Scottish Government

Chapter Five: Public perceptions of antisocial behaviour

5.1 The first part of this chapter describes public perceptions of the extent and nature of antisocial behaviour in local neighbourhoods. The data presented in this chapter is primarily drawn from a random sample household interview survey of 1613 residents conducted by Management Information Scotland Ltd in June 2006, and these findings are compared with national level Scottish data. The survey comprised interviews with approximately 200 residents in each of the 8 case study neighbourhoods, identified through a random selection of household addresses. This interviewee selection method means that the survey sample should not be expected to be representative of the case study neighbourhoods or the Scottish population as a whole. In particular, although surveyors did attempt to conduct some interviews in the evenings and attempted to ensure that interviews were drawn from addresses throughout the neighbourhood localities, there was an over-representation of females and older age groups. The survey includes a small sample of 25 individuals from an ethnic grouping other than White Scottish/Irish/British. A full account of the survey method, sample and additional findings from the survey is presented in Annex 3. A copy of the survey is provided in Annex 5. The second part of chapter provides some comparative context for the findings from our case study neighbourhoods by comparing our results with national figures and data from the local authority areas in which the neighbourhoods were located. This data also enables a comparison of the case study neighbourhoods with other deprived neighbourhoods in Scotland. In addition to data about perceived levels of antisocial behaviour and direct experience of types of antisocial behaviour, the chapter also presents comparative analysis of residents' feelings of personal safety and satisfaction with agency performance in addressing antisocial behaviour.

Anti-social Behaviour in the Neighbourhoods

5.2 The survey asked residents how common 8 types of antisocial behaviour ¹ were in their local neighbourhood. Table 5.1 shows the proportions of residents reporting that each type of antisocial behaviour was 'very common' or 'quite common'. A majority of respondents (52 percent) perceived rowdy behaviour to be common, whilst over 4 in 10 respondents perceived vandalism, rubbish and harassment to be common (44, 43 and 43 percent respectively). Over a third of respondents (36 percent) also perceived drug misuse or drug dealing to be common, while noisy neighbours (22 percent) and setting fires/burnt out cars (10 percent) were perceived to be less common problems. Table 5.1 also reveals significant variation in the perceived frequency of antisocial behaviour problems between the neighbourhoods. Antisocial behaviour, with the notable exception of harassment, was reported to be more common in the Edinburgh and Fife neighbourhoods. The Muirhouse and Methil neighbourhoods had the highest reported frequency of antisocial behaviour, whilst Carfin in North Lanarkshire had the lowest. The extent of variation between these neighbourhoods is illustrated by the fact that 57 per cent of respondents in Muirhouse perceived drug-related problems as common, a rate 14 times higher than in Carfin (4 per cent).

Table 5.1 - Percentage of respondents who perceive types of antisocial behaviour to be common

Base: All (1613). Source: Management Information Scotland Neighbourhood Survey, June 2006.

5.3 The variation between neighbourhoods is also evident in Figure 5.1, which presents the adjusted odds ratios for the relationship between neighbourhood and antisocial behaviour problems. Adjusted odds ratios control for other explanatory variables, such as the influence of the age and gender of residents, or the influence of housing tenure on the likelihood of perceiving or experiencing antisocial behaviour. Controlling for these other factors enables any 'neighbourhood effects' to be identified, where variations in our findings between different neighbourhoods will not be caused simply by the differences in the population characteristics between different neighbourhoods. The adjusted odd ratio analysis revealed a significant 'neighbourhood effect' particularly in the 2 Edinburgh and Fife neighbourhoods, which indicated a localised pattern of high levels of perceived levels of antisocial behaviour in these localities.

5.4 In Figure 5.1, the bars in black indicate where there is significant variation between the actual perceived levels of antisocial behaviour and the levels of perceived antisocial behaviour that may have been expected given the population characteristics of a neighbourhood. In Carfin, there is a significant value to the left of the y axis, indicating that the reported levels of antisocial behaviour were significantly less than may be expected given the population profile of the survey respondents in this neighbourhood. In contrast, the levels of antisocial behaviour reported in Broomhouse, Muirhouse and Methil were significantly higher than may be anticipated.

