![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
Millions of American citizens are denied the right to vote due to their criminal record. This study finds evidence that restoring the right to vote causes formerly disenfranchised individuals to develop higher levels of internal and external efficacy, generating citizens who are more confident in their own abilities and have stronger tendencies toward democratic engagement. A field experiment is embedded within a panel survey conducted before and after a statewide election in Virginia. All subjects are American citizens with a felony conviction who were once disenfranchised, but are now either eligible to vote, or are eligible to have their voting rights restored. Experimental treatments provide information about recent changes in restored voting rights, along with varying information and assistance with voter registration and turnout. The treatment that included a mobilization element generated the highest levels of efficacy. Additionally, the treatment that only included information about restoring voting rights also generated significant increases in political efficacy on its own. Thus the study finds novel evidence that the right to vote directly increases political efficacy.
Acknowledgements
Thank you to Ray Dutch, Lisa Hill, Marc Meredith, Melissa Michelson, Paul Legunas, Amy Lerman, Dave Nickerson, and Jill Sheppard for vital input and feedback during various stages of designing and analyzing this project. The experimental designs also benefitted from feedback received at the ISPS Experimental Research Methods seminar at Yale University, Bobst-CSDP Workshop on Experimental Research Design at Princeton University, Center for the Study of American Politics Workshop at Yale University, American Politics Seminar at New York University, Workshop on Experimental Research Design at the University of Pittsburgh, and the Evidence in Governance and Policy Workshop in Bogota, Colombia. The field experiment was funded through the Stephen D. Manners Faculty Development Award at the University of Pittsburgh. The field experiment was preregistered with a pre-analysis plan through EGAP: ID #20171109AA.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes on contributor
Victoria Shineman is an Assistant Professor of Political Science at the University of Pittsburgh.
Notes
1 Recruitment materials were placed online and in print in a variety of newspapers (Urban Views Weekly; Richmond News; Richmond Free Press), job search websites (indeed.com; craigslist.org; classifiedads.com), through Facebook ads, and were also distributed through flyers posted at outdoor street festivals, public transportation hubs, bodegas, and in offices and organizations that provided public services. Recruitment materials did not mention voting rights or any other political content. A sample recruitment ad is provided in the Appendix.
2 The complete materials for all experimental treatments can be viewed in the Appendix.
3 All volunteer opportunities were for events taking place between October 14th (the voter registration deadline) and November 7th (Election Day).
4 How strongly do you agree or disagree with the following statements: “My vote can make a difference”; “I feel politically empowered”
5 How strongly do you agree or disagree with the following statements: “I consider myself well-qualified to participate in politics”; “I feel that I could do as good a job in public office as most other people”; “I am just as qualified to serve on a jury as most other people”; “I am just as qualified to serve as a notary public as other people”
6 “How likely are you to vote in the upcoming [November 2018 General Midterm Election] / [November 2020 Presidential Election] / [November 2021 Virginia Statewide Election]?”
7 “Over the next year, how likely are you to [contribute time or money to a political campaign] / [contact an elected representative]”; “Are you willing to answer some extra survey questions for free?”
8 How strongly do you agree or disagree with the following statements: “I am currently active in my community”; “ I am aware of opportunities to be involved in my community”; “Over the next year, how likely are you to engage in the following types of participation … Volunteer Within Your Community”; “In the future, I would like to be active in my community”.
9 As specified in the pre-analysis plan, covariates include age, age2, gender, race, education, employment, and number of years at current address.
10 All directional hypothesis tests were pre-registered and specified in the pre-analysis plan.
11 One might be tempted to estimate effects among subgroups within the sample, such as generating a complier average causal effect (CACE) estimating the effects of learning one’s right to vote was restored. However, such analyses rely on the exclusion restriction to be unbiased, and this assumption is not met with the current design. There are several reasons to suspect that the treatments generated increases in efficacy beyond the effects they had through the increases in awareness of voting rights they generated. For example, subjects were informed that the Governor proactively restored voting rights to 150,000 citizens with felony convictions in Virginia. Learning about this policy (among those who had not heard of it before) might increase efficacy and confidence in oneself and one’s peer group – as the Governor made a statement that this community should be eligible to vote. Because the treatments likely caused direct increases in efficacy, models estimating complier effects would provide over-estimates. Estimating the average effect of receiving each treatment is the more conservative approach, and the estimated effects of receiving either treatment are also unbiased.
12 Overall, of the eight estimates where the pooled analysis estimated statistically significant treatment effects (1A, 2A, 2B, 3A, 3B, 3C, 4A, and 4C), the average level of power was 0.75.