3 Questionnaire Content and Development
3.1 Structure and coverage of the questionnaire
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 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. 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.
Each of these three elements contain various sections (for example, the self-completion questionnaire covers illicit drug use, partner abuse and sexual victimisation). The questionnaires are also linked (for example, incidents which have been mentioned already in the victim form are not recorded in as much detail in the self-completion questionnaire as those that have not been - section 5.4). Within some sections there is further filtering so that questions are only asked of sub-samples (for example, those who have had contact with the police in the last 12 months). It is therefore recommended that data users read the following section on the questionnaire carefully before starting analysis and should also refer to the specific questions being analysed in the actual questionnaire itself.
The SCJS 2009/10 questionnaire consisted, in order, of the following questionnaires / sections / modules:
Main questionnaire (16,036 respondents):
- General views on crime and social issues;
- Victim form screener.
Victim form (repeated up to five times, based on information collected in the victim form screener section):
- Incident details;
- Support and advice wanted / received;
- Perception of the incident;
- Experience of criminal justice system organisations (police contact, prosecution, information and assistance relating to the investigation, Procurator Fiscal);
- Attitudes towards offender prosecution and sentencing.
Full sample modules (16,036 respondents):
- Community sentencing;
- Local community;
- Criminal justice system.
Quarter-sample modules (c.4,000 respondents each):
- Road safety cameras;
- Fraud (card and identity);
- Civil law;
- Crime scenarios;
- Procurator Fiscal.
Main questionnaire continued (16,036 respondents):
Self-completion questionnaire (completed by 13,418 respondents): 22
- Illicit drug use;
- Stalking and harassment;
- Partner abuse;
- Sexual victimisation.
The main differences between the 2008/09 and 2009/10 survey questionnaires are presented in section 3.7.1.
Before the main questionnaire starts, a series of screener questions are asked on the age and gender of each of the household members at an address. These questions are asked to allow the CAPI software to make a random selection of a household member (aged 16 or over) for interview (see section 2.8). They also instruct the interviewer to ask parental permission if the selected household member is aged 16 or 17 old. The NIPOCAPI software and tablet PCs which TNS- BMRB interviewers use on the survey allows the retention of this screener data for use in the actual interview.
The basic structure of the questionnaire is shown in Figure 1 below. The complete questionnaire can be found on the survey website and UK Data Archive as a separate document. 23
Figure 1: Questionnaire structure
The structure and content of the SCJS questionnaire is explored below, providing users with a fairly comprehensive overview of the questionnaire contents. However, data users should also refer to the full questionnaire before conducting analysis.
3.2 Main questionnaire content
3.2.1 General views on crime and social issues
The survey begins with a series of attitudinal questions on how important various social issues, including crime, are in Scotland. This is followed by questions about the local area, including how long the respondent has lived in the local area; how much the crime rate has changed locally and in Scotland overall; and how safe the respondent feels, either at home or going out alone after dark. The next questions ask respondents how worried they are that specific crimes will happen to them and their views on how likely they are to be a victim of these crimes. The majority of this section of the questionnaire is asked of all respondents.
3.2.2 Victim form screener questions
Respondents are asked whether they have experienced certain incidents since the beginning of the reference period (section 5.1).
These screener questions are separated into three broad groups:
- Vehicle related incidents, including theft of vehicle, theft from vehicle, damage to vehicle and bicycle theft;
- Household property incidents, including whether the home or outbuildings were broken into and things stolen or damaged, or an attempt was made to do so, or whether any property outside of the home was stolen or damaged;
- Personal incidents, including whether any personal property was stolen, or an attempt was made to do so, whether any personal property was damaged, and whether the respondent had been a victim of force or violence (including from another household member) or threats.
All respondents are asked a maximum of 17 victim form screener questions. 24 The wording of the screener questions has been kept consistent with past Scottish crime surveys. They are designed to ensure that all incidents within the scope of the SCJS, including relatively minor ones, are mentioned. The screener questions deliberately avoid using terms such as burglary, robbery, or assault, all of which have a precise definition that respondents would not be expected to know. This is consistent with the design of the British Crime Survey ( BCS) questionnaire.
