VOTEADVICE is a four-year research project that investigates the impact of new technologies on political behaviour. It is a collaboration between the University of Exeter, as academic partner, Kieskompas, as industry partner developer of online voting advice applications (VAA) for a wide range of countries and elections and, Koç University as associate partner.
The scientific objectives of VOTEADVICE are to produce research related to how new technologies and social media influence political and social behaviour. In order to achieve this, we develop and apply techniques for the analysis of non-probability samples, online surveys and experiments and eye tracking tools. Our research projects analyse four different types of data: post-election studies that are comprised of random probability samples of citizens; online follow-up surveys after completing vote advice applications (VAA), online experiments and eye tracking data.
VOTEADVICE directly benefits three Early Stage Researchers (ESRs) whose involvement in this cutting-edge project through academic training and collaboration with industry and associated partners will impact their career development.
Description of the work performed since the beginning of the project,
VoteAdvice kick-off conference took place in Amsterdam (5-6 November 2014) with the attendance of 25 participants and 17 participants of our International Training Network (ITN) who delivered presentations on Experiments and Voluntary Samples. In January 2015 the ITN met in Exeter to deliver training sessions on Sample Matching Resampling-Weighting/ Bayesian Approaches to Non-responses and missing data (Exeter, 11 - 13 January, 2015); in February 2015, the three ESRs met with researchers of the College of Life and Environmental Science of the University of Exeter to undertake Eye Tracking Technology training. In March 2015, the team met again in Istanbul to undergo training on Online Experiments (Istanbul 12-14 March, 2015,). All presentations and an eye-tracking video with the recorded session are available on our website: www.voteadvice.org
The ESRs have received additional training on Quantitative Data Analysis; They attended the ECPR Winter School in Methods and Techniques in Bamberg, Germany, the Oxford Essex/Nuffield Summer School in Experimental Design, the 10th ECPR Summer School in Methods and Techniques and the EPOP Workshops Media in Context in Cardiff. In addition, they have accessed professional skills training such as project Management and product-based planning, Publishing and Academic Careers, and Entrepreneurial Skills.
As part of their training, they are taking secondments with our industry partner Kieskompas. The advent of the UK election allowed the researchers to take the first part of these secondments in the UK and develop a voting advice tool for the 2015 British General Election. After this initial stage in the UK, the first ESR moved to Amsterdam to completed the next part of his secondment which included design and implementation of the VAAs for Spain, Portugal, USA. The second ESR started her secondment in February 2016, the last ESRs’ secondment starts in May 2016.
Description of the main results achieved so far
Research to date has been presented in conference and working papers and in draft chapters of the ESRs individual PhD theses. The results are based on data collected as part of VAAs designed, developed and implement along with the industry partner. During the first half of the project, the main results can be summarised as:
(1) Design, development and implementation of 4 Voting Advice Applications (VAA): Portugal (https://www.bussolaeleitoral.pt ), Spain (https://brujulaelectoral.es/es ), USA (https://usa.electioncompass.org/en/#/ ) and United Kingdom (https://uk.electioncompass.org) and, supported the creation of Election Compass Wales (https://wales.electioncompass.org); data collected on approximately 200,000 respondents across the first 4 tools in four countries; (2) Analysis of samples, question items, issue positions and respondent response rates.
Scientific highlights: These scientific highlights are the basis for papers listed in our website www.voteadvice.org and for papers in progress. We continue to collaborate with other researchers who use similar tools, disseminate our results and innovations in conference presentations and informally through networks in order to encourage uptake of the innovations. In addition, we have made progress on the following: (1) Country specific political party issue placement items; (2) Post-stratification weighting of VAA data sets to allow further analysis by the Kieskompas team; (3) Kieskompas development of new interface; (3) Capturing social media accounts of users; (4) Data collected on media content surrounding the EU election, innovative cross-EU, tracking experiment of EU debate; (5)Country specific party-positions developed from VAA data; (6)Development of eye-tracking methodologies and experiment.
Intersectional and multidisciplinary collaboration: Our major multidisciplinary collaboration has been between political science and psychologists. We have achieved this collaboration in four ways: (1) Co-investigator Sam Vine is a psychologist who is leading training and research collaboration on eye-tracking; (2) Training events have been held with social-psychologists at Istanbul on eye-tracking; (3) Kick-off event was attended by and presentations were given by researchers from communications, sociology and psychology; (4) ESRs have initiated collaboration with social psychology experts for carry out current projects: eye-tracking experiments (with scholars at Koç University, Istanbul); (5) Face-morphing experiments by Kieskompas on leadership evaluations in the context of the 2016 US Presidential Elections (with psychologists at VU University, Amsterdam).
Outreach activity – Our team participated in the ESRC’s Festival of Social Science 2014 to deliver an event on data analysis with election related vote advice applications addressed to year-10 students from ISCA College. For the two-day event (November 2014), we had the support of University of Exeter students who have developed TickBox (another VAA). The VoteAdvice team has also participated in “Core Maths” a programme where maths students and teaching staff are invited to apply quantitative technique to relevant social and political issues. The dissemination activities have reached wide academic and non-academic activities. Conference papers have largely been presented at academic conferences. We have promoted to results to non-academic audiences largely through our social media activity (@VoteAdvice) and via an informal network of organisations that develop voting advice applications. The UK voting advice application was supported by LocalWorld news organisation with 22 million unique browsers across 83 websites.
Expected final results and their potential impact and use.
One of the major outputs of VOTEADVICE will be the application of techniques that will allow VAA data to be used for research purposes. Basic modes for sampling and data collection have been challenged by the evolution of mobile communication devices, which includes new analogue as well as the virtual modes of communication. The main criticism on these new access panels is related to volunteerism of panellists and the coverage, as surveys were traditionally used to obtain representative data on a target population. In a situation where even probability samples with low response rates are considered as volunteer samples, one needs to discuss the future of survey research and the development of new approaches regarding the basics of sampling and survey methodology, as well as innovations in statistical theories and the standardization of practices. Therefore, our applications can be used to make self-administered, non-probability samples an efficient way of conducting commercial research.
Our research will also have a direct impact on the debate about e-voting. Once people can vote electronically, one might envision voters from transferring their selection of candidates based on the recommendation given by a VAA into the official electronic ballot paper. Our research will indicate how VAAs impact voters in a variety of ways and whether VAA can lead to better informed voters and are likely to have a positive impact on electoral turnout. Our research can be used to evaluate how VAA might link to e-voting to make voting easier, more informed and thus enhance democratic participation.