The Scottish Government

« Previous | Contents | Next »

Listen

EVALUATION OF THE DRUGS DRIVING TV ADVERT

ANNEX A: Research methods

1. The quantitative element of this research consisted of a series of 6 questions included in the June 2002 Scottish Opinion Survey and a separate and more detailed survey of drivers.

The June 2002 Scottish Opinion Survey

2. In total, 1,025 people took part in the June 2002 Scottish Opinion Survey in the fourth week of June 2002. All interviewing took place in respondent's homes and was carried out by members of the NFO fieldforce using Computer Assisted Personal Interviewing (CAPI). The Scottish Opinion Survey uses a quota sample designed to be representative of the Scottish population in terms of geographical spread and the sex and working status of respondents.

3. The questions included in the Survey were designed to test basic levels of awareness and understanding of the drug driving television advert and of publicity relating to drug driving in general. A copy of the questions included is available from the research team on request (e-mail: rachel.ormston@nfoeurope.com)

Survey of drivers

4. In total, 730 interviews were completed between late July and September 2002. The target of 700 interviews was exceeded as interviewing was extended to boost numbers of interviews with people in younger age groups (17-24 year-olds in particular). As with the Scottish Opinion Survey, all interviewing took place in respondents' homes using CAPI.

Data collection methods and instruments

5. The questionnaire for the survey was developed by the NFO Social Research team in consultation with the Scottish Road Safety Campaign. It included the 6 questions from the Scottish Opinion Survey in order to provide an indication of whether awareness of the advert declined when more time had elapsed since it was shown. Two picture prompts were used during the interview - one featuring still frames from the drug driving television advert, and one colour copy of the 'Know the Score' logo. The questionnaire included a Computer Assisted Self Interviewing (CASI) section for questions about respondents experiences of drugs and drug driving. The final script took around 15 minutes to administer and covered topics including:

• Awareness of advertising and publicity on drugs and driving in general

• Awareness and understanding of the drug driving television advert

• Knowledge and beliefs about the legal consequences of drug driving

• Recognition of the 'Know the Score' logo and what respondents associate it with

• Experiences of cannabis and of driving under the influence of cannabis

• Experiences of drink driving

• Perceptions of the impact the drug driving advert on respondents' behaviour

6. The questions on experiences of drink and drug driving drew on previous research on drink and drug driving carried out by NFO System Three for the Scottish Executive (Ingram et al, 2000 and Anderson & Ingram, 2001). A copy of the full questionnaire is available from the research team upon request (e-mail: rachel.ormston@nfoeurope.com)

Sample design

7. The sample for the survey followed a quota design. This was felt to be the most appropriate strategy given the objective of achieving interviews with a disproportionately high number of interviews with younger drivers, aged 17-24 years old. Since there is no existing sampling frame for the population of young drivers, the only option for conducting a probability survey of this group would have been to screen houses to produce a sample from which a random selection could then have been made. This approach was deemed unnecessary given the purpose and objectives of the survey.

8. In order to obtain a sample likely to obtain a relatively high proportion of younger drivers for interviewers to select from, we employed a sampling agency to provide us with addresses for Enumeration Districts that fall into the top quartile for car ownership (i.e. the 25% of EDs with the highest numbers of cars), and which also fall into the top 2 quartiles for young people aged 17-29 (i.e. the 50% of EDs with the highest numbers of 17-29 year-olds).

We chose the top 2 quartiles rather than the top quartile for young people as young people tend to be concentrated in urban areas. Seventy-one percent of EDs in the top quartile in terms of concentration of 17-29 year-olds were in large cities or medium towns (compared with 60% of EDs generally). If we had selected from the top 25% we may have skewed the sample of EDs too much towards urban areas.