The focus of the victim form screener questions switches between incidents experienced by the household and those experienced by the individual respondent:
- All vehicle and household property incidents are classified in the questionnaire as household incidents. Respondents are asked about whether anyone currently residing in the household has experienced any incidents within the reference period. A typical example of a household incident is criminal damage to a car (owned or used by someone in the household). It is assumed that the respondent will be able to recall these incidents and provide information even in cases where they were not present when the incident happened because it involves household property;
- Personal incidents refer to all crimes against the individual and are only asked about incidents that have happened to the respondent personally (for example a personal assault), and not to any other people in the household. 25
The distinction between household and personal incidents also affects how the data is analysed (section 7.2).
The questions are also designed in a way that avoids the respondent mentioning the same incident more than once (though this does happen in a small number of cases - section 7.1.4). 26
At the end of the victim form screener questions, the interviewer is shown a list of all incidents recorded. The interviewer checks this list with the respondent to ensure that all incidents they / their household have experienced in the reference period have been recorded and nothing has been counted twice. If this is not the case, the information is corrected before proceeding. Responses to the screener questions then trigger the victim form questionnaire if a respondent has experienced at least one incident.
3.3 Victim form questionnaire structure
Up to five incidents identified by the victim form screener questions are followed through in much more detail in the victim form questionnaire. The victim form questionnaire is designed to elicit all of the relevant details of an incident, irrespective of what incident the victim form was triggered by. 27 This then allows the coders to assign the correct offence code to the incident, regardless of what type of incident in the screener section triggered the victim form (see section 6.1 for details of the offence coding process).
Respondents are asked to report all incidents that they / their household experienced in the reference period. However, regardless of the number of incidents the respondent reports, the survey only collects detailed information on up to five of these. Incidents are covered in a specific priority order as explained below. This priority order is consistent with previous surveys and the BCS.
3.3.1 Identification and ordering of incidents for victim forms
Where a respondent had experienced more than one incident in the reference period, the CAPI programme automatically determines which of the incidents are followed up with a detailed victim form questionnaire, and the order in which the incidents are asked. Neither the interviewer nor the respondent have any choice about which incidents are followed up with the victim form questionnaire (with the exception of incidents of domestic violence) 28 or which order they are asked in. The priority ordering used by the computer is as follows:
1. According to incident type: Victim forms are asked in reverse order to the victim form screener questions. Broadly speaking this means that all personal incidents are asked before household incidents. Within household incidents, property-related incidents are asked before vehicle-related incidents.
2. Chronologically within each type of crime: If a respondent reports more than one incident of the same type, victim forms are asked in chronological order with the most recent incident first. 29
If a respondent has experienced five or fewer incidents identified at the victim form screener section, then a victim form questionnaire is asked for all incidents (with the order based on the priority ordering above). If the respondent has experienced more than five incidents in the reference period, only five victim forms are asked (with the incidents and order based on the priority ordering above). As a result the survey does not collect details about all incidents which a respondent experienced in these cases.
The priority ordering means that the incidents which are not asked about are likely to be incidents that tend to be more common. For example, criminal damage to vehicles is the lowest priority in the victim form order, but one of the most common crimes (motor vehicle vandalism had the highest prevalence of all the groupings of crime used in the SCJS 2009/10 Main Findings Report - Table A1.5).
Section 5.2 provides information on the numbers of victim forms that were completed by respondents.
3.3.2 Series of incidents
The victim form screener section also determines how many times the respondent has experienced a particular incident within the reference period. Most victim forms represent a single incident. However, in a minority of cases a respondent may have experienced the same type of incident ( i.e. one of those asked about in the victim form screener) a number of times in succession. If more than one incident is reported, the respondent is asked whether these incidents represented a 'series' or not. A series is defined as:
" the same thing, done under the same circumstances and probably by the same people".
In common with the BCS, if a respondent regularly experiences incidents where the same thing is done under the same circumstances by the same type of people, this is recorded as a series of incidents (or series incident) rather than separate incidents. 30 This most usually happens in a work situation, where groups such as patients or the general public might be involved.
Where a series of incidents is identified, only a single victim form is completed for the series, and this relates to the most recent occurrence.
