A Choice‐Based Measure of Issue Importance in the Electorate

The forthcoming article “A Choice-Based Measure of Issues Importance in the Electorate” by Chris Hanretty, Benjamin E. Lauderdale and Nick Vivyan is summarized by the authors below.

A Choice‐Based Measure of Issue Importance in the Electorate

**Some issues are more important to voters than others.** 

This is a simple idea, but measuring “issue importance” is hard. 

If you ask people what issue is most important to them, they often mention issues that they’ve read or heard lots about, not necessarily issues that are personally important to them.  These self-reports aren’t that helpful: knowing what issue a respondent *says* is most important to them often [doesn’t help us make much sense of how they *vote*](https://doi.org/10.1007/s11109-018-9494-0). 

Our article takes a different approach. Instead of asking survey respondents to pick important issues, we ask respondents which positions they prefer themselves and then ask them to pick between sets of issue positions, presented as fictional candidate platforms. We fielded our survey in the UK, but nothing about our method is specific to the UK except the issues and positions we use. You can see an example of the issue positions presented to respondents in Figure 1; Figure 2 shows the choice between fictional candidates. Respondents get to pick one of the two bundles, or to say that they’re not sure. 

These choices don’t show which issues were most important for specific individuals, but they do show which issues tend to influence voting most strongly overall.  If respondents tend to pick a bundle which gives them their most-preferred policy on issue X, no matter what positions the hypothetical candidates take on issues Y and Z, then we learn that issue X is more important to them than Y and Z.  The core intuition is that what we actually mean by an issue being more “important” is that changing a candidate’s positions on that issue has a greater tendency to make people choose different candidates than they otherwise would have. 

A final element of our importance measure is that, for an issue to be politically important, respondents have to feel strongly about the range of alternatives that other respondents frequently hold. Healthcare in the UK is a good example of the need to consider the distribution of opinion.  One possible position that we presented on this issue is to privatize the National Health Service and let companies charge what they want for medical care. This position has a *really strong* impact on people’s choices: very few people say they would vote for candidates who want to privatize the NHS. But this position also isn’t one which many people support. We therefore weight how much people dislike different political positions by the frequency with which those positions are held in the population. For an issue to be important as we understand it, people need to really dislike positions that many other people say they support. 

One key finding is that while obvious high profile issues like the future relationship of the UK with the EU are important by our measure, there are also issues ignored by current political contestation where the public disagrees intensely.  The death penalty has not been used in the UK since the 1960s and is rarely a subject to political debate, but if parties were to take up differing positions on this issue, our experiment suggests that it would become very contentious.  There are substantial factions among UK citizens on opposing sides of the issue and they seem to put a lot of weight on the issue in candidate comparisons.  That lack of current contestation on this issue might be good or bad — but the latent conflict is something that our method can identify which previous methods cannot.  ​ 

About the Authors: Chris Hanretty is Professor, Royal Holloway, University of London, School of Politics, International Relations and Philosophy, Benjamin E. Lauderdale is Professor, University College London, Department of Political Science and Nick Vivyanis is Professor, Durham University, School of Government and International Affairs. Their research “A Choice-Based Measure of Issues Importance in the Electorate” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.