Figure 5.1 - Respondents who perceive antisocial behaviour to be common (adjusted odds ratios)

Base: All (1613) Source: Management Information Scotland Neighbourhood Survey, June 2006. Black indicates significant at the 0.05 level.

5.5 Survey respondents were also asked about whether they had personally witnessed antisocial behaviour in their neighbourhood in the previous 12 months. Table 5.2 shows that 4 in 10 respondents had experienced rowdy behaviour, whilst a quarter of respondents had witnessed vandalism, harassment and rubbish. The proportions of respondents who had witnessed particular forms of antisocial behaviour were lower than the proportion of residents who felt that these problems are common in their neighbourhood. For example, only a third of respondents perceiving drug misuse or dealing to be common had personally witnessed this problem. However, the high proportion of residents reporting directly witnessing antisocial behaviour suggests that it was a real and significant problem in the case study neighbourhoods. Residents in the 4 Edinburgh and Fife neighbourhoods and Whinhall in North Lanarkshire were more likely to have personally experienced forms of antisocial behaviour. Harassment and rowdy behaviour were a particular problem in almost all of the neighbourhoods.

Table 5.2 - Percentage of respondents who have witnessed different types of antisocial behaviour in the last 12 months

Base: All (1613) Source: Management Information Scotland Neighbourhood Survey, June 2006.

5.6 However, as Figure 5.2 reveals, using adjusted odds ratios reveals that neighbourhood effects are different when applied to direct personal experience of antisocial behaviour. Again the bars in black represent statistically significant variations between the actual levels of personal experience and what may be expected given the population profile of each neighbourhood. Whilst respondents in Muirhouse and Methil report higher levels of personal experience of antisocial behaviour than would be expected, given the characteristics of the local population, this localised effect is less pronounced in Broomhouse. In Whinhall there is actually a 'positive' statistically significant neighbourhood effect, with the incidence of personally experienced antisocial behaviour lower than might be anticipated given Whinhall's population profile. In contrast to this, the extent of personal experience of antisocial behaviour in Langlee in Galashiels is higher than may be anticipated.

Figure 5.2 - Respondents witnessing antisocial behaviour in the previous 12 months (adjusted odds ratios)

Base: All (1613). Source: Management Information Scotland Neighbourhood Survey, June 2006. Black indicates significant at the 0.05 level.

Changes in Antisocial Behaviour and Agency Performance

5.7 Survey respondents were asked whether they believed that antisocial behaviour problems had improved in the previous twelve months (May 2005-May 2006). Table 5.3 shows that, across the whole sample, rowdy behaviour (25 percent), vandalism (20 percent) and groups harassing people (19 percent) were the problems that the highest percentages of residents believed had improved. Importantly, these 3 were, along with littering, perceived to be the 4 most common types of anti-social behaviour. However less than 1 in 10 residents perceived the other types of antisocial behaviour to have decreased. The most striking findings were from Broomhouse in Edinburgh, where significant proportions of residents perceived improvements across all forms of antisocial behaviour. In contrast to this finding, residents in Muirhouse and Methil were the least likely to perceive improvements.

Table 5.3 - Percentage of respondents who perceive problems have got better in the last 12 months

Base: All lived in area for 12 months or more: Carfin (195), Whinhall (197), Burnfoot (197), Langlee (198), Broomhouse (192), Muirhouse (191), Abbeyview (198), Methil (195). Source: Management Information Scotland Neighbourhood Survey, June 2006.

5.8 This pattern is demonstrated in the adjusted odd ratios presented in Figure 5.3. These show that, controlling for the characteristics of the survey respondents in each neighbourhood, residents in Broomhouse and Whinhall had experienced significant improvements in antisocial behaviour in the previous 12 months, which, our research suggests, is linked to the antisocial behaviour interventions being undertaken in these neighbourhoods. In contrast, residents in Methil and Muirhouse were significantly less likely to report improvements in the levels of antisocial behaviour in the pervious 12 months.

