No CrossRef data available.
Published online by Cambridge University Press: 11 September 2020
The dynamics of choice and self-selection are central features of politics but absent from most experimental designs. We show how designs that incorporate choice, by allowing some subjects the option to receive or avoid treatment, can be extended by randomizing conditional on subjects’ treatment choices to answer further questions of interest while preserving statistical power. We apply this design to study how the gender of messengers for the #MeToo social movement conditions who receives the movement’s message and how they respond. Our results, from both convenience and nationally representative samples, suggest that #MeToo movement’s message reaches a wide audience with the intended effect. The potential for backlash in response to the message appears limited but more likely when this message is delivered by a woman.
Support for this research was provided by Brown University and the Carrie Chapman Catt Prize for Research on Women and Politics from Iowa State University. The authors report no conflict of interests associated with this project. No outside funding was used to conduct the study, and the authors hold no additional positions outside their academic posts. The authors thank the editor, Kaye Usry, Andy Bloeser, and three anonymous reviewers for helpful feedback. The data, code, and any additional materials required to replicate all analyses in this article are available at the Journal of Experimental Political Science Dataverse within the Harvard Dataverse Network, at: https://doi.org/10.7910/DVN/CWEHBA under Testa et al. (2020)