Scottish Election Study 2016

In May 2016 Scottish residents aged 16+ will vote in the fifth election to the Scottish Parliament. This takes place a year after the UK General Election in which the SNP won all but three Scottish seats, and 20 months after a referendum in which support for independence rose to 45%, a figure surpassed in some recent polls. This election provide an opportunity to explore long-term electoral realignment in Scottish politics, the specific impact of the referendum, and youth electoral engagement following the extension of the franchise. In addition we will address more habitual concerns about the impact of campaigning or opinion polls and the predictors of vote choice.

We will conduct an online survey before the election to identify the factors influencing vote intention, and return to the same respondents in an immediate post-election survey to identify how and why people voted the way they did. Via this panel design, we can determine how intentions translate into behaviour and see how expectations influence reactions to the election result. The pre-election survey will be conducted across the last four weeks of the campaign to capture reaction to campaign events and polls. In addition, we will collect data on campaigning through a survey of party agents, interviews with party officials, and data on leader visits and campaign spending. We will also collect data on opinion polls from January 2016, tracking these alongside the campaign survey data to assess the impact of these polls.

Primarily we are interested in exploring whether and how individuals decide to vote, how this process might be shaped by factors distinct from UK general elections, and also whether previous models of voting behaviour in devolved elections are still able to account for partisan preferences after the referendum. We will not only link preferences in the 2016 election to past voting behaviour in devolved and UK elections but also probe opinion on two wider issues: constitutional change in the UK (in England as well as Scotland) and the possibility of British exit from the EU. In each case, we can examine how this relates both to partisan preferences and campaign strategies in the 2016 SP elections.

Throughout we will locate Scottish findings within a comparative context through our collaboration with the 2016 Welsh Election Study team, with whom we will conduct a Massive Open Online Course on the devolved elections. We will also include a core of questions from the comparative Making Electoral Democracy Work project, which to date has conducted sub-state surveys in five states . Our project will provide the first UK data for analysis within this comparative project. Our research questions, discussed in detail in the case for support, relate to contested theories of voting behaviour in elections. We focus on the extent to which voting behaviour is a function of identity, issue positions (especially on the independence question), ideology, evaluations of government performance, and leader preferences. We draw on a rich body of research on Scottish voting behaviour, much of which has been generated by previous Scottish Election Study data.

We are eager to contribute to public debate and to facilitate engagement with the research process. We will hold consultation events before the survey begins, will invite academic researchers and school students to suggest survey questions, will provide the data in a format that is easy to analyse online and will employ a communications team to reach key stakeholders. Online research facilitates the quick release of research findings so that we will be able to contribute to wider debates about the election result. We will conduct briefing events and produce briefing papers as well as generate top quality academic research for journals and a monograph targeted at a top university press. We will also hold capacity building events to facilitate access to data for secondary school and university students.

*Who will benefit from this research?*
Election surveys generate insights about public opinion and political behaviour. Much of the impact of this research is diffuse and indirect but we know that our research findings (SES 2007, SES 2011, SRS 2014) has been sought by those in Scottish and UK government, political parties, broadcast and press media, campaigning organisations and members of the public and we expect that this will be true for the 2016 SES as well. We have already discussed this proposal with a number of practitioners, policy makers and other impact partners, including researchers in the Scottish Government, non-academic organisations such as the Electoral Reform Society, broadcasters - notably the BBC and STV - and teachers of Modern Studies and Mathematics. The extension of the franchise to 16 and 17 year olds provides another set of impact partners in terms of both students and schools.

*How will they benefit?*
Our data and analysis will provide political parties with information about the perceived effectiveness of their campaign strategies on, government officials with details about voter preferences for different party policies, and the public with data about the wider electorate. To bodies such as the BBC, Electoral Commission and Electoral Reform Society we will provide data about the perceived fairness of the electoral process and evaluations - and awareness - of these specific institutions. In addition, the data will be of direct benefit to academic users who will be able to analyse the data to pursue their own research agendas.

*How will we facilitate impact?*
We will engage with impact partners at three key stages:
1. Designing the surveys. In addition to inviting comments from our advisory board on drafts of the survey we will also engage with impact partners in two ways: first, by open consultation for all other interested parties in which we will invite proposal for questions; second, by two parallel competitions among researchers and among secondary school students for small amounts of questionnaire space. In these ways we will capitalise on existing links and forge new ones. These are practices we trialled with the Scottish Referendum Study to very useful effect: generating survey questions for use by other academics (Brandenburg et al., 2015) and of interest to young voters.

2. Initial headline findings. The speed of internet surveys means that findings from our pre-election survey will be available the day after polling and post-election findings will follow just days later. We have some experience with this. We successfully presented preliminary data from the Scottish Referendum Study shortly after the referendum, for example, to a large public dissemination event at Dynamic Earth in Edinburgh. The findings will be circulated to our impact partners and other interested parties via a range of channels coordinated by our communications officer such as press releases, media appearances, social media, the project website and policy blogs.

3. Further analysis. In parallel to our scientific outputs we will produce non-technical briefings on key issues as well as matching teaching materials for use in schools, with further readings and issues for discussion. Likely topics include but are not limited to youth engagement, new SNP voters and the impact of opinion polls. These will serve as the basis for workshops at which our impact partners can discuss the broader implications of our findings.