Figure 5.3 - Respondents who feel that antisocial behaviour problems have improved in the last 12 months (adjusted odds ratios)

Base: All lived in area for 12 months or more: Carfin (195), Whinhall (197), Burnfoot (197), Langlee (198), Broomhouse (192), Muirhouse (191), Abbeyview (198), Methil (195).
Source: Management Information Scotland Household Survey June 2006. Black indicates significant at the 0.05 level.

5.9 A third of surveyed residents were dissatisfied with what local agencies were doing to tackle antisocial behaviour in their area (Table 5.4). Residents in the 2 Edinburgh neighbourhoods were most likely to be dissatisfied, with almost half of residents in Muirhouse being dissatisfied. In contrast, residents in the 2 North Lanarkshire neighbourhoods and Langlee in Galashiels were less likely to be dissatisfied with local agency performance. A quarter of residents perceived that local agencies were performing better in tackling antisocial behaviour over the previous 12 months. This response appears directly linked to rates of dissatisfaction, with residents in the North Lanarkshire neighbourhoods and Langlee, Galashiels most likely to perceive an improvement in agency performance and those in the 2 Edinburgh neighbourhoods least likely to perceive an improvement. It is clear that the apparently positive impact of agencies and interventions in Broomhouse does not yet appear to translate into improved resident satisfaction with agencies' efforts to tackle antisocial behaviour, although (as discussed below) qualitative data did show that some residents were aware of an improvement.

Table 5.4 - Satisfaction with local agencies

Base: All (1613). Source: Management Information Scotland Neighbourhood Survey June 2006.

5.10 The survey asked respondents who had reported an incident of antisocial behaviour to a local agency whether they were satisfied with the response that they received. Figure 5.4 shows that, with the exception of Council responses to litter and vandalism and housing association responses to vandalism, less than half of complainants were satisfied with the response that they received. The levels of satisfaction vary between agencies and across different types of antisocial behaviour complaints. Agency responses to drugs and neighbourhood disputes were least likely to satisfy complainants. The sample size was too small to enable a statistically robust comparison between the neighbourhoods.

Figure 5.4 - Percentage of residents satisfied with agency response by type of antisocial behaviour

Base: Those reporting antisocial behaviour (Vandalism to police: 148; Rowdy behaviour to police: 139; Noisy neighbours to police: 103; Harassment to police; 102; Litter to police: 58; Neighbour disputes to police: 35; Drugs to police: 29; Setting fires to police: 20; Noisy neighbours to council: 81; Rowdy behaviour to council: 66; Harassment to council: 56; Neighbour disputes to council: 44; Vandalism to council: 40; Drugs to council: 25; Litter to council: 17; Setting fires to council: 15; Noisy neighbours to Housing Association: 31; Rowdy behaviour to Housing Association: 14; Vandalism to Housing Association: 12).
Source: Management Information Scotland Neighbourhood Survey, June 2006. Note: Only charted for cases in which 10 reports of a type of antisocial behaviour have been made to a particular agency

Comparing Findings at the Neighbourhood, Local Authority and Scotland Level

5.11 In order to provide a comparative benchmark of the findings from our neighbourhood survey, we compared our survey data with data from the 2005 Scottish Household Survey ( SHS) which used the same antisocial behaviour questions as our survey, supplemented with data from the 2003/04 Scottish Household Survey and the 2003 Scottish Crime Survey, where the questions used differ slightly from those in the survey conducted on our behalf. This analysis provides a wider context for the examination of antisocial behaviour in the 8 case study neighbourhoods., by comparing our survey findings to national averages and the averages for the 4 case study local authorities, with additional comparative data from the most deprived neighbourhoods at national and local authority level. The analysis includes comparisons of residents' perceptions of levels and personal experience of antisocial behaviour, feelings of safety and satisfaction with agency performance in tackling antisocial behaviour.

5.12 Table 5.5 shows comparative figures for the proportion of respondents who reported types of antisocial behaviour to be 'fairly common' or 'very common' in their local neighbourhood. The proportion of residents in the SHS living in the Edinburgh, Fife and North Lanarkshire local authority areas who perceived each type of antisocial behaviour to be common was close to the Scottish average, whilst the proportion of residents in the South of Scotland (an area including the Scottish Borders) perceiving these problems as being common in their local neighbourhood was slightly less than the national average.