In common with other victimisation surveys such as the BCS, only asking about the most recent incident where a series of similar incidents has occurred yields three practical advantages:
1. Many (although not all) incidents classified as a series tend to be minor incidents ( e.g. vandalism). Only asking about the most recent incident avoids asking a respondent the victim form questionnaire several times over when the detail of the incidents recorded will be very similar, therefore decreasing the likelihood that the respondent will terminate the interview or refuse to answer repetitive detailed questions about what can be very similar incidents;
2. It avoids using up the limit of five victim forms on incidents which tend to be less serious;
3. Respondent re-call of the incident details is likely to be more accurate for more recent incidents, and less so with earlier incidents.
In 2009/10 79% (3,827) of all victim forms (4,847) related to single incidents and 21% (1,020) related to a series of incidents.
In rare cases where respondents have experienced a mixture of single incidents and a series of incidents of the same type, the interview program has a complex routine which handles the sequence of individual and series incidents. This allows the priority ordering of the victim forms to be allocated, based on the date of the latest incidents first.
3.4 Victim form questionnaire contents
The victim form contains two basic sections; the first relates to the details of the incident itself (including details of the offender(s) if known), and the second to the follow-up of the incident with regard to victim's experience of the criminal justice system and related issues.
3.4.1 Incident details
The victim form is key to estimating victimisation in Scotland and collects three vital pieces of information about incidents:
1. When the incident or series of incidents occurred;
2. The respondent's description of the incident;
3. Important details of the incident to allow offence coding.
These are explored in turn below.
1. When the incident or series of incidents occurred.
For single incidents, where respondents are unsure about the exact month in which an incident happened, they are asked to narrow it down to a specific quarter ( e.g. between nine and 12 months prior to the month of interview). If they are unsure about the quarter in which the incident happened, they are asked to confirm if the incident happened in the 12 month reference period (section 5.1).
In some cases, respondents may report an incident in the victim form screener section as having happened within the reference period, which later turns out to be before the start of the reference period (and therefore outside the survey's coverage). In such cases, after this has been confirmed, the victim form is terminated and the questionnaire moves on to the next victim form (or the next section of the main questionnaire if the respondent has not experienced any further incidents). If the incident is in the month of interview, then details are collected, but the incident is not included in the survey statistics (section 5.1).
For incidents that were part of a series, respondents are asked how many incidents occurred in each quarter of the reference period and the month in which the most recent incident occurred. 31 If the most recent incident in the series occurred in the month of interview the victim form is still completed, but the number of incidents in the series is adjusted accordingly to only include those that happened in the reference period (section 5.1.1). If there are no incidents in the reference period or the month of interview then the victim form is non-valid and terminated in the same way as for single incidents.
In the CAPI questionnaire, reference dates (months, quarters etc) are automatically calculated based on the date of interview and appropriate text substitution is used to ensure that the questions always refer to the correct reference period. Because the 12 month reference period changes throughout the fieldwork year, some date-related questions in the victim form have different text each month to reflect this changing reference period.
2. The respondent's description of the incident.
At the start of the victim form respondents are asked to describe the details of the incident, with the interviewer probing for where it happened, who the victim was, who the perpetrator was and what they did. The interviewer then summarises these in an open-ended text entry. This summary description is vital to the accurate offence coding of incidents when used in combination with the series of pre-coded questions which are also asked about the incident (see section 6.1 for further detail of the offence coding process).
At the end of each victim form, the open-ended description is re-capped, along with the answers to some of the key pre-coded questions. By presenting this information on a single screen, interviewers have the chance to confirm with respondents that the information is correct and consistent. If the respondent and / or interviewer wish to add or clarify any information they have the opportunity to do so at this stage.
3. Important details of the incident.
Examples of the sort of information collected include where and when the incident took place; whether anything was stolen or damaged and if so, what; whether force or violence was used and if so, the nature of the force used and any injuries sustained.
The scope of the SCJS only includes incidents which happen within Scotland. For incidents occurring on-line to be included, the respondent must be living in Scotland at the time of the incident. If incidents occurred outside of Scotland then the victim form questionnaire terminates and the questionnaire moves on to the next victim form (or the start of the next section of the main questionnaire if the respondent has not experienced any further incidents). The victim form would ne assigned a non-valid offence code. The key questions within the victim form have remained largely unchanged from previous versions of the survey.
