Testing the Young Voter Registration Form for the 2014 Referendum on Scottish Independence

3 METHODOLOGY

3.1 The testing of the YVF was conducted through a series of one-to-one cognitive interviews which involved:

• giving participants an envelope containing an annual canvass form placed over the YVF and asking them to complete these as they would if they had received them in the post
• researchers observing how participants completed the YVFs
• an in-depth interview with participants to establish whether YVFs were completed accurately, the reasons behind any errors made and views on the usability of the YVF.

3.2 The testing method was qualitative in nature: it was primarily concerned with uncovering the range of potential problems experienced when participants completed the YVFs and why these problems arose. It was not designed to measure the likely prevalence of errors among the wider population and was therefore not a quantitative exercise.

Sample profile and recruitment

3.3 To ensure that the testing explored the experiences of a range of people in different circumstances, and thereby identify a range of potential problems with the YVF, interviews were carried out with the main householder in households containing:

• 15 year olds who will be 16 by the date of the referendum - referred to as 'eligible households' (householders should fill in the YVF)
• 15 year olds who will not be 16 by the date of the referendum, 13-14 year olds or 16-17 year olds - referred to as 'almost eligible households' (householders should not fill out the YVF but may have made errors because of their child's age being close to the eligible age)
• 0-12 year olds or no children- referred to as 'ineligible households' (householders should not fill out the YVF).

3.4 In addition, the testing targeted groups who are likely to experience problems with filling out forms because of language issues or low literacy.

3.5 Forty interviews were conducted in total. Table 2.1 shows the breakdown of interviews by household type.

Table 2.1: Sample profile

3.6 The testing included both men (12) and women (28) with different levels of qualifications (at least 12 participants had less than Higher qualifications) and working status.

3.7 To recruit participants in eligible and almost eligible households, the research team applied for, and were granted access, to the Scottish Household Survey (SHS) re-contact database. These participants were telephoned by Ipsos MORI and invited to take part in the testing.

3.8 Recruitment of participants from ineligible households was conducted on-street by experienced Ipsos MORI recruiters using a recruitment questionnaire developed by the research team.

3.9 Participants who used English as a second language were recruited through community organisations and snowballing^[4].

3.10 The research team had originally intended to recruit participants with low literacy through adult literacy groups. However, low literacy participants were identified in the course of undertaking interviews among eligible and ineligible households so it was not necessary to use alternative recruitment methods.

Fieldwork

3.11 Interviewing was carried out in Edinburgh, Glasgow, Paisley, West Lothian and Fife by members of the Ipsos MORI research team. Interviews lasted on average 25 minutes each and participants were given £25 to thank them for their time.

3.12 Fieldwork took place between the 11^th and 28^th of March 2013. Fieldwork started in advance of obtaining access to the SHS re-contact database because of the timescales available for the study. This meant that most of the interviews with participants in ineligible households, those who use English as a second language and those with low literacy were conducted towards the start of the fieldwork period, while interviews with eligible and almost eligible householders took place towards the end.

The Young Voter Registration Form and other research materials

3.13 To begin with, the draft version of the YVF designed by the Scottish Government (version 1) was tested in 23 interviews. In light of the findings which emerged in these interviews, the YVF was slightly amended (see section 4.10). The amended version of the YVF (version 2) was tested in the remaining 17 interviews.

3.14 In agreement with the Scottish Government, a dummy date for the referendum was used on the YVF as the actual date had not been announced by the start of the fieldwork period. The dummy date was 30^th January 2014 to provide a similar timeframe (and therefore age for eligibility) from the canvass to the date of the referendum as will be used in the actual canvass.

3.15 While the precise date of the referendum had not been set by the start of fieldwork, it had been announced that the referendum would take place in autumn 2014 and announcement of the precise date was made during the fieldwork period. However, the discrepancy between the dummy date on the YVF and these announcements was rarely brought up by participants in the testing. In cases where it was, it did not appear to have an impact on the testing.

3.16 Other fieldwork materials developed included a generic annual canvass form and a discussion guide. To ensure that the testing replicated real life experiences of completing canvass forms as much as possible, the design of the generic annual canvass form was largely based on the existing form used by the Lothian Valuation Joint Board, which is designed in line with Electoral Commission guidelines.

3.17 Copies of all research materials are provided in Appendix A.