Table 5.5 - Percentage of SHS respondents who perceive antisocial behaviour to be either 'very' or 'fairly' common

Sources: Scottish Household Survey, 2005.

Note: In the Scottish Household Survey 2005, the Scottish Borders is combined with Dumfries and Galloway to create a Southern Scotland categorisation. The figures presented here relate to that combined locality rather than the Scottish Borders local authority area. The most deprived 10% rows represent figures for neighbourhoods in the bottom decile of the Scottish Indices of Multiple Deprivation 2004. Fife and the South of Scotland are excluded from this bottom decile analysis due to low bases.

5.13 The SHS figures also revealed that residents in the most deprived neighbourhoods in Scotland were twice as likely to report antisocial behaviour to be common in their neighbourhoods than the Scottish average, and that this finding was consistent for all types of antisocial behaviour. Table 5.6 reflects this by highlighting how the resident survey respondents in the majority of our case study neighbourhoods were significantly more likely to perceive each type of antisocial behaviour to be common compared to their SHS local authority average. For example, 62 percent of residents in Muirhouse perceived rowdy behaviour to be common compared to 17 percent of respondents across the City of Edinburgh local authority, and similar proportions were found in Methil and Fife respectively. The one exception to this pattern was the Carfin neighbourhood which had a lower or similar proportion of residents reporting 4 of the types of antisocial behaviour to be common when compared to North Lanarkshire as a whole.

Table 5.6 - Comparison of percentage of respondents at LA and neighbourhood level who perceive antisocial behaviour to be either 'very' or 'fairly' common

Sources: (1) Scottish Household Survey, 2005. (2) Management Information Scotland Neighbourhood Survey, June 2006.

Note: In the Scottish Household Survey 2005, the Scottish Borders is combined with Dumfries and Galloway to create a Southern Scotland categorisation. The figures presented here relate to that combined locality rather than the Scottish Borders local authority area.

5.14 A comparative analysis of SHS data from the most deprived neighbourhoods showed similar proportions of residents in the Edinburgh and North Lanarkshire local authority areas perceived each type of antisocial behaviour to be common, and that these proportions were close to the Scottish average for deprived neighbourhoods. Data for Fife and the South of Scotland was not released for this period by the SHS due to the small size of the relevant SHS sample. Comparing our case study neighbourhoods with other deprived neighbourhoods in their local authority areas showed that higher proportions of residents in Broomhouse and especially Muirhouse perceived antisocial behaviour to be common, with the exception of neighbour disputes. Muirhouse appeared to have a particular problem with drugs-related antisocial behaviour. In North Lanarkshire, residents in Carfin were less likely to perceive antisocial behaviour to be common, reflecting lower levels of deprivation in this neighbourhood, although there did appear to be a particular problem with harassment from groups. Residents in Whinhall were less likely than residents in other deprived neighbourhoods in North Lanarkshire to perceive noisy neighbours, vandalism, rubbish and neighbour disputes to be common, but were more likely to perceive drug misuse and rowdy behaviour to be common and almost twice as likely to perceive harassment from groups to be common.

5.15 A similar pattern emerged when analysing SHS respondents' direct personal experience of antisocial behaviour in the previous 12 months (Table 5.7 and Table 5.8). The proportions of residents in the case study local authorities reporting personal experience of antisocial behaviour were similar to the Scottish average, with the exception of the 2 Scottish Borders neighbourhoods where personal experience of antisocial behaviour was lower than the Scottish average. There were some exceptions to this pattern depending on the specific type of antisocial behaviour, with residents in Edinburgh and Fife considerably more likely than the Scottish average to have experienced noisy neighbours and rowdy behaviour. Table 5.8 also highlights very starkly the gap between the case study neighbourhoods, including the Scottish Borders neighbourhoods, and their wider local authority areas, with residents in the case study neighbourhoods being more likely - and in some cases, several times more likely - to have personally experienced each type of antisocial behaviour. The one exception to this pattern was Carfin in North Lanarkshire.