As well as details of the incident (along with experience of the criminal justice system and related issues - described below), respondents are also questioned about the characteristics of the offender(s), victim's perceptions of the offender(s) motivation and emotions felt as a result of the incident.
The victim form also contains a number of questions which are designed to help explain inconsistent answers which may arise within the questionnaire (for example, if a victim form was triggered because of an incident of theft in the victim form screener questions but nothing is recorded as having been stolen).
Several questions are included to allow the interviewer to terminate the victim form if the incident being recorded is a duplicate of a previous victim form (section 7.1.4).
3.4.2 Victim's experience of the criminal justice system and related issues
Several sections follow the questions on the details of the incident on various subjects related to the victim's experience of the criminal justice system and related issues: 32
- Support and advice wanted / received: what victims would liked to have received after the incident, what they did receive and from whom, and how satisfied they were with it;
- Perceptions of the incident: whether anyone other than the offender(s) was responsible for the incident, whether the respondent themselves used force or was under the influence of drugs or alcohol prior to the incident;
- Police contact: how the police came to know about the incident and how satisfied victims were with police handling of the incident;
- Prosecution: whether the offender(s) was prosecuted and what sentence they received;
- Information and assistance relating to the investigation: what information / assistance was received and from whom (including the police, Victim Support Scotland and the Witness Service), satisfaction with it, reasons for dissatisfaction and what information / assistance victims would liked to have received;
- Procurator Fiscal: contact with and type of assistance received, views on the helpfulness of this assistance and satisfaction with contact with the Procurator Fiscal.
- Attitudes to offender prosecution and sentencing: whether the offender(s) should have been prosecuted in court, reasons why not, what alternate punishment should be used as an alternative to prosecution in court, whether they should have received a prison sentence and how long this should be, what type of non-prison sentence they should receive and perception of the incident as a crime or not.
3.5 Main questionnaire content (continued)
After the victim form screener (or victim form) has been completed, the main questionnaire continues with thee full-sample module sections.
3.5.1 Community sentencing (full sample)
This section is the continuation of the main questionnaire and all respondents are asked it. Respondents are asked about what community sentences they are aware of, whether these sentences would make an offender less likely to commit a crime in the future, and whether they agree or disagree with a series of statements about community sentences. They are also asked whether they have ever personally been on remand or served a prison sentence in Scotland.
3.5.2 Local community (full sample)
A new section for the 2009/10 survey, this section asks respondents to imagine a scenario in which a group of young boys are damaging a bus shelter in the respondent's local area. They are then asked how likely it is that they themselves would take a series of actions (for example, speaking to the boys parents). Respondents are then read a list of statements about people in their local area and asked how far they agree or disagree with each statement (for example, 'people in this local area pull together to prevent crime'), before being asked how many people they know in the local area.
3.5.3 Scottish criminal justice system (full sample)
The criminal justice system in Scotland is defined to respondents as:
" the shared name for all the organisations in Scotland that deal with finding offenders and arresting them, then taking them through the court system and deciding what sentence they are given if they are found guilty".
Questions are asked of respondents' level of awareness of the system as a whole and confidence in it via a series of statements about it. 33 Respondents are then asked which of the component organisations that make up the criminal justice system they have heard of, and which they have personally been in touch with. Finally, some of the responsibilities of local police forces are read out and the respondent is asked to say how confident they are in their local police forces' ability to fulfil them. All respondents are asked the questions in this section.
3.5.4 Quarter-sample modules (A-D)
Addresses are randomly allocated to one of four modules at the sampling stage. Allocations are equal so that one quarter of addresses are allocated to each module. In the final achieved sample this percentage varies slightly due to small differences in response rates between modules (see Table 3.1).
Table 3.1: Quarter-sample module sample sizes
Source: SCJS 2009/10.
Base: All respondents (16,036).
Sample size (n)
3.5.5 Module A: fear of crime
Fear of crime: The short section on fear of crime includes questions about how common respondents think various crimes are and where they have got this impression from. All of these are with reference to the respondent's local area (that is within a 15 minute walk of the respondent's home). Respondents are asked whether they have changed anything in their everyday lives as a result of these issues, and if so what (new questions for the 2009/10 survey).
