The Scottish Government

« Previous | Contents | Next »

Listen

Annex 3: Additional Findings from the Neighbourhood Household Survey

Introduction

This annex provides further information about the neighbourhood household survey and presents additional findings and analysis which complement the key findings from the survey presented in Chapters 5 and 6 of the main report. A copy of the household survey is presented in annex 5 of this report.

Survey Methods

The survey was conducted in June 2006 by researchers from Management Information Scotland Ltd. Residents were interviewed on their doorstep, with each interview lasting approximately 25 minutes. The majority of the survey comprised standard questions drawn from existing surveys, including: the Scottish Household Survey, the Scottish Crime Survey, the British Crime Survey and local neighbourhood perception and fear of crime surveys conducted by selected Scottish local authorities; with a few additional questions specifically developed for this research. Although the survey comprised closed response options, the survey team also noted down any comments that residents made and also made their own assessment of neighbourhood conditions as they were conducting the survey. These residents' and survey team members' comments have informed the qualitative findings presented in our main report. The survey was conducted in June in order to provide a balance between the lower rates of antisocial behaviour typically experienced in the winter months with shorter hours of daylight and the higher rates of antisocial behaviour usually associated with the summer months of longer daylight and the long schools holidays.

The survey sample was generated by utilising the neighbourhoods constructed using Scottish Neighbourhoods Statistics data zones (see Annex 1). The boundaries of these neighbourhoods were then further refined using on the ground and map analysis to remove anomalies (for example, where a data zone contained one area of settlement which was a considerable distance from the rest of the neighbourhood). The survey team were provided with a map of the area and a list of streets and survey team members then randomly selected addresses. Attempts were made to ensure that the survey sample was drawn from all streets in the neighbourhood. Visits to the neighbourhoods were conducted in the daytime and the evenings to ensure that residents in employment were included. However, no attempt was made to ensure representativeness on other dimensions such as age and gender. The target of 200 completed interviews was achieved in all eight neighbourhoods.

The total survey sample consisted of 1613 interviews, with just over 12 per cent (200 interviews) in each of the eight case study neighbourhoods. Questions were asked under the headings:

• Personal and household details
• The neighbourhood and neighbours
• Neighbourhood problems ( i.e. antisocial behaviour)
• Personal experiences of neighbourhood problems and agency responses
• Feelings of safety

Some of the survey findings are presented as basic percentages. In addition, further analysis was undertaken, utilising logistic regression techniques to identify the extent to which different factors helped explain why and how one variable was associated with another. This technique is useful as it allows a number of underlying variables, such as age, to be taken into account when calculating the extent to which other factors, for example educational attainment, may be associated with antisocial behaviour in an area. Results can be presented as a series of Odds Ratios. Odds Ratios reflect the probability of a person being in one group rather than another after all other factors in the model have been taken into account. For example an Odds Ratio of 2 means a person with a known attribute, for example, living in social housing, is, on average twice as likely to have witnessed an act of antisocial behaviour, compared with those who do not, after all other factors have been taken into account. Adjusted logistic regression models in this report have been adjusted for: sex, age, ethnicity, tenure, household composition and an in-employment indicator.

Characteristics of the Survey Respondents

Figure A3.1 shows the personal and household characteristics of respondents by neighbourhood. The key characteristics of the survey sample are summarised below.

• In total 59% of respondents were female, 41% were male.
• For all neighbourhoods 54% of respondents were aged 45 or over. Only six percent were under-25; ranging from one percent in Carfin to 11 per cent in Muirhouse.
• 28 per cent of respondents were in either couples or couples with children households; 27 per cent were in single households and 15 per cent were single with children.
• 39 per cent were in employment, of which 28 per cent were in full-time employment. Seven percent were unemployed and ten percent were not working due to disability or ill health.
• In total 98 per cent of respondents were white Scottish/Irish/British. Only one per cent of respondents were non-white. This prevented analysis of any differential experience of antisocial behaviour by respondents from a black and minority ethnic background.
• 36 per cent of all respondents lived in either detached houses/bungalows or semi-detached housing; 45 per cent lived in flats/maisonettes
• In total 38 per cent of respondents were owner-occupiers; ranging from 20 per cent in Muirhouse to 67 per cent in Carfin. The majority of survey respondents were in social housing in all neighbourhoods except Carfin
• Overall the neighbourhoods had fairly stable populations with 66 per cent of all respondents living in the neighbourhood for over five years and 48 per cent living in the neighbourhood for more than 10 years.

