THE 2000 SCOTTISH CRIME SURVEY: ANALYSIS OF THE ETHNIC MINORITY BOOSTER SAMPLE
1 INTRODUCTION
The Scottish Crime Survey
1.1 Crime surveys provide a picture of the extent of most types of personal and household victimisation, and enable issues related to crime, such as its impact on victims, anxieties and concerns about crime, and attitudes towards the police, to be explored. Crime surveys are an important complement to the statistics on crime compiled by the police, as not all crimes are reported to the police and, of those that are, not all are subsequently recorded.
1.2 The 2000 Scottish Crime Survey (SCS) is the latest in a series of surveys of public experiences and perceptions of crime that have been carried out in Scotland. In 1982 and 1988 respondents from Central and Southern Scotland were included in sweeps of the British Crime Survey (BCS). Subsequently, separate Scottish Crime Surveys were run in 1993, 1996 and 2000, covering the whole of mainland Scotland plus the larger islands. The 2000 SCS is, therefore, the fifth such survey to be carried out in Scotland. It is, however, the first to include an ethnic minority booster sample in addition to the main sample.
The Scottish Crime Survey ethnic minority booster
1.3 The purpose of a booster sample is to obtain information on a larger minority population than would be available from analysis of the main sample which contains minority respondents only in proportion to their representation in the total population. One of the difficulties of drawing an ethnic booster sample is identifying members of ethnic minorities and where they live. The main sample is drawn by approaching addresses in 334 specified sampling points throughout Scotland. Adults are selected at random within each household to ensure that the sample is representative of the total population. This approach would not work for an ethnic booster in Scotland, however, as it would be prohibitively expensive to send interviewers out at random in search of members of such a comparatively small population.
1.4 A major concern in achieving the main sample was representativeness and comparability with previous sweeps of the SCS and with other crime surveys like the BCS. The biggest difficulty in achieving an ethnic booster is in obtaining a sufficiently large sample size so that relatively reliable data can be obtained. The system of ethnic boosting used in the BCS (focused enumeration, where addresses two doors to the left and right of the core sample address are screened) is not suitable for use in Scotland because there are too few areas with a high concentration of ethnic minority population to make the method practicable.
1.5 Because of these difficulties and the associated costs, the possibility of drawing a representative ethnic minority sample had been discounted in the planning of previous sweeps of the SCS. Given the increasing public concern over racially motivated crime, however, it was decided that for the 2000 SCS an attempt would be made to arrive at an ethnic minority sample which might be large enough and sufficiently representative to allow for separate analyses by ethnic background.
Objectives of the ethnic minority booster sample
1.6 As this was the first time an ethnic minority booster had been tried for the SCS, comparability was far less of an issue. Although a previous crime survey contrasting the attitudes and experiences of white and ethnic minority respondents in the former Strathclyde Region of Scotland was carried out in 1995 (Ditton et al, 1998 1), it was not designed to provide estimates of rates of crimes experiences and households and fewer (and different) questions were asked of the samples than in the SCS. Although some of the findings in this report may seem broadly similar to those in the Ditton study, direct comparisons are invalid because of the different methods used.
How the ethnic minority booster sample was selected
1.7 The ethnic minority booster sample for the SCS was drawn by searching for appropriate-looking names in the 1999 Electoral Register for the 334 sample points used for the main sample and the two pseudo-enumeration districts closest to each of these sample points. Using information from each area, addresses were screened into the sample on the basis of the surname of the occupant listed in the electoral register. The selected sample was then checked against the main sample addresses to ensure no duplication had occurred. Given the limited information available on Scotland's ethnic minority population (dating from the 1991 Census 2), it was believed that the electoral register was the only viable sampling frame.
1.8 The ethnic minority booster sampling frame consisted of 782 addresses selected from 192 of the 334 sample points used in the main sample and their adjoining areas. The other sample points did not yield any suitable names. Due to the distribution of the ethnic minority population, the selected addresses were mostly located in urban, lowland areas of Scotland. All regions represented in the main sample were, however, represented in the ethnic booster with the exception of Dumfries & Galloway, Orkney and Shetland and the Western Isles.
1.9 There are a number of benefits and drawbacks to consider when using electoral register information as a sampling frame. It is vulnerable to under-registration and the information is likely to become progressively out of date the older the register, although even newly updated registers are by no means foolproof. The method used to select addresses also meant that only those persons with 'ethnic-sounding' surnames were selected. This method tends to exclude from the sample many African-Caribbean residents and any non-white residents who, because of marriage or parentage do not have an 'ethnic-sounding' surname. However, the main ethnic groups in Scotland (Asian and Chinese) would normally be identified by this method.
1.10 Ethnic minority addresses were given to interviewers employed by MVA, the fieldwork contractors, in addition to their allocation of addresses for the main sample when any were located within their assigned sample points and adjacent areas. Initially, interviewers were assigned a maximum of eight ethnic minority booster addresses per main sample point. The interviewers were asked to obtain as many ethnic minority interviews as possible from their addresses, and were required to return to an address at least three times to contact and interview the selected person. 3 Additional screening was required for the booster to ensure that ethnic minorities were resident in the sampled household and that the person selected from that household was of non-white ethnic origin. However, the selection of respondents for both the Adult Self-Completion and Young Persons Self-Completion questionnaires was similar to the method used for the white sample.