3.5.6 Module B: police and safety cameras
Police: The new section for 2009/10 begins with questions about police visibility in the local area, including how important it is to the respondent to know a community police officer who patrols the local area, how frequently patrols are seen and in what form, and opinions on police presence and why these are held. Respondents are then asked about their level of agreement / disagreement with a series of statements about the police in their local area (for example, 'they can be relied on to be there when you need them'). Finally a series of questions are asked of those who have been stopped by the police in the 12 month reference period, including whether a reason was provided for being stopped; what this was; whether the respondent thinks this was the real reason; whether a search took place; how the respondent viewed the contact with the police and how satisfied they were with it.
Safety cameras: Road safety cameras are defined as both speed cameras and red traffic light safety cameras. Respondents are asked how far they agree with the use of each, and whether they agree or disagree with a battery of statements about them.
3.5.7 Module C: fraud and civil law
Fraud: The fraud section of the questionnaire focuses on two types of fraud; card fraud and identity theft. The section is included to provide a measure of the extent of both card and identity fraud, as incidents of these crimes are not specifically included in the victim form screener questions, and hence would not necessarily be recorded by the survey.
Card fraud: This covers both the unauthorised use of credit and bank cards to buy or pay for things or withdraw cash, and the use of card details for the same purposes. Respondents who experience this during the 12 month reference period are asked where the cards / details were used, and, if they were used online, where they were living at the time. If the use of cards / details occurred outside of Scotland or the respondent was not living in Scotland at the time the cards / details were used then follow-up questions are not asked.
Various follow-up questions were included to find out more about the incident and, where appropriate, contact with the police in relation to the incident.
Identity fraud: These questions relate to someone pretending to be the respondent or using their personal details (including name, address, date of birth or National Insurance number). The section covers their use to do things such as obtain credit, open a bank account, apply for a mobile phone contract or state benefits or for official documents such as a driving licence or passport or to commit some other kind of fraud. If the fraud was not perpetrated in Scotland then the follow-up questions are not asked. The follow-up questions broadly follow the same format as those for card fraud.
Civil law:34 This section relates to problems and disputes that the respondent may have experienced in their everyday life in the last three years that can be settled in court. 35 The section is carefully introduced to the respondent due to both the extension in the re-call period and the shift towards incidents which relate to civil law rather than criminal law:
"I am now going to ask you some questions about different kinds of problems or disputes you might have had in the past three years. These are problems that are not directly related to crime but to other issues you might have to deal with in your everyday life. Of course, everyone has problems in their lives from time to time which they deal with. We are particularly interested in problems or disputes you had that you found difficult to deal with or that you could not solve easily."
Civil law issues are grouped into four specific types:
1. Those concerning home, family or living arrangements (neighbours, family, housing and immigration);
2. Those concerning health and well-being (injury because of an accident or medical negligence and mental health issues);
3. Those concerning money, finances or any purchased good or service (debt, benefits and faulty goods and services);
4. Those concerning unfair treatment (discrimination, unfair treatment by the police and employment related issues).
Respondents are asked how important it was that they solve the problem, and which is the most important (if they have had more than one). For the most important or only problem respondents are asked whether it was resolved or not; if they have solved or are trying or planning to solve the problem they are asked if they used / are using help and advice from others; if the problem has been solved then they are asked how satisfied they are with the results. Three new questions were added to the 2009/10 survey: if they used / are using help and advice from others then they are asked who from, and if they are not then they are asked why not. Finally, those who are not planning on trying to resolve the issue are asked why not.
3.5.8 Module D: crime scenario and Procurator Fiscal
Crime scenario: This is a new section for the 2009/10 survey. Respondents are presented 36 with an imaginary scenario where a crime is committed and asked their opinion of the punishment that the offender(s) should receive (if any). The amount of information respondents are in possession of about the incident is then incrementally increased over five stages; after each new piece of information is provided they are asked if they would change their opinion of the punishment the offender(s) should receive (if any).
