Annex 2: Self-completion Methodology
Crime and victimisation surveys have been carried out in Scotland since the early 1980s. The geographical coverage, sample size, method and reference period have varied across these surveys. The Scottish Crime and Justice Survey (SCJS) was launched in April 2008 and represented a major shift in design, methodology and sample size from previous Scottish crime surveys.
This section provides more detail about aspects of the self-completion section of the questionnaire and specifically about the questions covering illicit drug use. Details are also included about the self-completion interview, the weighting used on the data and the classifications of illicit drugs used in this report.
Further information about the design of the SCJS is contained in Annex 2 of 2010/11 Scottish Crime and Justice Survey: Main Findings report and in the accompanying Technical Report.
A2.1 The self-completion interview
Interviews for the SCJS 2010/11 were conducted in the respondent's home using Computer Assisted Personal Interviewing (CAPI) and, for more sensitive questions in the self-completion section of the questionnaire such as those reported here, Computer Assisted Self-completion Interviewing (CASI). Interviews lasted 37 minutes on average in total, with the self-completion section taking just over 10 minutes on average to complete.
The results included in this report are based on interviews conducted by TNS-BMRB between 01 June 2010 and 31 March 2011. Respondents to the SCJS had the option of refusing to participate in the self-completion section which means this part of the questionnaire was not completed by all 13,010 respondents to the main survey. 10,999 (85%) respondents agreed to participate in the self-completion questionnaire.
The majority of respondents who answered the self-completion questions used the touch sensitive tablet PC on their own without any help from the interviewer (67%). A third (33%) of respondents asked the interviewer to enter their answers for them. A dislike of computers was the most common reason why respondents asked the interviewer to enter their answers for them (mentioned by 66%), while running out of time was the most common reason given for respondents refusing to complete it (mentioned by 52%). Only 6% of respondents refused to complete the self-completion questionnaire because of worries about confidentiality.
The SCJS questionnaire has a complex structure, but basically consists of three elements:
- The main questionnaire consists of a set of core modules asked of the whole sample, including demographics; and a set of full and quarter-sample modules, containing questions on a variety of topics;
- A victim form questionnaire which collects details about the separate incidents a respondent may have experienced during the reference period. This victim form can be repeated up to five times. The number of victim forms completed depends on the number and nature of incidents respondents experienced;
- A self-completion questionnaire covering sensitive issues, including illicit drug use. All respondents were asked to complete a self-completion questionnaire, but had the option to refuse this due to the sensitive nature of the questions.
The illicit drugs section is included at the start of the self-completion questionnaire.
Respondents are asked whether they have ever used 16 illicit drugs (see Chapter 1 for information on how these are classified in this report). While under-reporting of illicit behaviour by respondents is by far the main concern on a survey such as this, it is also recognised that some people may report taking particular drugs when they have not actually done so for reasons of bravado or other reasons. Respondents are therefore asked if they have ever taken 'semeron', a fictitious drug. Respondents who said they had taken semeron were then excluded from the final data outputs and data displayed in this report. There were 22 cases of respondents reporting that they had taken semeron in the SCJS 2010/11.
Those respondents who have taken drugs in the past are then asked a series of follow-up questions, including:
- Whether they have taken the drug in the last year. Those that have are asked whether they have taken the drug in the last month and, if so, which one they have taken most and how hard it is to get hold of it;
- What drug was the first ever taken; at what age they first took drugs, and what methods of drug taking they have ever tried;
- Whether they have ever mixed the drug they had used most often in the last month with either alcohol or other drugs, and in the case of the latter which drugs they have mixed with it;
- Whether, in the last month, they have felt dependent on the drug taken most often in the last month and have tried to cut down but were not able to do so.
The questions are asked in a loop, circulating through each drug (i.e. "Have you ever taken <drug name>?") rather than by selection from a single list of drugs. This approach has been shown to improve survey estimates of illegal drug-taking (Mayhew, 1995).
A2.3 Disclosure of sensitive information
Given the sensitive nature of the questions, especially as the majority of the questions on illicit drug use were on offending behaviours rather than victimisation (as opposed to the remainder of the self-completion questionnaire and the main questionnaire), a separate 'Don't wish to answer' button was provided at the top of the screen at every question in the self-completion section of the questionnaire.
At the start of the questions on illicit drug use, respondents were reminded that the answers they gave were completely confidential, reminded not to answer the questions including any drugs for which they had a prescription, and asked to answer the questions honestly.
"The following questions ask whether or not you have ever used drugs. Please answer them honestly. The answers you give are completely confidential. Please DO NOT tell us about drugs you have been given on prescription."
The classifications of drugs used in this report are detailed in section 1.4. However, there are two points of clarification which readers should note:
- Amphetamines can be classified as either Class A (when prepared for injection) or Class B (in powdered form). Since questions used in the SCJS do not distinguish between the forms of the drug taken for the purposes of analysis, the report assumes that all amphetamine use is of the Class B type. This is consistent with the approach adopted by previous surveys, including the 2008/09 and 2009/10 SCJS.
- The category "not classified" indicates that possession of these substances (poppers and glues, solvents, gas or aerosols) is not illegal but it is an offence to supply these substances if it is likely that the product is intended for abuse.
Table A2.1 compares the profile of respondents who answered the self-completion section of the questionnaire (including those who did so where the interviewer administered the questionnaire) and those who did not answer it:
- Equal percentages of males and females answering the main questionnaire also answered the self-completion section (85%);
- The percentages of respondents to the main questionnaire who answered the self-completion section decreased as age increased (for example, 90% of 16-24 year olds answered the self-completion section compared with 80% of those aged 60 or over):
- o Similar percentages of males and females within age groups answered the self-completion section;
- There were only minor differences in response between those who had been a victim of crime, as identified by the SCJS 2010/11 and those who had not (86% and 84% respectively).
Table A2.1: % of respondents to the main questionnaire overall and in selected sub-groups who did and who did not answer the self-completion section 
Base: All respondents (13,010).
|Row percentages|| Self-completion |
| No self-completion |
A2.6 Sample Profile
All sample surveys are not necessarily representative of a cross-section of the population due to a variety of reasons including whether potential respondents were available for interview and their willingness to participate in the survey. In the SCJS 2010/11, the achieved sample under-represented younger adults and over-represented older adults. Weighting was applied to correct for differences in the level of response among groups of individuals on key attributes (section A2.7).
The differential response of younger and older respondents to the self-completion section of the questionnaire discussed in section A2.5 brought the unweighted sample profile for the self-completion questionnaire slightly closer to the adult population profile. Weighting was still required as differences in the level of response remained.
Table A2.2 shows the profile of the achieved sample of the self-completion section before weighting was applied and the weighted sample profile.
Base: all respondents to the self-completion section (10,999).
| Unweighted sample 1 |
| Weighted sample 2 |
1. The unweighted sample includes twenty two respondents who said they had taken semeron at some point in their lives and who were subsequently excluded from the data for the drugs questions (section A2.2 above).
2. The weighted sample targets were based on the General Register of Scotland (GROS) mid year population estimates for 2010:
The results obtained in the SCJS were weighted to correct for the unequal probability of selection for interview caused by the sample design and for differences in the level of response among groups of individuals.
In view of the reduced response to the self-completion section, some additional weighting was necessary for use when analysing this sub-sample. The self-completion weights were calculated in a similar way to the main individual and household weights but based only on respondents who had answered the self-completion section of the questionnaire. The individual weight was applied to all of the questions included in this report as they all related to the experiences and opinions of individuals.
Email: Stuart King