. Given these small numbers it is dangerous to read too much into this change. However, the very public and victimless nature of this offence means that it is possible that this increase in recorded incidents was driven by an increase in detection rates which could, in part, be due to the presence of the wardens. Despite this we feel that no conclusion as to impact can be drawn.

10.17 The data collected on fire raising incidents only covers the year 1 and year 2 monitoring periods, as no baseline data were collected. Between years 1 and 2 there was a 5% decline in the warden areas. The picture across the sub-areas was mixed, with there being:-

• A 15% decline in Tillydrone and Woodside between years 1 and 2;
• A 3% decline in Torry and a 1% drop in Mastrick; and
• An increased number of incidents in Seaton, Old Aberdeen and Powmis and Northfield and Cummings Park (21% and 16% respectively).

While it is not possible to arrive at robust conclusions from just one year's data the figures would appear to suggest that this change fits with the Positive Intervention theory. Initiatives such as the Gramps scheme will have played a key part in this positive change. This was developed under the banner of the Torry Total Communities initiative and involves the wardens working with a wide range of local partners, including the fire and rescue service, neighbourhood police and local nature reserve volunteers to reduce wilful fire raising in the Tullos Hill and Kincorth Nature Reserve. It may be that initiatives such as this are now being reflected in the statistics. However, without the wider local authority comparator data it is not possible to reach any definitive conclusions as to what was driving change.

10.18 Overall, it is difficult to assess the wider changes in the Aberdeen scheme. The broad pattern appears to suggest that the instances of reported crimes and offences have increased in the warden areas, with some showing a Rise and Fall pattern, but most showing a continual rise. Of the 4 indicators that show a rise in recorded crime, 2 (vandalism and breach of the peace) may conform to the Rise (Negative) theory because the increases in the wardened areas have been greater than those in the wider area. For the Drunk and Incapable indicator the increases in the wardened areas were smaller than in the wider comparator. This can, therefore, be explained by the Rise (Positive) theory. For the fourth indicator for which there was a rise in the number of incidents (Urinating in Public) we have not been able to speculate as to which theory may apply. Two indicators show a decline (racially aggravated and fire raising incidents). However, the lack of local authority comparator data means that it is not possible to explain this adequately. Overall, therefore, the analysis of the statistics shows a mixed picture with limited evidence of any across the board impact. However, this negative picture is, to some extent, contradicted by the perception surveys. For example, between the base and endline surveys there was a 48% increase in the numbers of respondents rating their Neighbourhood as "Very Good" whilst in the endline survey 51% of respondents felt that the wardens had improved the quality of life in their locality.

Dumfries and Galloway (North West Dumfries)

10.19 In the Dumfries and Galloway case study 9 indicators were to be collected to monitor the scheme, which covers the 5 communities of Lochside, Lincluden, Stakeford, Summerhill and Summerville. The data was, however, not collected for the individual sub-areas but for the scheme as a whole. The indicators were:-

• Neighbourhood disputes;
• Breaches of the Peace;
• Assaults;
• Cases of Shoplifting;
• Number of needles collected;
• Housebreaking;
• Drug Related Crimes;
• Malicious False Alarm; and
• Vandalism.

In retrospect the manager now feels that several of these indicators are things that the wardens can have little impact upon, for example assaults, shoplifting and burglaries. However, they are still being used to monitor progress and accordingly are analysed here.

10.20 There are 2 gaps in the data across the Dumfries scheme. There is no year 2 figure for the number of Drug Related Crimes and no baseline for the number of instances of Vandalism. Accordingly these indicators have not been included in the analysis.

10.21 The number of neighbourhood disputes in the Dumfries scheme has decline from a baseline of 405 to a year 2 position of 176, a reduction of 57%. This change reflects a substantial decline in the number of disputes between years 1 and 2, as there had been a considerable increase from the baseline to year 1. This seems to conform to the Rise and Fall (Confidence) theory. This is backed up by the perception survey (Table 12.4) which shows that the perception of the prevalence of noisy neighbours as an issue has decreased by 12% between the base and endline surveys. However, the absence of a local authority comparator means that this interpretation cannot be proved.

10.22 The number of Breach of the Peace offences has increased from a baseline of 79 to a year 2 figure of 262, an increase of 232%. This compares to a 6% reduction in offences across Dumfries and Galloway. However, the wardens in Dumfries take a proactive approach to public drinking offences by targeting drinking dens in the Whitesands area of the town. The increase in the number of offences may, therefore, reflect this (particularly as detection of Breach of the Peace depends heavily on surveillance) and provides support for the Rise (Positive) theory. This positive view is backed up by the perception survey (Table 12.4) where perceptions of the prevalence of 4 out of 5 types of antisocial behaviour show decreases. However, given the size of the increase, it is difficult to rule out the Rise (Negative) theory, indeed given the scale of the increase it is probably cautious to ascribe impact to this theory.