9. Once the sample was issued to interviewers, they were set quotas for interviews with drivers of different age groups, in addition to (non-interlocking) age and sex quotas. These quotas were designed to boost numbers in the 17-24, 25-34 and 35-44 year-old age groups at the expense of older drivers. The table below shows the proportion of drivers who fall into the different age groups across the Scottish population (based on figures from the Scottish Household Survey). The second indicates the target number of interviews that would have been achieved if quotas had been set proportionate to the age spread of the population of drivers, while the third column indicates the actual number of interviews achieved and the fourth the proportion of the total number of interviews conducted that this figure represents.

Table 1 Proportions of drivers in different age groups

Bases - Scottish Household Survey years 1-3 drivers = 26,664

10. As the table shows, drivers in the youngest age group (17-24 year-olds) made up 13% of our interviews, although they account for only 7% of the driving population. A higher proportion of interviews were also achieved in the 25-34 year-old and 35-44 year-old age group compared with the proportion of drivers in the general population that fall into these age groups.

Qualitative component

11. The purpose of the qualitative research was to explore in more depth the reactions of young drivers and passengers aged 17-24 to the advert. Two main sorts of interview were conducted: peer focus groups with mixed sex groups of friends, and paired depth interviews with couples and pairs of friends.

Paired depth interviews

12. By interviewing friends together, we hoped that participants would feel more confident about engaging with the topic, but also that they would reflect on, build on or challenge each other's accounts.

Peer group interviews

13. An alternative to conventional focus group work, where participants are typically not known to each other in advance, is to convene 'peer groups'. This is an approach that we have used successfully in a recent study of drinking and driving (Anderson and Ingram, 2001) and involves recruiting one person who meets certain demographic criteria and 'snowballing' from them to their friends or family. The advantage of this approach is that there is generally a certain level of existing trust within the group and that participants' awareness of each other's lives tends to act as a check on misrepresentation of views or behaviours. It also has the advantage of allowing discussion of drug-driving to take place in the social context in which the behaviour itself may occur - i.e. in groups of friends.

14. A total of 10 interviews (5 of each type) were conducted in September 2002 in both rural and urban areas in Scotland. We felt that this would be a sufficient number to cover a range of attitudes and experiences and to identify the main themes, without involving a high degree of duplication and inefficiency.

Recruitment of participants

15. For each interview, a 'core' participant who met certain demographic criteria was recruited initially, and then asked to help recruit friends with whom they regularly socialise. All core participants held full driving licences and had some experience of cannabis use (either past or present), which was viewed as putting them at higher than average risk of having experienced drug-driving, either as a driver or passenger.

16. Around half the core participants were recruited through the drivers survey. The survey included a question asking if people were happy to be contacted to take part in follow-up research. The results of the survey were analysed to find respondents who met the criteria in terms of age and cannabis use and who had consented to be approached about taking part in follow-up research. However, door-to-door recruitment by NFO interviewers was also used as we were unable to recruit sufficient core participants through the survey.

17. The structure of the sample is detailed below.

18. Four of the focus groups took place in neutral venues such as a room in a hotel. One took place in the core participant's home at their request. All the paired depths were conducted in the participants' own homes. In keeping with standard research practice for work of this kind, all respondents were paid a small financial incentive to attend in order to offset any childcare or travel costs incurred and to encourage participation.

19. The discussion moderators were Simon Anderson, Neil Cafferty and Rachel Ormston of NFO Social Research. A video of the advert was shown twice during the interview. The general content and flow of the discussions were dictated by the participants themselves, within the framework of a series of topics or areas for discussion introduced by the moderator. These topics included:

• Unprompted recall of the advert

• General impressions of the advert after the first showing of the video - message; content; style etc.

• Participants own driving behaviour, in particular experiences of drug driving as a driver or a passenger

• Detailed views of the advert after the second showing of the video - characters shown; the two-part structure; recollection and thoughts on the messages shown in the advert etc.

• Views on the target audience for the advert and the likely impact on this target group

• General views on the effectiveness of road safety campaigns

• Recognition and views on the 'Know the Score' and 'SRSC' logos

20. With the consent of participants, the discussions were tape-recorded for subsequent transcription and analysis.

« Previous | Contents | Next »