Procurator Fiscal: this short section includes questions on awareness of the Procurator Fiscal's role, contact with them and satisfaction with this contact.
3.5.9 Demographics section
A variety of demographic information is collected from all respondents (many using Scottish Government harmonised questions), 37 including:
- Newspaper readership;
- Age, gender, marital status, ethnicity, religion and health status;
- Tenure and property type;
- Employment status, including questions to allow Office for National Statistics Socio-Economic Classification ( NS- SEC) coding; 38
- Household income and ability to afford an unexpected expense.
Age and gender of other persons in the household are collected at the interview screening stage (see section 3.1 for details). This information is used to establish the household reference person ( HRP). 39 This standard classification is used on most government surveys and is based on the following criteria:
- The HRP is the member of the household in whose name the accommodation is owned or rented, or is otherwise responsible for the accommodation. In households with a sole householder, that person is the HRP.
- In households with joint householders (for example, two people's name on the mortgage) the person with the highest income is taken as the HRP.
- If both householders have exactly the same income, the older is taken as the HRP.
At the end of this section respondents are asked whether they are willing to provide their contact details and survey answers to the Scottish Government or research organisations who are acting on their behalf for the purpose of further research.
3.6 Self-completion questionnaire contents
The self-completion questionnaire is asked of all members of the sample after the demographics section of the main questionnaire has been completed - there are no upper age restrictions. 40 Respondents could, however, refuse to answer the self-completion questionnaire: 41 84% of respondents to the main survey completed the self-completion questionnaire (section 4.6).
The self-completion questionnaire covers the following topics:
- Illicit drug use and availability;
- Stalking and harassment;
- Partner abuse (including both psychological and physical abuse by a partner);
- Sexual victimisation.
Details of stalking and harassment, partner abuse or sexual victimisation incidents recorded in the self-completion questionnaire are not included in the statistics 'all SCJS crime' (see section 7.1.5 for details) unless the incident is also mentioned by respondents in the victim form and assigned an offence code in the normal way. Incidents reported only in the self-completion questionnaire could not be assigned offence codes in the same was as those collected in the victim form as only a limited number of follow-up questions were asked about incidents (reflecting an ethical decision based on potential respondent distress at having to disclose detailed information on very sensitive incidents).
Section 5.8 provides further information on the administration of the self-completion questionnaire.
3.6.1 Illicit drug use
Respondents are asked whether they have ever used 16 illicit drugs. 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 have said that have taken semeron are then excluded from the final data outputs and reporting for the drugs section of the questionnaire. 42 There were nine cases of respondents reporting that they had ever taken semeron.
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 12 months. 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 ( 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).
3.6.2 Stalking and harassment
This section begins with a screener section collecting information about respondents' relationship history and sexual orientation. 43
Respondents are then asked about whether they have experienced any of four forms of stalking and harassment in the 12 month reference period. As measured by the SCJS stalking and harassment included: 44
- Receiving obscene or threatening correspondence, such as letters, emails, text messages or cards;
- Receiving obscene, threatening, nuisance or silent telephone calls;
- Having someone waiting outside a home or workplace on more than one occasion;
- Being followed around and watched on more than one occasion.
If they have, they are asked (for the most recent incident of each form of stalking and harassment if they had experienced more than one) who the offender(s) was and what their relationship to the respondent was. The respondent is also asked whether the police came to know about the incident, and if not, why not.
3.6.3 Partner abuse
This section is only asked of those who have had a partner at any time since they were 16. It is introduced carefully to ensure that respondents are clear on the coverage of the questions:
" We would now like to ask you some questions about your own relationships with any partners you may have had since you were 16. By partner we mean a boyfriend, girlfriend, husband, wife or civil partner."