10.23 The number of assaults has increased from a baseline of 45 to a year 2 figure of 194, an increase of 331%, in contrast to the 17% decline in serious assaults and the 8% decline in minor assaults across Dumfries and Galloway. As such we conclude that this change can be explained by the Rise (Negative) theory, with crime appearing to be increasing quite substantially as against a reduction in Dumfries and Galloway as a whole. However, as mentioned later (Paragraph 11.3) it seems unlikely that the wardens can have a direct impact on assaults. As such this may not be a sensible, or fair, indicator against which to measure progress.

10.24 The number of cases of shoplifting has risen from the baseline of 15 to a year 2 figure of 120, an increase of 380%: substantially greater than the 14% increase across Dumfries and Galloway. This rapid increase, from a low baseline figure, again highlights what appears to be a consistent trend across the warden areas, of an increase in recorded crime. The change has been significantly faster than the increase in shoplifting across the local authority as a whole. As such we classify this as Rise (Negative). Yet, as with assaults, the wardens may have difficulties in having any impact upon shoplifting, which calls into question the relevance of this indicator.

10.25 The numbers of needles collected increased substantially, from a baseline of 99 to a year 2 figure of 402, a 308% increase. However, this indicator differs from the others in that it is an output, not an outcome, measure: that is it is a measure of activity not impact. As such it does not lend itself to interpretation in terms of the theories outlined in Chapter 9.

10.26 The number of housebreakings has more than doubled (+113%) from a baseline of 23. This increase was in stark contrast to a 42% decline across Dumfries and Galloway. This change is heavily skewed by the low baseline, which means that the overall change is not as extreme as it might appear. Again, the pattern shows a clear increase at a time when recorded incidents in the whole local authority area have fallen. This suggests the Rise (Negative) theory, with the wardens being unable to influence this indicator.

10.27 The number of malicious false alarm calls has increased from a baseline of 8 to a year 2 total of 21, an increase of 263%. However, given the small absolute numbers, it is probably foolish to read too much into this. Accordingly no conclusions as to impact have been drawn.

10.28 Overall there appears to be a clear trend in the Dumfries scheme that suggests that there has been limited impact on crime or antisocial behaviour by the wardens. On almost all indicators the figures have increased rapidly from the baseline and between years 1 and 2 at the same time as the wider local authority area has been experiencing sustained reductions in the same types of crimes and offences. However, this rather negative interpretation has to be balanced by the generally positive picture that merges from the perception survey (see, for example, Tables 12.4 and 12.7).

Dundee

10.29 In Dundee 8 monitoring indicators were selected for the 6 warden areas in the city: Hilltown and Bowbridge, Douglas, Whitfield, Stobswell, Lochee West and Charleston and Pitkerro. The indicators are:-

• Youth Disturbance calls;
• Noise related calls;
• Vandalism;
• Break-ins to Motor Vehicles;
• Theft of Motor Vehicles;
• Fire incidents Reported for Malicious Intent;
• Malicious Fire Incidents Attended; and
• Malicious Fire False Alarms.

10.30 Although the indicators are being used to monitor the scheme, and are reported to the Executive in the templates, the Dundee scheme manager is now reviewing the 3 indicators associated with malicious fire raising, in partnership with the fire and rescue service, to assess their suitability. The manager also now feels that it is unclear what impact the wardens can have on the 2 motor vehicle offences. However, as these 5 indicators are being reported to the Executive they will be analysed, although they will only be assessed for the whole scheme, rather than the sub areas, in order to provide some assessment of the direction and scale of change.

10.31 Overall, the number of Youth Disturbance Calls to the police has declined from a baseline of 3,109 to a year 2 figure of 2,423: a 22% reduction. This overall decline was generally matched in the individual sub-areas (see Chart 10.3). Thus:-

• There was a 46% decline in the number of calls in Douglas;
• The were substantial decreases in Stobswell (33%), Pitkerro (30%) and Hilltown and Bowbridge (29%);
• There was a slower decline in the number of calls in Whitfield (5%); and
• The only area to experience an increase was Lochee West and Charleston where the number of calls increased by 12%.

Dundee-wide data is only available for years 1 and 2. This shows a decrease of 19% (9,096 to 7,390), which is close to the decline in the wardened areas overall. Given this, the conclusion is that change in this indicator is Neutral. Although this is unlikely to reflect the No Impact theory (given what is known about the wardens' activities in Dundee) it may be due to Spillover or Compensation.

CHART 10.3 Changes in Reported Youth Disturbances in Dundee

10.32 Related to this indicator is the number of Noise Related Calls to the police. Overall, there has been a 9% increase in these between the baseline and year 2. This means the number of incidents has increased from 1,220 to 1,335 in the 2 years for which the warden service has been collecting data. Comparator data is only available for years 1 and 2 for the whole of Dundee. This shows a 24% increase (from 3,915 calls to 4,843). The comparable increase in the wardened areas was 3% (from 1,292 to 1,335 incidents).

10.33 There was a mixed pattern across the 6 warden areas:-

• There was a 79% increase in the number of noise related calls in Douglas;
• There was a 28% increase in Lochee West and Charleston. However, this hides a 5% decline between years 1 and 2;
• There was a 5% increase in Hilltown and Bowbridge;
• There was a 2% increase in Stobswell, although this masks a 6% reduction between years 1 and 2; and
• There were reductions in both Pitkerro (18%) and Whitfield (8%).