1.11 The main sample for the SCS was achieved by May 2000. Fieldwork for the ethnic minority booster sample, however, ran on until August 2000 and, in the event, achieved a sample of only 380 interviews from the in-scope addresses. The target sample was 450 (plus an estimated 50 from the main survey). The delay in achieving the ethnic booster was possibly exacerbated by its dependence on a sampling frame that was, latterly, several months out of date. The shortfall in the number of successful interviews is likely to be due to the data inaccuracies of the electoral register. (At 19% of sampled addresses, adults of a non-white ethnic origin were no longer resident.) Of those who were contacted and were eligible, 69.2% agreed to take part in comparison with % for the main sample. Further details of issues surrounding the conduct of the survey are given at Annex A.
Components of the Scottish Crime Survey
1.12 Most of the material from which the analysis in this report is drawn comes from two components of the SCS: The main questionnaire and the Victim Forms. To enable a wider range of questions to be asked and prevent the interview process taking too long, the main questionnaire comprises a core set of questions asked of everyone, along with two separate follow-up sections (A and B), each of which is asked of approximately half the sample.
1.13 The main questionnaire has screening questions that enable the interviewer to determine whether respondents have been the victim of any crime in the last year. It also includes a range of questions about the respondent's own circumstances and the household that allow analysis by various socio-economic factors.
1.14 The follow-up A questionnaire asks respondents about their experience of contacting or being stopped by the police, and their attitudes towards the police and other criminal justice agencies in Scotland. The follow-up B questionnaire asks respondents about their concerns about becoming a victim of crime and, for those who say they have been a victim of crime at some time in the past, the impact that this has had on them and their families.
1.15 In the event of people reporting more than one crime, two full Victim Forms and up to three more in summary form are completed. In addition, where victims report a series of victimisations which are of the same nature and probably perpetrated by the same people, only one form is completed for the series, to avoid repetition, but the number of serial incidents is recorded, up to a maximum of five.
1.16 An Adult Self-Completion questionnaire is given to respondents aged between 16 and 59, containing questions about awareness of drugs, experience of taking drugs and experience of domestic violence. There is also a Young Persons' Self-Completion questionnaire (covering a range of issues, including their experience of crime, policing and drugs) for persons aged between 12 and 15 resident in households identified as part of the sample.
Limitations of the data
1.17 Crime surveys cannot provide a complete picture of crime. For example, they do not collect information about crimes against public or corporate bodies, or individuals not resident in households (e.g. the homeless, prisoners, hospital patients, members of the armed forces, etc.). In addition, they do not collect information on victimless crimes such as certain driving offences and breach of vehicle construction and use regulations. Finally, they depend on respondents' willingness to participate in the survey and to remember and report their experiences accurately.
1.18 Although care was taken to make the main sample representative of individuals and households in Scotland, like any sample survey the SCS is subject to sampling error, whereby those who respond may not always accurately reflect the views and experiences of the Scottish population. Findings should not, therefore, be interpreted as exact measures, but as indicators set within margins of error.
1.19 A further difficulty arises from the fact that the bulk of the ethnic minority booster sample was drawn using a different sampling method to the main sample. Although reliance on ethnic-sounding names provided a sampling frame for Scotland's largest ethnic minority groups, it excluded ethnic minority householders with surnames that did not appear to be of non-white origin.
1.20 The smaller size of the ethnic minority booster is also problematic. Where raw numbers are very low, it is impractical to disaggregate the booster sample into sub-groups such as ethnic group, gender, age and social class, and this limits the amount of analysis that can be carried out.
1.21 The ethnic minority sample was broadly similar to the ethnic minority profile derived from the 1991 Census and was, therefore, likely to differ demographically from the white sample. In addition, it is not known accurately to what extent the size and characteristics of the Scottish population may have changed since 1991. There were, however, important differences between the ethnic minority sub-groups. While this report provides an analysis of ethnic minority views and experiences, it should be borne in mind that the group under discussion is far from being homogenous.
1.22 As a consequence, while the booster sample provides valuable information on views and experiences, any comparisons and contrasts with the white sample have to be read alongside an understanding of the important demographic differences between the two samples.
1.23 As with police recorded crime, the SCS (and the ethnic minority booster) does not claim to measure the 'true' level of crime in Scotland. However, for many types of crime, including the most prevalent, the SCS provides a better indicator of levels of victimisation than police recorded crime statistics. The findings from the SCS ethnic minority booster sample provide valuable baseline data that identify possible areas of concern which may be addressed by further, more qualitative, research.
Structure of the report
1.24 Chapter 2 describes the characteristics of the main and ethnic minority booster samples. Chapter 3 contains an overview of crime against ethnic minorities. Chapter 4 explores the impact of crime on people who said they were victims in 1999. Chapter 5 considers respondents' concerns about crime. Chapter 6 looks at respondents' experiences of contact with the police; Chapter 7 considers their attitudes towards the police and other agencies. Chapter 8 includes a limited amount of analysis of questions about respondents' awareness of and attitudes towards drugs. Finally, Chapter 9 draws together the main findings and presents conclusions.