Two questions present a list firstly of types of psychological abuse and secondly types of physical abuse; respondents were asked if they had experienced any of these since they were aged 16, and if so, how many partners perpetrated these acts. If any of these types of abuse have taken place within the 12 month reference period, a series of follow-up questions are asked, the majority about the most recent / only incident in that time, including:
- Where they happened (in Scotland or elsewhere) and how many incidents happened since the beginning of the 12 month reference period;
- Whether any children were in the household, whether they saw or heard what happened or were involved or hurt in the incident;
- What physical and psychological consequences they experienced;
- What people or organisations, if any, the respondent informed of the incident;
- Whether the police came to know about the incident and follow-up questions including: satisfaction with the way police dealt with the incident; why they did or didn't report the incident to the police; whether it was reported as a crime; if the report resulted in a prosecution and whether there was a conviction; satisfaction with the police handling of the incident;
- Whether the perpetrator was living with the respondent at the time of the incident and whether they are living with them at the time of the interview;
- Whether the respondent considered what happened to be a crime or not.
At the end of this section, all those who have had a partner since they were 16 are asked whether they consider themselves to have ever been a victim of domestic abuse. The term domestic abuse is not defined to the respondent.
3.6.4 Sexual victimisation
The questionnaire asks about all types of sexual offences. These are categorised into two groups, which are termed serious sexual assault and less serious sexual assault. 45 Less serious sexual assault includes:
- Indecent exposure;
- Sexual threats;
- Touching sexually when it was not wanted.
Serious sexual assault includes:
- Forcing someone to have sexual intercourse when they did not want to;
- Attempting to force someone to have sexual intercourse when they did not want to;
- Forcing someone to take part in other sexual activity when they did not want to;
- Attempting to force someone to take part in other sexual activity when they did not want to.
Respondents are reminded that they may skip such sensitive questions via using the 'Don't wish to answer' button at the top of the screen.
Different follow-up questions are asked of respondents depending on the nature of the incident(s) they have experienced ( i.e. whether they are classified as less serious or serious sexual assault) and when they experienced them (in the last 12 months or since the age of 16).
Less serious sexual assault
Victims of less serious sexual assault are asked the following questions for each of the four forms they have been the victim of: 46
- When the incidents(s) happened (in the last 12 months, longer ago or both); and how many times they occurred during the 12 month reference period;
- What the relationship was between the respondent and the offender(s) and the gender of the offender(s) for all incidents in the 12 month reference period and the latest incident in the reference period, as well as for incidents longer ago than the last 12 months but since the age of 16;
- For the latest incident in the 12 month reference period: whether it happened in Scotland; whether the police came to know about the incident: how it was reported or if it was not, then the reason why.
Serious sexual assault
Respondents who have experienced each form of serious sexual assault are asked additional follow-up questions about the incident(s) compared to the follow-up questions for less serious sexual assault. In addition, the time-period which some of the follow-up questions reference was increased compared to the 2008/09 survey from incidents occurring in the 12 month reference period to incidents occurring since the age of 16. This amendment to the questionnaire was made to increase the number of cases available as numbers in 2008/09 were insufficient to allow robust analysis (see Box 3).
Box 3: Serious sexual assault - time periods covered
The 2008/09 survey reported very few cases of serious sexual assault in the 12 months before interview (less than 50 cases across all four types combined). Therefore, a decision was made to increase the time-period to which the follow-up questions referred in order to increase the number of cases available for analysis.
The follow-up questions in the 2008/09 survey followed the same pattern as the questions for less serious sexual assault with respect to the time period, asking only about the latest incident within the 12 month reference period (questions <FS,AFS,OS,AOS>_WHR through to _12i).
In the 2009/10 survey some of the follow-up questions for serious sexual assault are asked referring to the latest incident the respondent has experienced since the age of 16 ( i.e. rather than the latest incident within the last 12 months).
The potential draw back of this approach is that respondent re-call error could be increased. For this reason, only the more factual of the follow-up questions are asked for the latest incident since the age of 16 (such as whether the police came to know about the incident. These include:
- Whether the incident happened in Scotland or elsewhere;
- Physical injuries sustained
- Whether the police came to know about the incident, how it was reported and if not then the reason why;
- Whether the incident was reported as a crime and if so, whether there was a prosecution and conviction.
Analysts should be careful when making comparisons between the 2008/09 and 2009/10 data due to the differing time periods covered.
