Scottish Crime and Justice Survey: consultation

3. Part One: Overview of the Current Scottish Crime and Justice Survey & Past Changes to the Survey

To provide further context to this consultation the following pages contain information on the current Scottish Crime and Justice Survey (SCJS), and the history of crime surveying in Scotland. Please also see the 2019/20 Technical Report and Questionnaire for further details.

3.1 Overview of Current SCJS

As it stands, the SCJS is a survey of public experiences and perceptions of crime in Scotland. The 2019/20 survey is the ninth year of the SCJS, with the first being conducted in 2008/09. The survey interviews adults aged 16 or over who live at private residential addresses in Scotland.

At present, the survey aims to:

• enable people in Scotland to tell us about their experiences of, and attitudes to, a range of issues related to crime, policing and the justice system, including crime not reported to the police
• provide a valid and reliable measure of adults' experience of crime, including services provided to victims of crime
• examine trends over time in the number and nature of crimes in Scotland, providing a complementary measure of crime compared with police recorded crime statistics
• examine the varying risk and characteristics of crime for different groups of adults in the population

The statistics produced from the SCJS provide a picture of the level of crime in Scotland as SCJS respondents are asked directly about their experience of incidents which have happened to them, irrespective of whether or not they reported them to the police. Therefore, the survey provides a record of peoples' experiences of crime, which is unaffected by variations in reporting behaviour of victims or any changes in crime recording practices by the police. However, the SCJS and police recorded crime statistics should be seen as a complementary series, which together provide a more complete picture of crime than could be obtained from either series alone.

The survey also provides evidence for a number of performance targets for the public sector in Scotland, at a national and a local level, including informing progress against the Scottish Government's National Performance Framework (NPF)^[1].

The survey uses a victim form questionnaire to collect extensive details about the nature of each incident that respondents report, such as when and where it occurred and details about the offender(s) and other relevant information. This allows classification, and hence counts, of crimes in Scotland.

The SCJS collects information on incidents occurring in the previous 12 calendar months before the month in which the interview takes place. This time period is referred to as the survey 'reference period'. The reference period varies depending on the month in which the interview took place, although the reference period covers an equal length of time (12 calendar months) for each respondent.

The SCJS only collects data on incidents occurring in Scotland in the reference period. Incidents which happen abroad are not covered by the survey (termed non-valid incidents). Incidents which happened in England and Wales will be recorded in the Crime Survey for England and Wales (CSEW) where householders are resident in either England or Wales. Crimes experienced in England and Wales by people normally resident in Scotland and incidents occurring in Scotland to people who live in England and Wales will not be captured in either the SCJS or CSEW.

Incidents which meet the above criteria and which are identified as crimes within the scope of the survey are used to produce the 'all SCJS crime' statistics which are published in the 2019/20 SCJS Main Findings report.

The survey collects socio-demographic information from respondents which allows a picture to be built up about the nature of crime in Scotland and variation in experiences of victimisation among subgroups of the population. For example, the 2019/20 SCJS found that men were more likely to experience violent crime, as well as people in the 15% most deprived areas of Scotland, and those living in urban locations whilst people aged 60 and over were less likely to be victims of violent crime.

The survey collects information on a number of sensitive issues, including sexual victimisation, stalking and harassment, and partner abuse (collected via the self-completion element of the questionnaire). It also collects attitudinal information on a range of issues related to crime, policing and the justice system.

Further information about what the SCJS covers, how the survey is delivered and who is included can be found in the 2019/20 Main Findings report 'Introduction and Background to the SCJS' chapter.

3.2 Scottish Surveys Core Questions

The Scottish Surveys Core Questions (SSCQ) is an annual Official Statistics publication. From January 2012 a set of 20 core questions have been asked in the Scottish Household Survey, the Scottish Health Survey and the Scottish Crime and Justice Survey. In addition, where two or more surveys ask questions on the same topic, wording has been harmonised. Results from the three surveys on the core questions are pooled together each year to offer a larger sample size, enabling more precise and granular breakdowns of results for equality groups and at sub-national geographies. It is anticipated that the SCJS's function as one of the three core question surveys will not be affected by the re-procurement process.

3.3 History of Crime Surveys in Scotland

Prior to the 2019/20 survey, there have been 16 previous surveys of victimisation in Scotland, beginning with the 1982 and 1988 years of the British Crime Survey (BCS) coordinated by the Home Office. BCS coverage in Scotland was limited to south of the Caledonian Canal. The first independent Scotland-only survey was commissioned by the Scottish Office in 1993 under the title of the Scottish Crime Survey (SCS) and was followed by repeated surveys in 1996, 2000 and 2003. In 2004, following an external review, the survey underwent both a name change, under the title of the Scottish Crime and Victimisation Survey (SCVS), and a major methodological change, with a move away from in-home face-to-face interviewing to telephone interviewing. However, the 2006 survey returned to face-to-face interviewing after it was shown that the robustness of the data produced by the 2004 telephone survey could not be substantiated. Following the 2006 SCVS, a further review of the crime survey was carried out, which resulted in the new Scottish Crime and Justice Survey (SCJS) being launched in April 2008.