The changes are illustrated in Chart 10.4.

CHART 10.4 Changes in Reported Noise Related Calls in Dundee

10.34 The picture, therefore, varies across the 6 areas, with 4 experiencing an overall increase in calls. Interpretation is, however, complicated by the introduction of Noise Wardens by the local authority. The scheme's manager feels that this has contributed to the overall increase in calls. Given this, it is very difficult to relate any changes specifically to the interventions made by the wardens and to explain these in terms of the earlier theories. Accordingly no conclusion is drawn as to impact.

10.35 The number of reported incidents of vandalism has gone from 1,175 at the baseline to 1,240 in year 2, a 6% increase. This was in contrast to the 3% decline across Dundee as a whole. However, the overall increase hides a 5% decline between years 1 and 2 (although for the same period the decline was 6% across Dundee). This means that, even accounting for the decline between years 1 and 2, the overall picture is one of a worsening situation in the warden scheme areas.

10.36 Again, there was a mixed pattern across the 6 areas, as Chart 10.5 shows. The key trends were:-

• A 36% decline in Stobswell;
• A 26% decline in Douglas;
• There were also lesser falls in Lochee West/Charleston (3%) and Pitkerro (1%);
• However, there was a 55% increase in Hilltown and Bowbridge, although this all took place in year 1 with there being virtually no change between years 1 and 2; and
• Vandalism increased by 37% in Whitfield, although this hides a 2% decline in the area between year 1 and year 2.

CHART 10.5 Changes in Reported Occurrences of Vandalism in Dundee

10.37 Whilst there is a degree of variation across the 6 areas, there has been a reduction in the number of reported occurrences of vandalism in 4 areas overall, whilst one other exhibits rise and fall. There was an overall increase in the final area, although this hides little change in the number of crimes between years 1 and 2. The perception surveys show that perceptions of vandalism have increased (Table 12.4). Given this it would seem that the best theoretical explanation for these changes is the Rise (Negative) pattern.

10.38 The number of break-ins to motor vehicles increased from a baseline of 256 to a year 2 figure of 266. This was a 4% increase from the baseline, in contrast to the 20% decline across Dundee as a whole. However, the overall increase in the Dundee warden scheme hides a 12% decline in the number of incidents between years 1 and 2, albeit that this was less than the Dundee wide change (a decrease of 38%). This might reflect the Rise and Fall (Confidence) theory. However, the views of the scheme's manager as to the wardens' abilities to impact on motor vehicle offences may mean that this interpretation is too positive and that a more realistic interpretation would be Rise (Negative).

10.39 The number of reported thefts of motor vehicles also increased by 4% between the baseline and year 1. This increase contrasts sharply with the 15% decline in such thefts across Dundee for the same period. Again, this suggests there has been a worsening of this crime in the wardened areas relative to the change across Dundee. Theoretically this can be explained by the Rise (Negative) theory which would reflect the views of the scheme's manager.

10.40 There were declines in: the number of malicious incidents attended (down by 17%); incidents reported for malicious intent (a 14% fall); and malicious false alarms (a 7% decline). This overall decline in each of these 3 indicators hides an increase between years 1 and 2. Given this, it is difficult to relate these changes definitively to any theory, especially as the scheme's manager now questions their relevance.

10.41 There are, therefore, mixed messages in relation to the overall performance of the Dundee scheme. There were slight increases in 4 of the indicators. For 3 of these (vandalism and break-ins to, and theft of, motor vehicles) this increase occurred whilst there was a decline in the wider local authority area. This seems to conform to the Rise (Negative) theory. For the fourth indicator showing an increase (noise related calls) this was far below the increase in the wider local authority. However, as comparator data were not available for the whole period it is not possible to draw definitive conclusions. For 3 indicators related to Malicious Intent there was an overall decline. However, the absence of comparators means that firm conclusions cannot be drawn. For the final indicator (Youth Disturbance Calls) the overall impact seems to be neutral. Given this, it is not possible from the analysis of crime statistics to definitively ascribe impacts to the wardens. However, the scheme manager now feels that some of the original indicators have proved to be less relevant than originally thought for monitoring the wardens' impact. The introduction of initiatives such as Noise Wardens also makes attribution difficult. Drawing on the perception surveys seems to show that for some antisocial behaviours (Youth Disturbances) the wardens may be having an impact, but their effect on other activities is far less clear.

East Renfrewshire

10.42 East Renfrewshire collected 3 monitoring indicators. These were:-

• Vandalism to void properties;
• Graffiti; and
• Fire Raising Incidents.

The data was aggregated for the whole of the wardens' patrol areas (Thornliebank, Barrhead and Auchenbach and Neilston).

10.43 The reduction of vandalism to void properties was a primary aim of the East Renfrewshire wardens. The number of instances declined from 229 in the baseline to 167 in year 2: a fall of 27%, or 62 less instances over the period. This drop has been consistent from the baseline to year 1 and then to year 2. Closely related to the vandalism indicator is the decline in the number of removal requests for graffiti across the scheme areas. There were 59 requests in year 2, down 26% from the baseline figure of 75. Again the decline has been consistent from the baseline to year 1 and then on to year 2.