Although time-periods differ, and with the exception of some additional questions that are asked, a broadly similar set of follow-up questions are asked for each form of serious sexual assault the respondent has experienced compared with those asked for less serious sexual assault. These include:
- When the incidents(s) happened (in the last 12 months, longer ago or both); and how many times they occurred during the 12 month reference period and ever (since the age of 16);
- What the relationship was between the respondent and the perpetrator(s) and the gender of the perpetrator(s) for all incidents and the latest incident both for those happening in the 12 month reference period and those happening prior to the 12 month reference period (since the age of 16);
- For the latest incident (irrespective of when this was), whether it happened in Scotland; what physical injuries were sustained as a result of the assault, whether the police came to know about the incident, how it was reported or if it was not, then the reason why, whether it was reported as a crime and if so, whether there was a prosecution and conviction;
- For the latest incident in the 12 month reference period, the location of the assault, the psychological impacts, whether the respondent perceived the incident as a crime or not, satisfaction with police handling of the matter and reasons for dissatisfaction (where applicable), people or organisations informed of the assault, and whether the respondent or the offender(s) were under the influence / had been given drugs or alcohol at the time of the assault.
3.7 Questionnaire changes and piloting
The 2009/10 survey questionnaire remained to a large extent the same as the 2008/09 questionnaire. A number of changes were made (section 3.7.1), principally to the quarter sample modules, as a result of an internal Scottish Government consultation with policy colleagues with an interest in the survey as well as advice from the survey Technical Advisory Group ( TAG).
These new modules and selected other questions underwent cognitive testing (section 3.7.2) before a limited CAPI pilot was conducted (section 3.7.3).
3.7.1 Questionnaire changes
The major changes to the questionnaire between the 2008/09 and 2009/10 survey are summarised below.
Full sample modules:
- Local community: section added.
Quarter sample modules:
- Module A:
- Workplace violence: section removed.
- Module B:
- Insulted / pestered / intimidated: section removed;
- Police: section added.
- Module D:
- Civil justice: section removed (but retained in quarter sample module C);
- Crime scenario: section added.
- Sexual victimisation: reference period for incidents of serious sexual assault extended to cover the last incident experienced since the age of 16 - see section 3.6.4).
The content of questionnaire sections is summarised in section 3.2 onwards. A list of questions that have been added, deleted and amended is provided in Annex 6. Copies of the 2008/09 and 2009/10 survey questionnaires are available from the survey website. 47
3.7.2 Cognitive question testing
Cognitive testing uses probing techniques to try and understand the thought processes that a respondent uses in answering a survey question. It is designed to see whether the respondent understands the question, or specific words and phrases contained within the question; what sort of information the respondent needs to retrieve in order to answer the question; and what decision processes the respondent uses in coming to an answer
Cognitive testing was undertaken for some of the questions in the local community (full sample module) and the police sections (module B), all of the questions in the crime scenario (module D) and some other additional questions, including some which were not added to the final questionnaire. 20 questions were tested in total using two questionnaires. Other new questions did not undergo cognitive questions, either because they were from, or based on, existing questions from other surveys or were simple factual questions (such as those used in the victim form for offence coding purposes).
The process of cognitive interviewing is similar to conducting a semi-structured, depth interview, and employs both qualitative and quantitative interviewing techniques. The interview involved:
- Selected questions asked of respondents, replicating conditions of the actual interview as far as possible ( e.g. asking some questions that did not need testing but set the context of those that did);
- Respondents verbalising their thoughts (thinking-aloud) while answering questions;
- Researchers using different probes and prompts to encourage provision of more in depth / different information.
Research staff from TNS- BMRB conducted 32 interviews with members of the general public in two locations, one in Glasgow and the other in Edinburgh in January / February 2009. Members of the public were selected on the basis of quotas by age, gender and social grade. The purpose of this selection was to achieve a breadth of respondents' views from across the range of different types of individuals that would be interviewed in the actual survey.
The results were presented to the Scottish Government in a summary report containing a series of recommendations relating to the tested questions ( TNS- BMRB, 2009).
3.7.3 CAPI pilot
Once the final questionnaire had been agreed, a small scale technical CAPI pilot was conducted by two TNS- BMRB interviewers testing the CAPI questionnaire to ensure that the questionnaire script functioned correctly and that the links between the questions and modules were operational. It also allowed early inspection of raw data to check questionnaire routing as well as testing the data collection, performance monitoring and management systems and offence coding systems.