The SCJS has retained the same basic design from 2008/09 onwards. However, throughout the SCJS series of surveys, there have been a reduction in sample sizes and some small changes to the sample design in relation to clustering and stratification and the length of the fieldwork period for each survey (summarised in Table 2). Whilst the fundamental structure of the questionnaire has remained consistent, the survey questionnaire is designed to allow the rotation of questionnaire sections in and out of the survey according to the policy and research requirements of the Scottish Government and stakeholders.

After the 2010/11 survey was completed, the survey moved to a biennial frequency, with the survey conducted in 2012/13 and 2014/15 (but no surveys in 2011/12, 2013/14 or 2015/16). From the 2016/17 survey onwards, the SCJS has returned to a continuous fieldwork model, but with the achieved sample size reduced to around half that of the 2014/15 survey (11,500 down to sample target of 6,000). This means that some sections of the questionnaire and breakdowns of the data are only available on a biennial basis (e.g. when the 2018/19 and 2019/20 surveys are combined). The 2019/20 survey represents the fourth survey in this continuous series.

3.4 Changes as a Result of COVID-19

As a result of the COVID-19 pandemic and associated public health measures of social distancing and lockdown, the SCJS was suspended and no 2020/21 survey was produced. Instead, the Scottish Victimisation Telephone Survey 2020 (SVTS) was delivered. This social survey, which was designed to be nationally representative, asked people about their experiences and perceptions of crime, safety, and policing during the COVID-19 pandemic. The results are based on a sample of around 2,700 telephone interviews conducted in September and October 2020. It is important to note, however, that due to a number of methodological differences the findings from the SVTS cannot be compared to SCJS results.

3.5 Plans for 2021/22 Survey

We plan to restart interviews in November 2021 operating a knock-to-nudge approach to begin with and then moving to in-home face-to-face interviewing when it is considered safe to do so. Respondents will be given the option to complete their interview over the phone if they prefer.

Knock-to-nudge involves an interviewer knocking on the door of a randomly selected household and asking the selected adult to take part in an interview over the phone or by video. It does not involve an interviewer entering the respondent's home.

The self-completion section of the SCJS (on drug use, partner abuse, sexual victimisation, and stalking and harassment) is usually completed by the respondent confidentially using the interviewer's laptop. This will remain the case for face-to-face interviews^[2], however we plan to introduce the option to complete this section through a web-based survey when the respondent has completed the main survey over the phone.

This should allow us to produce a final year of data under our existing contractual obligations, using a flexible survey design which can be amended as required should social-distancing restrictions return in future.

Conducting a mixed-mode SCJS provides an opportunity to learn about how survey modes work together and compare. However, whilst mitigating actions will be taken wherever possible, there are a number of risks associated with telephone interviewing, in particular. For example, there can be greater bias in those who choose to take part by telephone and response rates are known to be significantly lower than with in-person interviewing. Therefore, although a central aim, it may not be possible to compare 2021/22 results to the SCJS time series.

Results of the 2021/22 survey are expected to be published in mid-2023. For more information on the 2021/2022 survey, please visit the SCJS webpage.

Table 2: Summary of Methodology Changes since 2008/2009 SCJS

Survey company	2008/09	2009/10	2010/115	2012/13	2014/15	2016/17	2017/18	2018/19	2019/20
	TNS-BMRB6	TNS-BMRB	TNS-BMRB	TNS-BMRB	TNS-BMRB	Ipsos MORI & ScotCen	Ipsos MORI & ScotCen	Ipsos MORI & ScotCen	Ipsos MORI & ScotCen
Core achieved sample	16,003	16,036	13,010	12,045	11,472	5,567	5,475	5,537	5,568
Response rate	70.9%	70%	67%	67.7%	63.8%	63.2%	62.4%	63.4%	63.4%
Sample frame	Royal Mail PAF1	Royal Mail PAF	Royal Mail PAF	Royal Mail PAF	Royal Mail PAF	Royal Mail PAF	Royal Mail PAF	Royal Mail PAF	Royal Mail PAF
Survey weights	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold	Incident, Indiv, Hhold
Self-completion	√	√	√	√	√	√*		√*
Reference period 2	12 months	12 months	12 months	12 months	12 months	12 months	12 months	12 months	12 months
CAPI 3	√	√	√	√	√	√	√	√	√
No of victim forms	5	5	5	5	5	5	5	5	5
Cap on series of incidents	√(5+)	√(5+)	√(5+)	√(5+)	√(5+)	√(5+)	√(5+)	√(5+)	√(5+)
Sample type	Stratified sample design, rural areas clustered			Single stage unclustered stratified sample design
Design factor	1.5	1.5	1.5	1.3	1.2	1.34	1.22	1.17	1.21
Geographical coverage	Scotland (excluding smaller Island Communities)					Sampling frame includes all Islands
Police Force Area (PFA)	√	√	√	√	√	√	√	√	√
Police Division (PD) 4					√	√*		√*
Community Criminal Justice Areas (CCJA)	√	√	√	√	√	√	x	x	x

1. PAF – Postal Address File

2. The SCJS only counts data on incidents occurring in Scotland and in the reference period for crime statistics

3. CAPI stays for 'Computer-assisted personal interviewing'

4. Police Division were introduced 1 April 2013; estimates can be derived for pre-2013 data. PFA results can still be derived by aggregating divisions in the underlying dataset

5. There were no surveys conducted in 2011/12, 2013/14, or 2015/16

6. TNS-BMRB is now Kantar Public UK

* Note that data from the self-completion modules and data by Police Division (PD) is reported on biennially