10.44 Both of these trends are closely related and show a similar picture, namely a decline in the number of incidents. This suggests that we can categorise both indicators as conforming to the Positive Intervention theory, with an immediate fall in the warden area. This suggests that the wardens might be having a positive impact across the scheme areas. However, in the absence of comparator data this must be a speculative so that it may be wiser to reach no conclusion.

10.45 There was no baseline figure for the number of incidents of fire raising so it is not possible to fully assess change. However, the available data shows that there has been a decline in the number of incidents from 420 to 347: a fall of 17% between years 1 and 2. As there is a lack of complete data for this indicator, as well as no local authority comparator, it is difficult to come to firm conclusions as to the reasons for this change.

10.46 In the absence of comparator data for the wider local authority area it is difficult to interpret the changes in East Renfrewshire. However, as Chapter 12 shows (Tables 12.4, 12.6 12.7 and 12.8) there have been improvements in residents' perceptions of the prevalence of a variety of types of antisocial behaviour as well as increases in feelings of safety. Given this the statistical interpretation may be too pessimistic.

Edinburgh

10.47 In the Edinburgh South case study 3 indicators were collected retrospectively by GEN. The reason for this was that the scheme did not appear to have collected data on a consistent basis and therefore had little useful monitoring information. The indicators covered:-

• Vandalism;
• Housebreaking; and
• Fire Raising.

These indicators cover the Edinburgh South area, as defined by the EH16 6 and EH17 7 postcode sectors. These include the areas of Burdihouse/Southhouse, Gracemount and Liberton.

10.48 The number of vandalism crimes in Edinburgh South has increased from a baseline of 281 to a year 2 figure of 469. This is an increase of 66%, or 188 more incidents in the wardened area. This was greater than the change across Edinburgh as a whole (14%). Given this, this would seem to be best explained in terms of the Rise (Negative) theory, as the problem would appear to have got worse locally.

10.49 The number of housebreaking crimes in Edinburgh South declined from a baseline of 195 to a year 2 figure of 174: a fall of 11% at a time when housebreaking increased by 10% in Edinburgh as a whole. It also represents a consistent fall since the introduction of the warden service. This would appear to suggest that the wardens are having an impact. When coupled with the wider increase in the number of crimes, this suggests that this change can be explained by the Positive Intervention theory and indicates a major improvement locally.

10.50 The number of fire raising incidents in Edinburgh South has remained relatively steady, at around 10 incidents a year, over the lifetime of the warden service. Across Edinburgh as a whole, incidents of this nature increased by around a third. The wardens work locally with the fire and rescue service on the "Cool Down Crew". This involves both services working with local youths to promote fire safety. The wardens have a good relationship with the fire and rescue service locally and the service acknowledges the wardens' role in removing combustible material from the streets. This may help to explain why there has been little change in the number of incidents at a time when they have been increasing in Edinburgh as a whole. As such we would classify this as fitting with the Positive Intervention theory, as conditions locally have improved relative to Edinburgh as a whole.

10.51 Overall, there are signs that the Edinburgh warden scheme is having an impact on the local area, with fewer fire raising incidents and less housebreaking relative to Edinburgh as a whole. Despite these positive changes, there has been a sharp rise in the number of vandalism crimes. Paradoxically, the perception survey (Tables 12.4 and 12.6) shows that vandalism was perceived to have become less of a problem between the base and endline surveys, both absolutely and relatively. This might imply that the wardens' presence is providing reassurance, regardless of the reality of the scale of crime.

Inverclyde

10.52 In Inverclyde 4 indicators were collected. These were:-

• Detection rates for Disorderly Conduct;
• Detection rates for Vandalism;
• Issuing restorative warnings ²⁶; and
• The number of persistent offenders.

10.53 The use of detection rates, rather than reported incidents, is interesting. It could be argued that this is more a measure of the effectiveness of the police service than of the wardens. However, the scheme sees this as being a measure of the way that it works closely with the police to overcome problems and ensures that offenders are caught. Despite this view, the data used are not a measure of detection rates at all. They simply count the number of detected offences. Without knowing what proportion of offences are detected we have no way of knowing whether any rise in detection is due to the increased effectiveness of police and wardens or to a rise in the number of offences that are being committed. Although the data are summarised below we do not, therefore, draw any conclusions from it. The final 2 indicators are essentially measures of long term change and reflect the work that the wardens do with Council Street Mediators and the police. The aim is to target persistent offenders and, through the use of such things as restorative warnings, ensure that they do not enter the criminal justice system. The measure of the success of these interventions will be falls in the number of persistent offenders.

10.54 The indicators cover the localities of Port Glasgow, Central Greenock, Larkfield, Bow Road and Fancy Farm. However, the scheme aggregates the data rather than collating it for each sub-area. For most of the indicators local authority-wide comparators were not available.

10.55 Detection rates for disorderly conduct have increased from a baseline of 1,200 to a year 2 total of 1,297: an increase of 8% on the baseline. This may suggest that the wardened areas have been more effective at tackling this form of antisocial behaviour but, as explained earlier, it might also suggest that there has been a rise in Breach of the Peace offences. There are no comparator local authority data. Accordingly no conclusions as to impact have been reached.