Due to the sampling techniques used in this study, the survey sample is not accurately representative of either the wider population of the case study neighbourhoods or the Scottish population as a whole. Caution should therefore be taken in interpreting the results presented here and in the main part of the report.

Table A3.1: Characteristics of the survey respondents

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

Perceptions of Local Neighbourhoods, Social Relations and Personal Safety

This section presents the survey findings about respondents' perceptions of their neighbourhood and their neighbours, including whether respondents perceived their neighbourhood to be: a good place to live, to have good aspects, to have good social relations, and to be a safe place.

In total 78 percent of survey respondents rated their neighbourhood as a good place to live (Figure A3.1). However there was considerable variation between the neighbourhoods. In both of the North Lanarkshire and Scottish Borders neighbourhoods almost nine in ten respondents believed that their neighbourhoods were good places to live. In contrast, less than three quarters of residents in the Edinburgh and Fife neighbourhoods felt this to be the case. Just over one half of surveyed residents in Muirhouse (57 per cent) and two thirds (67 percent) of surveyed residents in Methil reported that their neighbourhood was a good place to live.

Figure A3.1: Percentage of respondents who think their neighbourhood is a good place to live

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

Two sets of questions in the survey were used to explore residents' perceptions about 'aspects of neighbourhood' and 'social relations'. An aspects of neighbourhood score was developed, which encompassed residents' views on: the manner in which streets and public spaces were maintained, relations with neighbours, sense of community, shops and facilities, and facilities for children and young people. Respondents with an aggregate 'positive' score, perceived that overall these aspects of their neighbourhood were good. Figure A3. 2 shows that, in total 45 per cent of respondents reported an overall 'good' aspects of neighbourhood score. As may be expected from the results shown in Figure A3.1 above, residents in the North Lanarkshire and Scottish Borders neighbourhoods were more likely to have a 'positive' aspects of neighbourhood score than residents in the Edinburgh and Fife neighbourhoods. Again, Muirhouse (14 per cent) and Methil (28 per cent), had the lowest proportions of residents reporting overall positive aspects to their neighbourhood.

Figure A3.2: Percentage of respondents with a 'good' aspects of neighbourhood score

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

A social relations score was constructed using analysis of the following three questions from the survey which asked respondents whether they agreed or disagreed with the following statements:

• If I were alone and needed help I could rely on friends/relatives in this neighbourhood to help me.
• If my home was empty, I could count on friends/relatives in this neighbourhood to keep an eye on my home.
• I feel I could turn to my friends/relatives in this neighbourhood for advice or support.

Respondents with a 'good' social relations score overall agreed with these statements. In total 77 per cent of survey respondents had a 'good' social relations score (Figure A3.3). The findings suggest relatively strong social relations within the case study neighbourhoods. In both North Lanarkshire neighbourhoods over 90 per cent of respondents had a 'good' social relations score whereas in Edinburgh, social relations appeared to be less robust, with 68 per cent of residents in Broomhouse and only a half of residents in Muirhouse reporting an overall 'good' social relations score.

Figure A3.3: Percentage of respondents with a 'good' social relations score

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

Residents' perceptions of safety are discussed in Chapter 5 of the main report. In addition to the findings presented there, survey responses were used to construct a feelings of safety score for each respondent. This score aggregated individual residents' perceptions of safety in different scenarios (in the home or walking in the neighbourhood during daylight or at night). Respondents with an 'unsafe' feelings of safety score overall would feel unsafe in the four scenarios. Across the entire survey sample, only six per cent of respondents had an overall 'unsafe' feelings of safety score (Figure A3.4). However, there was some variation between neighbourhoods, with 14 per cent of residents in Muirhouse generally feeling unsafe compared to no residents in Carfin.

Figure A3.4: Percentage of respondents with an 'unsafe' feelings of safety score

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

Willingness to Report Incidents of Antisocial Behaviour and to Act as a Witness

Chapter 6 of the main report identifies the significant problem of the under-reporting of antisocial behaviour in the case study neighbourhoods, which mirrors similar findings at the national level from the Scottish Household Surveys and British Crime Surveys. Our survey asked residents about how likely they would be report an incident of vandalism or graffiti to local agencies. Table A3.3 shows that a majority of residents would be likely to report such an incident. However, this varied from over nine in ten residents in Carfin to two thirds of residents in Muirhouse and Methil. The police are the local agency most reports would be directed to, followed by the Council. The impact of community wardens is most discernible in Langlee in Galashiels (Scottish Borders) and Abbeyview in Dunfermline (Fife) where a quarter and a fifth of residents respectively would report the incident to this service. These findings reveal a considerable discrepancy between residents' apparent potential willingness to report incidents and the actual levels of reporting that occur within neighbourhoods.

