New Paper Examines Impact of "Convenience" on Voters With Disabilities

[Image courtesy of BET]

The question of how best to assist voters with disabilities is important, but often difficult to address because there is so little empirical data on how those voters experience the system. Fortunately, we are starting to see researchers trying to overcome that lack of information by using existing surveys that capture voting and disability information – though not always linked – in a such a way as to give us a fuller understanding of how those voters interact with the voting process.

The latest example is a paper by Penn’s Peter Miller and UC-Irvine’s Sierra Powell entitled “Overcoming Voting Obstacles: The Use of Convenience Voting by Voters With Disabilities” which will be published in an upcoming issue of American Politics Research.

In short, Miller and Powell find that people with disabilities vote at lower rates but are more likely to participate when given the option of casting votes by mail. Here’s the abstract:

We examine the extent to which Americans with disabilities vote at the polls or take advantage of convenience voting reforms relative to voters without disabilities. The Help America Vote Act (2002) sought, in part, to increase polling place accessibility for the voters with disabilities, with a particular focus on those with vision disabilities. We construct two operationalizations of disability from the November Voting and Registration Supplement of the Current Population Survey over eight elections. A multinomial logistic model shows voters with a disability are significantly less likely to vote but are more likely to vote by mail ballot. Early in-person voting reforms have a marginal effect on the voting behavior of voters with disabilities. A matching model reveals adopting no-excuse absentee voting, permanent absentee voting, or both reforms increases the likelihood of voters with a disability casting a mail ballot.

The authors seek to test three hypotheses:

Hypothesis 1 (H1): Having a disability has a negative effect on one’s likelihood of voting.

Hypothesis 2 (H2): Having a disability has a negative effect on one’s
likelihood of voting early in-person.

Hypothesis 3 (H3): Having a disability has a positive effect on one’s likelihood of using a mail ballot.

They do so by constructing two separate measures of disability using Current Population Survey (CPS) data:

The ADA defines disability as “a physical or mental impairment that substantially limits one or more major life activities of such individual.” The CPS includes two operationalizations of disability: A set of 6 questions asking respondents whether they have a specific disability, which we refer to as the “pan-disability measure,” and a set of 12 questions that include a response option indicating disability, which we use to identify respondents with a “labor-precluding disability.” The set of questions related to the labor-precluding disability measure is present in each CPS survey; the set of questions we use to create the pan-disability measure is present in the 2008, 2010, and 2012 surveys. [p. 10, citations omitted]

Based on their analysis, Miller and Powell find they were correct on two out of three of their original hypotheses:

We expected disability to have a negative effect on voting (H1), and–all else being equal–we find strong support for this hypothesis. Both measures of disability significantly increase the likelihood a voter will abstain from voting …

There is no clear relationship between disability and voting early in-person (H2). The change in the likelihood of voting early in-person for respondents with a labor-precluding disability is in the anticipated direction, but the effect is indiscernible from zero…

[The data] supports our hypothesis that having a disability and mail voting are positively related (H3) …The decrease in the likelihood of voting for each operationalization of disability is almost offset by an increase in the likelihood of voting by a mail ballot. On average, having a labor-precluding disability increases the likelihood of abstaining by about 78%, while the same status increases the likelihood of voting by a mail ballot by about 68%.
Similarly, one of the pan-disability types not only increases the likelihood of abstaining by about 69% but also increases the likelihood of casting a mail ballot by about 51%. [p. 18]

In conclusion, the authors suggest that polling place accessibility, a key feature of HAVA, is not having the expected impact – but that growing availability of vote by mail is providing an avenue for participation for voters with disabilities:

These data suggest reforms designed to increase the accessibility of the polling place for voters with disabilities (e.g., HAVA) will have a minimal effect on turnout among people with disabilities. There is little evidence to suggest voters with a disability cast their ballots at the polling place on Election Day. Instead, these data show voters with a disability in large part cast mail ballots. [p. 21]

Best of all, Miller and Powell sketch out a potential future research agenda for anyone interested in digging further into the issue:

A panel study of the voting behavior of respondents with disabilities may shed additional light on the effect of convenience voting reforms on voting over time. We hope future research in this area would expand on these findings to consider the role of political parties and interest groups in working with people with disabilities to take advantage of convenience voting reforms while striving to maintain ballot secrecy for people with disabilities who opt to use a mail ballot. [p. 21, references omitted].

Thanks to Miller and Powell for their work – and to UC-Irvine’s Rick Hasen, who originally shared this research on his Election Law Blog. This is an fascinating look at an important issue, and I hope the authors get takers on their research suggestions (if they don’t do it themselves)!

Stay tuned …