10.56 Detection rates for vandalism have declined from the baseline of 294 to the year 2 figure of 258 detected incidents: a 12% decline. Again there are no comparator local authority data. However, the perception survey (Table 12.4) found that the incidence of vandalism was thought to have increased between the base and endline survey dates. Given this, it would seem that there is limited evidence to show that the wardens are having an impact. Accordingly no conclusions are reached.

10.57 The introduction of a service from the police to issue restorative warnings has resulted in 14 warnings being issued in year 1 and a further 24 in year 2. The wardens work closely with the police and help to identify persistent offenders who can then be issues with warnings. The 24 warnings were seen as a major achievement by the Inverclyde scheme manager. However, this can be seen as a measure of activity rather than of impact. For this reason, and in the absence of local authority comparator data, it has not been interpreted in terms of any of the theories of change.

10.58 The number of persistent offenders recorded by the Scottish Children's Reporters Administration has fallen from a baseline of 21 to 14 in year 2, with the entire decline taking place between years 1 and 2.

10.59 The fall is felt by the scheme's manager to be a real success. A number of the case study areas raised the issue of the impact that persistent or troublesome individuals can have on recorded crime and, in particular, the quality of life for residents. The wardens' role is not direct but is very much a reflection of the help they give to the police to identify those who should be issued with restorative warnings. In the longer tem it is hoped that this will result in behaviour modification. However, the lack of comparators makes interpretation difficult in any theoretical sense. Accordingly no conclusion as to impact is drawn.

10.60 Overall there are some mixed patterns in the Inverclyde scheme with an increase in detections of Breach of the Peace incidents at the same time as detection rates for Vandalism have declined. However, the number of persistent offenders has declined and the number of restorative warnings issues has increased which suggests that there are positive trends in the warden areas. This positive interpretation is backed up by the perception surveys which show an increase from 17% to 40% in residents rating their area as being "Very Good" in which to live. However, the lack of comparators makes any more definitive conclusions hard to draw.

Orkney

10.61 Three indicators were collected in Orkney. These were for:-

• Vandalism;
• Complaints of antisocial behaviour; and
• The financial cost of antisocial behaviour.

The indicators cover the whole of the local authority area so that there are no local comparators.

10.62 Crimes of Vandalism account for the largest single category of crime in the Orkney area. Overall, the number of crimes increased from a baseline of 240 to a year 2 figure of 270, a 13% increase. This overall increase masks a 3% decline in the number of vandalism crimes between years 1 and 2.

10.63 Consultations with the scheme manager suggested that, given the relatively low level of vandalism in Orkney, the statistics can vary significantly as a result of the actions of a small number of people. The manager felt that much of the increase could be explained in this way. Given that there is no comparative date it is not possible to relate this change to any of the theories outlined in Chapter Nine.

10.64 The total number of complaints of antisocial behaviour in Orkney increased from a baseline of 112 to 176 in year 2: a 57% increase. Most of this increase was between years and 2 as there was an initial fall in complaints in year 1.

10.65 Accordingly there is a mixed picture on this indicator, with an initial fall being overtaken by a rapid increase in the number of complaints. This may be the result of the warden service bedding down in the area with a greater number of people now complaining because they believe that something will be done. However, the perception surveys (Table 12.4) show that for 4 out of 5 indicators of antisocial behaviour the situation is felt to have become worse between the base and endline surveys. One theoretical explanation of this might be the Rise (Negative) theory. However, it needs to be remembered that behavioural norms in Orkney are likely to be different to those on the mainland, with levels of tolerance being far lower for what many would see as being very minor offences. This, coupled with the lack of comparators, makes interpretation doubly difficult. Accordingly no conclusions are drawn as to impact.

10.66 The final indicator, looking at the cost of antisocial behaviour to the council, only has data for years 1 and 2. The figure declined from a cost of £6,408 per annum to £3,806 in 2005/06. This was 40% lower and could indicate significant progress in the warden scheme and a substantial cost reduction for the council.

10.67 As data is not available for the whole time period it is difficult to draw any firm conclusions. However, consultations with the scheme manager suggested that this was a longer term trend and that the wardens were now patrolling schools, play areas and other council properties to reduce vandalism and graffiti. It is felt that these actions explain the changes. Accordingly this could be explained in terms of the Positive Intervention theory, with the wardens having an immediate impact. Yet they lack of a comparator means that no definitive conclusions have been drawn.

10.68 Overall the indicators for Orkney show a mixed picture. The number of vandalism incidents and the number of complaints of antisocial behaviour have increased as have perceptions of the prevalence on antisocial behaviour. However, the absence of comparators makes interpretation difficult.

Perth and Kinross

10.69 In Perth and Kinross 3 indicators were selected, covering:-

• Vandalism;
• Complaints of antisocial behaviour; and
• Street cleanliness.

The indicators cover the areas of North and South Letham and Hilliland. Again the data was collated for the whole of the scheme area.