Table A3.3: Likelihood of residents reporting an incident of vandalism or graffiti to agencies (percentages)

Base: Likely to report: All (1613). Given likely to report; Broomhouse (153), Muirhouse (136), Abbeyview (162), Methil (137), Carfin (186, Whinhall (170), Burnfoot (162), Langlee (154).
Source: Management Information Scotland Household Survey June 2006.

The survey also asked respondents about their willingness to act as a witness in a case involving serious antisocial behaviour (Table A3.4). Just under half of the survey sample (47 per cent) stated that they would be willing to act as a witness in a case of serious antisocial behaviour. This varied from over two thirds of respondents in Carfin and Whinhall to less than a third of respondents in Muirhouse. The difference between the levels of willingness to report an incident and to act as a witness were reflected in some of the challenges facing local agencies in investigating and taking action against antisocial behaviour which are discussed in Chapters 4, 5 and 6 of the main report.

Table A3. 4: Respondents' willing to act as a witness (percentages)

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

Residents' Satisfaction with the Responses of Local Agencies to Antisocial Behaviour

Chapters 5 and 6 of the main report discuss residents' satisfaction with the responses of local agencies to antisocial behaviour in the case study neighbourhoods. This analysis is based on two questions in the household survey:

• to what extent are you satisfied with what local agencies are doing to tackle antisocial behaviour in this area, and
• how has the performance of local agencies changed in tackling antisocial behaviour in the last 12 months?

Almost a third (32 per cent) of all respondents claimed that they were dissatisfied with what local agencies were doing to tackle anti-social behaviour (Table A3.5). This percentage was highest in the Edinburgh neighbourhoods and lowest in the North Lanarkshire neighbourhoods. Only one in five respondents were dissatisfied in Whinhall compared to nearly half of respondents in Muirhouse. A quarter (24 per cent) of all respondents thought that the performance of local agencies in tackling anti-social behaviour had improved in the previous 12 months. Respondents in Whinhall (34 per cent), Carfin (32 per cent) and Langlee (31 per cent) were most likely to report an improvement. In contrast only six percent of respondents in Muirhouse perceived agency performance to have improved in the previous 12 months.

Table A3.5: Residents' satisfaction with the performance of local agencies

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

Relationships between Indicators

Analysis was conducted of the key relationships between variables in the household survey, including: perceptions and experiences of antisocial behaviour, aspects of the neighbourhood and perceptions/experiences of antisocial behaviour, fear of crime and perceptions/experiences of antisocial behaviour, reporting by types of antisocial behaviour experienced, and reporting and agency performance.

Was there a relationship between witnessing antisocial behaviour and perceiving antisocial behaviour to be common?

It is known that there will be a range of influences on perceptions of antisocial behaviour, only some of which relate to individuals' actual direct experiences of the problem. Figure A3.5 shows the percentage of those survey respondents reporting that antisocial behaviour is 'very' or 'quite' common in their neighbourhood who had actually witnessed an act of antisocial behaviour. 92 per cent of the respondents who perceived that antisocial behaviour was common had witnessed at least one of the eight forms of antisocial behaviour listed in the survey. However this percentage differs for the eight types of anti-social behaviour listed. 70 per cent of those residents who felt that noisy neighbours were common in their neighbourhood had experienced such a situation, whilst only 30 per cent of those that perceived drugs-related problems to be common had actually personally experienced them.

Figure A3.5: Percentages of respondents who believe a type of antisocial behaviour is common who have experienced that type of antisocial behaviour

Base: All respondents who perceived an act of antisocial behaviour to be 'very' or 'quite' common; Overall (255), Noisy neighbour (235), Rowdy behaviour (572), Vandalism (478), Harassing (493), Rubbish (497), Neighbour disputes (138), Setting fires/Burnt out cars (110) Drugs (383).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Note: The 'Overall' column refers to the percentage of respondents who perceive antisocial behaviour to be 'very' or 'quite' common and have personally experienced at least one type of antisocial behaviour.

The adjusted odds ratios in Table A3.6 show that respondents who had witnessed an act of antisocial behaviour were significantly more likely to perceive that overall acts of antisocial behaviour were common in their neighbourhood. For example, those respondents who had witnessed noisy neighbours were around 50-times more likely to perceive that this was a problem in their neighbourhood compared to those respondents that had not witnessed noisy neighbours. Respondents who had witnessed at least one of the forms of antisocial behaviour listed were more likely, with an odds ratio of 7, to feel overall that anti-social behaviour was common in their neighbourhood than those who had not witnessed one of the acts of antisocial behaviour.