5.16 At the national level, SHS respondents living in the most deprived neighbourhoods were twice as likely as the national average to have personally experienced each type of antisocial behaviour. The proportions of residents in the most deprived neighbourhoods in the Edinburgh and North Lanarkshire local authority areas who had personally experienced antisocial behaviour were similar to the Scottish figures for the most deprived neighbourhoods, with the exceptions of higher rates of noisy neighbours in Edinburgh and lower rates of vandalism and graffiti in North Lanarkshire (data for Fife and the South of Scotland was not available due to small sample sizes). A comparison of the case study neighbourhoods in Edinburgh and North Lanarkshire with other deprived neighbourhoods in their respective local authority areas revealed a complex picture. The proportion of residents in Broomhouse and Muirhouse who had experienced either vandalism/graffiti, harassment from groups, or rowdy behaviour was far higher than the average for deprived neighbourhoods in Edinburgh. The rates of personal experience of antisocial behaviour in Muirhouse were higher than the city average for deprived neighbourhoods for all types of antisocial behaviour, and highlighted a particular problem with drugs-related antisocial behaviour. In contrast, the rates of personal experience of antisocial behaviour in Broomhouse were similar to or lower than the Edinburgh average, apart from the 3 types of antisocial behaviour identified above. In North Lanarkshire, residents in Carfin were less likely to have personally experienced all types of antisocial behaviour than the deprived neighbourhood average for North Lanarkshire, with the exception of vandalism and graffiti. The comparative findings from Whinhall were more varied. Whilst personal experience of vandalism/graffiti, harassment from groups and rowdy behaviour was more common than the average for deprived neighbourhoods in North Lanarkshire, the proportions of Whinhall residents personally experiencing noisy neighbours, rubbish, neighbour disputes and drug misuse were considerably lower than the average.

Table 5.7 - Percentage of SHS respondents who have personally experienced antisocial behaviour in the last 12 months

Sources: Scottish Household Survey, 2005.

Note: In the Scottish Household Survey 2005, the Scottish Borders is combined with Dumfries and Galloway to create a Southern Scotland categorisation. The figures presented here relate to that combined locality rather than the Scottish Borders local authority area. The most deprived 10% rows represent figures for neighbourhoods in the bottom decile of the Scottish Indices of Multiple Deprivation 2004. Fife and the South of Scotland are excluded from this bottom decile analysis due to low bases.

Table 5.8 - Comparison of percentage of respondents at LA and neighbourhood level who have personally experienced antisocial behaviour in the last 12 months

Sources: (1) Scottish Household Survey, 2005. (2) Management Information Scotland Neighbourhood Survey, June 2006.

Note: In the Scottish Household Survey 2005, the Scottish Borders is combined with Dumfries and Galloway to create a Southern Scotland categorisation. The figures presented here relate to that combined locality rather than the Scottish Borders local authority area.

5.17 Table 5.9 provides a comparative analysis of the proportions of residents who report feeling unsafe in their local neighbourhoods. Perceptions of safety when alone in the home after dark amongst residents in our case study neighbourhoods were similar to national averages. This is a particularly positive finding, given that our household survey comprised higher proportions of elderly and female respondents than the national comparator surveys. There was some variation between neighbourhoods, with 1 in 10 residents in Muirhouse reporting feeling unsafe compared to no residents in Carfin. Residents in our case study neighbourhoods were however significantly more likely than the Scottish population as a whole to feel unsafe whilst walking alone in their neighbourhood after dark, based on a comparison of data from both the 2005 Scottish Household Survey and the 2003 Scottish Crime Survey. The extent of this problem is evidenced by the fact that in 7 of the 8 case study neighbourhoods at least half of residents reported feeling unsafe when walking alone after dark in their neighbourhood and in 2 of the case study neighbourhoods (Muirhouse and Methil) approximately two thirds of residents felt unsafe.

Table 5.9 - Residents' perceptions of safety in their home and neighbourhood

Sources: Scottish Household Survey, 2005; Scottish Household Survey 2003/2004; Scottish Crime Survey, 2003; Management Information Scotland Neighbourhood Survey, June 2006.