10.70 The number of incidents of vandalism increased from a baseline of 1,100 to 1,169 in year 2: a 6% increase. This was faster than the change in the local authority area as a whole (a 1% increase). All of this change took place between years 1 and 2, there being little change from the baseline to year 1. This suggests that there has been some negative change in the area, with incidents increasing faster in the warden areas than across the authority as a whole, resulting in an overall deterioration in the relative position of the wardened areas. However, this is not backed up by the perception surveys (Table 12.4) which show a 77% decrease in the proportion of residents reporting vandalism to be "Very" or "Fairly Common" between the base and endline surveys. Residents therefore feel that vandalism is far less of a problem at a time the statistics show it to be increasing. It may be that the increase therefore reflects increased reporting and, as such, this may reflect confidence in the wardens and could be interpreted in terms of the Rise (Positive) theory.

10.71 Complaints of antisocial behaviour declined from a baseline of 2,082 to a year 2 figure of 2,045: a reduction of 2%. This masks an increase of 15% between years and 2. Whilst the overall change on this indicator was positive, the scheme manager believed that the jump in incidents between year 1 and year 2 could be explained by the return to the area of offenders from prison. The location of temporary accommodation centre in Letham, which houses people with chaotic lifestyles, was felt to have added to these problems, whereby those moving to the centre created local difficulties. This was seen as an ongoing issue that could result in fluctuations in data used to monitor the service. Interpreting this in terms of theories needs to take account of the improvements of perceptions of antisocial behaviours that were positive across all 5 types of behaviour (Table 12.4). Given this our interpretation is positive, the Rise (Positive) theory, with increases in complaints between years 1 and 2 reflecting greater confidence in the wardens' ability to take action. However, the absence of a local authority comparator means that this must be speculative.

10.72 There has been an increase in Environmental Street Cleanliness from a baseline score of 73 to a year 2 position of 75. However, this represents a lower score than in year 1 when it stood at 83. A baseline for the local authority is not available, but the decline between years 1 and 2 was also experienced across the local authority, with the year 2 score being lower than the year 1. ²⁷ The overall direction of change is, however, positive. The scheme manager believes that the wardens' role in securing the removal of abandoned vehicles and dealing with dog fouling and lighting faults had helped to improve the streets and hence the street cleanliness indicators. This is backed up by the wardens' survey ²⁸. As such, this can be explained in terms of the Positive Intervention theory there being an overall improvement in the areas in which the wardens are working.

10.73 As with other case study schemes the picture that emerges is mixed, when the indicators are considered in isolation. However, when taken in conjunction with the positive results from the perception surveys, the conclusion is that the wardens are having a positive impact in the area.

South Lanarkshire

10.74 In South Lanarkshire 9 indicators were collected. These covered:-

• Vandalism;
• Breach of the Peace;
• Graffiti;
• Street Drinking offences;
• Fly Tipping;
• Abandoned Vehicle Fires;
• Refuse Container Fires;
• Vandalism to Fire Hydrants; and
• Racial Harassment.

They were collected for the 4 warden areas: East Kilbride, Cambuslang and Rutherglen, Hamilton and Clydesdale. The 4 areas cover the entire local authority. This means that there is no wider local authority comparator for this scheme.

10.75 The analysis concentrates on the first 5 of the indicators. This is because the other 4 (abandoned vehicle fires, refuse container fires, vandalism to fire hydrants and racial harassment) have low baseline figures. Given this, we did not feel that it was appropriate to assess the change in each of the 4 warden areas, as small absolute changes can result in apparently significant percentage movements.

10.76 Vandalism accounts for the single largest number of crimes in the South Lanarkshire warden scheme areas. Overall, the number of crimes has increased from a baseline of 5,446 to 6,443 in year 2: an 18% increase on the baseline but a fall of 8% from year 1.

10.77 There was a broad increase in the number of vandalism incidents across the 4 wardens' areas (Chart 10.6), although some exhibited greater change than others, in particular:-

• There was a 70% increase in Clydesdale, with most of this taking place between the baseline and year 1;
• There was a 13% increase in Hamilton between the baseline and year 2. However, this hides an 8% decline between years 1 and 2;
• There was a 10% increase in Cambuslang and Rutherglen., although this again masks an 8% reduction between years 1 and 2; and
• There was a 4% increase in East Kilbride. However, this has to be set against a 17% decline between years 1 and 2.

CHART 10.6 Changes in Reported Occurrences of Vandalism in South Lanarkshire

10.78 These trends suggest that the Rise and Fall (Confidence) theory could apply, with most warden areas experiencing an increase in reported incidents in the first year followed by a reduction between years 1 and 2. The scheme's manager believes that the proactive approach to vandalism has driven these changes. The schools initiative provides a good example of this approach. The wardens visited 51 schools regularly during the holiday periods to identify issues such as vandalism and take appropriate action. A total of 7,165 visits were carried out in the 24 month period with a number of incidents being reported to the police. Interestingly perceptions of vandalism as a problem have increased, albeit only by 5% (Table 12.4). However, the lack of comparator data means that interpretation must be speculative and it is probably wisest to draw no conclusions.