Table A3.6: Adjusted odds ratios: Perceptions and experiences of antisocial behaviour

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

Correlations between neighbourhood percentages of those respondents who had witnessed an act of antisocial behaviour and those respondents who perceived it to be common are presented in Table A3.7. At a neighbourhood level there was a significant positive relationship between the percentage of respondents in a neighbourhood that perceived overall antisocial behaviour to be common and the percentage of respondents that had experienced at least one of the types of anti-social behaviour types listed. Breaking this down by the eight types of antisocial behaviour:

• significant at 0.01 level were: noisy neighbours, vandalism, rubbish, and setting fires/burnt out cars,
• significant at 0.05 level were: neighbour disputes, harassing, and drugs, and
• a positive but not significant correlation: rowdy behaviour.

Squaring the correlation coefficient gives the r squared statistic or proportion of explained variation. This statistic gives the variation in one variable that is explained by variation in the other. 92 per cent of the variation in the neighbourhood percentage who perceived noisy neighbours to be common was explained by the variation in the neighbourhood percentage of respondents who had witnessed noisy neighbours. However, only 58 per cent of the variation in the neighbourhood percentage who perceived drugs to be common was explained by variation in the percentage of respondents who had witnessed drugs problems.

Table A3.7: Correlation between residents perceiving types of antisocial behaviour as common and personally experiencing types of antisocial behaviour

Base: All (1613).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Bold Indicates significant at the 0.05 level (two-tailed).
Bold and * indicates significant at the 0.01 level (two-tailed).
Note: 'Overall ASB' represents respondents who perceive antisocial behaviour to be common and have experienced personally experienced at least one type of antisocial behaviour.

What was the relationship between aspects of the neighbourhood and perceived and experienced antisocial behaviour?

Table A3.8 presents the relationships between:

• The impact of 'aspects of neighbourhood', encompassing: the way streets and public spaces are maintained; relations between neighbours; a sense of community; local shops and facilities; facilities for children (aged up to 11); and facilities for young people (aged 12 to 18) on experiences/perceptions of antisocial behaviour;
• The association between 'good social relations' and experienced/perceived antisocial behaviour; and
• The association between perceiving the neighbourhood as a good place to live and experienced/perceived antisocial behaviour.

Table A3.8: Adjusted odd ratios: Neighbourhood well-being and experience/perceptions of antisocial behaviour

Base: All (1613).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Bold indicates significant at the 0.05 level.
(1) 'Other' includes negative neighbourhood scores and 'don't know' responses.

We would expect that perceiving antisocial behaviour to be common or directly experiencing antisocial behaviour would have a detrimental effect on feelings that a neighbourhood is a good place to live. Table A3.8 confirms that respondents who perceived that their neighbourhood was a good place to live were significantly less likely to have witnessed an act of antisocial behaviour (odds ratio of 0.21) and to perceive antisocial behaviour to be common (odds ratio of 0.17). There is an interesting relationship between residents' feeling that aspects of their neighbourhood were good and their experience and perception of antisocial behaviour. Whilst both relationships were significant, residents who perceived that aspects of their neighbourhood were good overall were 68 per cent less likely to perceive antisocial behaviour as a problem, but only 29 per cent less likely to have experienced antisocial behaviour: Perceptions of good aspects of neighbourhood were therefore associated to some extent with reduced experience of antisocial behaviour but more so with reduced perceptions of antisocial behaviour as a neighbourhood problem. Table A3.8 also presents the relationship between a respondent's social relations score and witnessing and/or experiencing antisocial behaviour. Respondents who had a 'good' social relations score were about a third less likely have experienced antisocial behaviour and/or to perceive it to be common in their neighbourhood.

What was the relationship between fear of crime and perceptions/experience of antisocial behaviour?

Analysis of the household survey reveals an association between a respondent feeling 'overall' unsafe in their neighbourhood and their experience and/or perception of antisocial behaviour in their neighbourhood (Table A3.9).

Table A3.9: Adjusted odds ratios: Feelings of safety and experience/perceptions of anti-social behaviour

Base: All (1613).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Bold indicates significant at the 0.05 level.
(1) includes negative feelings of safety scores and 'don't know' responses.

Table A3.9 shows a strong and significant relationship between feelings of safety and experience/perceptions of antisocial behaviour. Residents who felt unsafe were over six and a half times more likely to have experienced antisocial behaviour and seven and a half times more likely to perceive antisocial behaviour to be common in their neighbourhood.

