ABSTRACT
One of the big success stories with randomized controlled trials in recent years, which happened over a relatively short period of time, has been in the study of voter turnout and more generally political participation, including vote choice. The basic idea of the voter turnout experiments, sometimes called Get Out the Vote (GOTV) campaigns, is that a group of voters can be randomly assigned to receive an intervention designed to increase their participation and the outcomes, turnout, or vote choice can be compared to those observed in a control group whose members get nothing (or a placebo). It is possible to test for the effects of the intervention using public records, which are (depending upon the country) relatively easy to access. Other participation outcomes, such as interest group membership or party choice, can be targeted in much the same way as GOTV, but often involve close collaboration with an organization. The rate of change has been rapid. The landmark contemporary study
was carried out in 1998 (Gerber and Green 2000a); by 2012 there were just over 200 experiments on mail alone (Green et al. 2010) and several hundred voter turnout experiments overall (Green and Gerber 2015; Green et al. 2013). While most of these experiments are non-partisan, Green et al. (2013) draw attention to partisan experiments as a more recent area of interest where researchers work with political parties to affect vote choice as well as turnout. There are fewer experiments on other aspects of participation, though there are more of these occurring as well. This chapter starts with an account of how these experiments work in
practice. Then it offers some background information and reasons why experimentation took off in this area. After this discussion, it gives an outline of the type of studies that have emerged. These studies are put through the ten steps of experimental design as set out in Chapter 2, followed by how researchers have responded to the nine threats summarized in Chapter 3.