10.79 The number of Breach of the Peace offences increased from a baseline of 4,043 to a year 2 figure of 4,872: an increase of 21%, though there was a 4% decline between years 1 and 2.

10.80 There were broadly similar increases in each of the 4 areas between the baseline and year 2 (Chart 10.7). For example:-

• There was a 31% increase in offences in East Kilbride. However, this hides a fall in the number of offences between years 1 and 2;
• There was a 23% increase in Clydesdale between the baseline and year 2;
• There was a 17% increase in Cambuslang and Rutherglen, although most of this was between the baseline and year 1; and
• There was a 15% increase in Hamilton between the baseline and year 2. However, the overall change hides a 9% decline between years 1 and 2.

CHART 10.7 Changes in Reported Breach of the Peace Offences in South Lanarkshire

10.81 These trends suggest that the Rise and Fall (Confidence) theory might apply, as the broad trend is for there to be a rapid increase in the number of offences in the first year followed by a gentler decline over time. Consultations with the scheme manager suggested that, whilst there has been an overall increase, the subsequent decline might be evidence that the wardens are now beginning to have an impact. This reflects the introduction of a number of targeted initiatives. For example, the wardens are working with the police to curb youth drinking, with the wardens identifying outdoor drinking dens that are then raided by the police. However, the statistical pattern could also indicate that the Rise and Fall (Misplaced Confidence) theory was applicable, especially as the surveys show that perceptions of the incidence of the types of antisocial behaviours most closely aligned to Breach of the Peace (noisy neighbours, vandalism and intimidation) had all increased (Table 12.4). This was, however, countered to some extent by the 63% increase in residents thinking that the area was a "Very Good" place to live (from 24 to 39 respondents). Again the absence of a comparator for a wider area makes conclusions speculative so that no definitive conclusion as to impact has been drawn.

10.82 The number of reports of graffiti increased from a baseline of 2,238 to a year 2 figure of 2,677: an increase of 20%, although between years 1 and 2 there was a 21% decline.

10.83 There was a general increase in each of the 4 warden areas (Chart 10.8). For example:-

• There was a 164% increase in the number of reports of Graffiti in Clydesdale. This was, however, from a baseline of just 81 incidents. There was also a decline in the number of incidents between year 1 and year 2;
• There was a 21% increase in East Kilbride, although most of this occurred in year 2;
• There was a 17% increase in Hamilton, which occurred after an initial fall between the baseline and year 1; and
• There was a 10% increase in Cambuslang and Rutherglen between the baseline and year 2. However, this hides a 43% decline between years 1 and 2.

CHART 10.8 Changes in Reported Occurrences of Graffiti in South Lanarkshire

10.84 The scheme manager believes that these trends have been driven by the local authority's proactive approach to graffiti removal, close links with the police and improved reporting systems. One example of this approach covers the Fernhill graffiti and vandalism initiative. The wardens assisted Strathclyde Police in cataloguing graffiti "tags" and took digital photographs which were subsequently made into intelligence packages. These allowed the Community Police to charge 3 of the most persistent offenders. However, the surveys (Table 12.4) show that perceptions of the incidence of vandalism and graffiti have increased slightly (5% between the base and endline). It may be that this is a refection of the increased publicity given to this type of behaviour so that, paradoxically, launching initiatives to tackle it results in perceptions of its frequency increasing. The theoretical models explaining the statistical changes are the Rise and Fall, (Confidence) or (Misplaced Confidence). On balance it is perhaps best not to make any theoretical attribution.

10.85 The number of Street Drinking offences reported to the police is also seen by the scheme as a key indicator of progress. Overall, there was an increase in the number of offences: from a baseline of 1,912 to a year 2 figure of 4,880: an increase of 155%, or 2,968 more incidents over a 2 year period.

10.86 There were increases on this indicator in each of the 4 areas:-

• Increases of 286% and 238% in Hamilton and East Kilbride respectively; and
• Less extreme changes in Cambuslang and Rutherglen and Clydesdale (44% and 13% respectively), but incidents have still increased between the baseline and year 2.

10.87 The scheme's manager suggested that these changes were driven by the local authority's approach to tackling youth alcohol and related antisocial behaviour. This was a broader council initiative, supported by the wardens. Through this the wardens identified "hot spots" where street and public drinking was occurring, including wooded areas and children's play areas. They then carried out environmental visual audits, recording their findings, taking photographs and arranging clean ups of the discarded bottles, cans and litter. In addition, the wardens also gathered intelligence for the Police on sources of sales of alcohol to under-aged drinkers. The wardens worked closely with the Community Police and patrolled the hot spot areas. They were also involved in developing and running diversionary activities for young people in the targeted areas (including sport activities, youth clubs, art projects and music events). The scheme manager suggested that this approach may have increased the number of reported incidents but was tackling the problem and would lead to more sustainable longer term outcomes. If this interpretation is correct then the Rise and Fall (Confidence) theory would explain the statistical changes although comparator data would be needed to support this view. Without this, firm conclusions cannot be drawn.