Table A3.10 explores this relationship further by looking at the four components of our constructed feelings of safety score. Feeling unsafe, either at home or walking alone in the neighbourhood, during the day was not statistically related to either experience or perceptions of antisocial behaviour. On the other hand, feelings of safety after dark were strongly related to experiences and perceptions of antisocial behaviour. For example, residents who indicated feeling unsafe in their own home after dark were almost six times more likely to perceive antisocial behaviour to be common in their neighbourhood and over eight and a half times more likely to have experienced antisocial behaviour.

Table A3.10: Adjusted odds ratios: Feelings of safety and experience/perceptions of anti-social behaviour in different situations

Base: All (1613).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Bold indicates significant at the 0.05 level.
(1) 'Other' includes responses indicating positive feelings of safety and 'don't know' responses.

The impact of respondents' perceptions of antisocial behaviour upon their feelings of safety within the neighbourhood is shown in Figure A3.6. 29 per cent of survey respondents stated that the eight types of antisocial behaviour listed had at least some impact on their feelings of safety when they were in their neighbourhood. Rowdy behaviour (26 per cent) and harassment (23 per cent) had the highest percentages of respondents stating that these types of antisocial behaviour impacted on their feelings of safety. Only seven percent of respondents felt that neighbourhood disputes or setting fires/burnt out cars impacted on their feelings of safety.

Figure A3.6: Percentage of residents believing that anti-social behaviour has an impact on their feelings of safety, by type of antisocial behaviour

Base: All (1613).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Note: The 'Overall' column represents the proportion of residents who stated that the overall impact of anti-social behaviour in their neighbourhoods had 'some' or 'a lot' of impact on their feelings of safety in their neighbourhood.

Were those respondents who would report antisocial behaviour also those who perceived that local agencies had improved in performance or that performance was satisfactory?

The adjusted odds ratio Table A3.11 allows us to explore two relationships:

• to what extent were respondents satisfied or dissatisfied with what local agencies (such as the council, police and social landlords) were doing to tackle antisocial behaviour in this area?, and
• how had the performance of local agencies (such as the Council, police and landlords) changed in tackling antisocial behaviour in the last 12 months?

Those survey respondents who indicated that they were dissatisfied with local agencies were significantly less likely to be willing to act as a witness (odds ratio of 0.78) or to report an incident of antisocial behaviour to the police (odds ratio of 0.68) and/or council (odds ratio of 0.74). However, they were over two times more likely to report an incident to their neighbourhood watch or residents groups (although the numbers here are very small). Respondents who thought that the performance of local agencies had improved were also significantly more likely to act as a witness. They were also significantly more likely to report an incident of antisocial behaviour to community wardens (odds ration of 4.16) and/or a neighbourhood watch scheme (odds ratio of 2.85).

Table A3.11: Adjusted odds ratios: Willingness to report an incident/be a witness and satisfaction with local agencies

Base: All (1613).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Bold indicates significant at the 0.05 level.
(1) 'Other' includes satisfied, neutral or don't know responses.
(2) 'Other' includes negative, neutral or 'don't know' responses.

Were those respondents who perceived that antisocial behaviour had improved also those who perceived that local agency performance has improved?

Table A3.12 indicates that respondents who reported that antisocial behaviour problems in their neighbourhood had got better in the last 12 months were almost three and a half times more likely to feel that the performance of local agencies had also got better in the last 12 months.

Table A3.12: Adjusted odds ratios: Improvement in antisocial behaviour and agency performance in the last 12 months

Base: All lived in the area for at least 12 months (1476).
Source: Management Information Scotland Neighbourhood Survey June 2006.
Bold indicates significant at the 0.05 level.
(1) 'Other' includes negative and 'don't know' responses.

Residents' Satisfaction with Agency Responses to Different Types of Antisocial Behaviour

Residents' perceptions of agency performance are discussed in Chapters 5 and 6 of the main report. An additional analysis of the whole household survey sample found the following key findings.

• Councils had notably high levels of satisfaction in their response to reports of vandalism (63 per cent satisfied) and litter (100 per cent satisfied). However, less than a fifth of respondents were satisfied with the Councils' response to reports of rowdy behaviour (17 per cent) and neighbour disputes (14 per cent).
• Over four in ten respondents were satisfied with the police response to reports of: vandalism (48 per cent); noisy neighbours (46 per cent); setting fires/burnt out cars (45 percent); and rowdy behaviour (43 percent). However only three per cent of respondents were satisfied with the police response to litter complaints.
• The majority of respondents (58 percent) were satisfied with a Housing Association response to vandalism.

« Previous | Contents | Next »