5.18 Our case study local authorities were benchmarked with Scottish averages derived from national survey evidence. In Edinburgh, Fife and North Lanarkshire, the proportion of residents perceiving antisocial behaviour to be common in the neighbourhood, or to have personally experienced antisocial behaviour in the previous 12 months, is similar to the Scottish population as a whole. Residents in the Scottish Borders are less likely to perceive antisocial behaviour to be common. However residents in the case study neighbourhoods, with the exception of Carfin, were more likely to perceive antisocial behaviour problems as common when compared to their local authority averages. Similarly, residents in the case study neighbourhoods, again with the exception of Carfin, were more likely, and in some cases several times more likely, to have personally experienced antisocial behaviour compared to their local authority averages. There was however considerable variation in the incidence of personal experience of antisocial behaviour between the 8 case study neighbourhoods. In general, the proportion of residents in the case study neighbourhoods in Edinburgh and North Lanarkshire who perceived antisocial behaviour to be common and who had personally experienced it in the previous twelve months was also higher than the average for deprived neighbourhoods in their local authority areas, although there was some variation depending on the type of antisocial behaviour. Although the proportion of residents in the case study neighbourhoods feeling unsafe in their home after dark was similar to those found in national surveys, these residents were more likely to feel unsafe when walking alone after dark in their local neighbourhood, with at least half of respondents feeling unsafe in these circumstances.

5.19 Data from the 2005 Scottish Household Survey revealed that the levels of dissatisfaction with local agencies' responses to tackling antisocial behaviour amongst residents in Fife and North Lanarkshire were similar to the Scottish average, whilst the levels of dissatisfaction in Edinburgh and the South of Scotland were lower than national average (Table 5.10). Figures for Edinburgh and North Lanarkshire reveal greater levels of dissatisfaction in deprived neighbourhoods, which is also the case nationally (data for Fife and the South of Scotland is not available due to small sample sizes). In Edinburgh, Fife and the Scottish Borders, the proportion of residents in the case study neighbourhoods who were dissatisfied with agency responses was greater than their local authority averages. In Edinburgh, these levels of dissatisfaction were also significantly greater than the city average for deprived neighbourhoods. In contrast, the 2 North Lanarkshire neighbourhoods had levels of dissatisfaction similar to (Carfin) or below (Whinhall) their local authority average.

Table 5.10 - Comparison of dissatisfaction with local agencies' responses

Sources: (1) Scottish Household Survey, 2005. (2) Management Information Scotland Neighbourhood Survey, June 2006.

Notes: In the Scottish Household Survey 2005, the Scottish Borders is combined with Dumfries and Galloway to create a Southern Scotland categorisation. The figures presented here relate to that combined locality rather than the Scottish Borders local authority area. The most deprived 10% rows represent figures for neighbourhoods in the bottom decile of the Scottish Indices of Multiple Deprivation 2004. Fife and the South of Scotland are excluded from this bottom decile analysis due to low bases.

Chapter Summary

5.20 A significant proportion of residents in the case study neighbourhoods perceived most forms of antisocial behaviour to be common or very common. Whilst the proportions of residents who had personally experienced antisocial behaviour were lower than the proportions of residents who thought antisocial behaviour was common or very common, the findings suggest that a large number of residents had been directly affected by antisocial behaviour either as victims or witnesses. There was considerable variation between residents' perceptions and experiences of antisocial behaviour across the case study neighbourhoods, with lower proportions of residents in the Scottish Borders being affected and higher proportions in the Edinburgh and Fife neighbourhoods. Statistical analysis identifies a clear neighbourhood effect, which our research suggests is linked to social dynamics and community relations in these localities and cannot entirely be attributed to the socio-demographic profile of each neighbourhood.

5.21 Although comparison with national survey data is problematic, it does appear as if most of the case study neighbourhoods were experiencing levels of antisocial behaviour that were considerably higher than national and local authority averages, including comparator averages for the most deprived neighbourhoods in Scotland. Only a minority of residents perceived that the levels of antisocial behaviour and the performance of local agencies in their neighbourhoods had improved in the previous twelve months. With the exception of North Lanarkshire, respondents in our case study neighbourhoods appeared to be more dissatisfied with the performance of local agencies in tackling antisocial behaviour than other residents in their local authority areas.