10.88 The number of reports of fly tipping reported to the council was also used as a measure of progress. Overall, there had been an increase from 1,511 reports to 3,072 reports between the baseline and year 2: an increase of 103%. However, the broader pattern of change reflects a substantial increase between the baseline and year 1 and then a reduction between years 1 and 2.

10.89 There was an increase in each of the 4 warden areas, though 2 in particular exhibited a substantial increase over the period:-

• There were increases of 145% and 107% in Hamilton and Cambuslang and Rutherglen; and
• There were less extreme changes in East Kilbride and Clydesdale (74% and 29% respectively), but both still showed increases between the baseline and year 2.

10.90 The pattern of statistical change would seem to reflect Rise and Fall. Given that the surveys indicate that perceptions of the prevalence of litter have decreased quite substantially (a fall of 46%) the Rise and Fall (Confidence) theory would seem to offer the best explanation. Again the absence of comparator data makes this conclusion difficult to justify.

10.91 As with most of the case studies, interpretation of the changes is open to debate. Our analysis has shown that the statistical changes for the 5 key indicators follow the Rise and Fall pattern, although the lack of comparator data makes it difficult to know if the theoretical explanation is the Confidence or Misplaced Confidence theory. There is evidence to support the Confidence theory, for example the scheme manager's views on the various initiatives that are being implemented and which have an impact upon reported levels of offences and the perception surveys ²⁹. The broad change has been an increase in the number of incidents between the baseline and year 2. However, this then masks a fall between years 1 and 2. This would seem to indicate that positive changes are starting to take place. It is, however, important to note that this is the overriding trend. Underneath this there is a picture of considerable complexity both within and across the scheme the interpretation of which is made more difficult by the absence of comparator data.

Conclusion

10.92 Analysis of changes across the case study areas is complicated. The lack of complete time series data for all indicators means that there is insufficient evidence on which to draw firm conclusions for all indicators and all schemes. There are also many other initiatives that have been implemented that impact upon recorded levels of crime and antisocial behaviour. Given this, it should be no surprise that the picture across the individual warden schemes is complex, with rises in some sub-areas being offset by falls in others. This makes drawing firm conclusions both challenging and complicated.

10.93 In order to simplify the analysis and draw out some overarching conclusions, Appendix 4, Table 4.1 summarises the conclusions reached in the earlier text. It can be seen that overall there are 39 indicators in the Table for the 9 case studies. Of these:-

• For 23 (59%) it has not been possible to draw definitive conclusions as to which of the impact theories accounts for the observed changes;
• The changes for a further 9 (23%) have been interpreted in terms of the Rise (Negative) theory, that is the increases in the reported occurrences of crimes and antisocial behaviours have been faster in the wardened areas than in the wider local authority;
• Changes in a further 3 indicators (8%) have been interpreted in terms of the Rise (Positive) theory. Here occurrences of crimes and antisocial behaviours have increased in the wardened areas but this increase has been less than in the wider local authority area;
• A further 3 (8%) have been interpreted in terms of the Positive Intervention theory. The implication of this is that the wardens' interventions have begun to have an almost immediate impact; and
• The final indicator has been interpreted in terms of the No Impact theory. Here there are no differences between the changes in the indicator in the wardened area and those in the wider local authority area.

10.94 Looking at the change in each of the case study areas:-

• In 2 areas (Dumfries and Galloway and Dundee) there is no statistical evidence that the wardens have had an impact on the reported instances of crimes and antisocial behaviours on which the schemes wanted to be evaluated;
• For Aberdeen and Edinburgh the selected indicators show some evidence of impact, which is backed up by the findings of the perception surveys;
• For 4 of the case studies (East Renfrewshire, Inverclyde, Orkney and South Lanarkshire) the lack of comparators means that definitive conclusions cannot be drawn, although:-
• • In East Renfrewshire the perception surveys show considerable improvements;
  • In Inverclyde residents feel that the areas have improved; and
  • In South Lanarkshire all of the indicators conform to the Rise and Fall pattern. Given that perceptions of neighbourhood quality have improved, and our awareness of the various initiatives that the wardens are involved in, there seems a strong possibility that the theoretical explanation for these changes is Rise and Fall (Confidence); and
• In Perth and Kinross there is clear statistical evidence that the wardens are having an impact. This is backed up by the results of the perception surveys.

10.95 It is disappointing that, for so many of the case study indicators, it has not been possible to arrive at definitive conclusions as to the impact of the wardens. The main reason for this is data inadequacies, in particular failures to collect data, especially comparators. Given this the conclusions summarised at the foot of Table 10.1 are very cautious and are based solely on the crime and antisocial behaviour statistics, which are notoriously difficult to interpret (see Paragraphs 9.21 to 9.24). At various places in the analysis other sources of evidence, such as the perception surveys and the views of scheme managers have been highlighted. Often these are far more positive about the impact of the wardens' interventions. Accordingly any final conclusions about impact need to wait until all of the evidence has been presented. Given that so much of the analysis of the indicators in the case study areas was thwarted by lack of data, in the next Chapter we extend the analysis to non-case study schemes. The focus is on the indicators that were used to monitor progress by more than one scheme and for which, therefore, more data was available.

« Previous | Contents | Next »