514 10TH S TREET NW, S UITE 1000
WASHINGTON, DC 20004
TEL: 202.628.0871 • FAX: 202.628.1091
S TAFF@S ENTENCINGPROJECT.ORG
WWW.S ENTENCINGPROJECT.ORG

STUDIES OF VOTING BEHAVIOR AND FELONY DISENFRANCHISEMENT
AMONG INDIVIDUALS IN THE CRIMINAL JUSTICE SYSTEM IN NEW
YORK, C ONNECTICUT, AND OHIO

ERNEST DRUCKER AND RICARDO BARRERAS

MONTEFIORE MEDICAL CENTER/ALBERT EINSTEIN COLLEGE OF MEDICINE
September 2005

This report was written by Ernest Drucker and Ricardo Barreras, both of Montefiore Medical
Center.
The Sentencing Project is a national non-profit organization engaged in research and advocacy
on criminal justice policy issues. Funding for this project was made possible by support from the
Ford Foundation. The Sentencing Project is a partner in the Right to Vote Campaign, a national
campaign to remove barriers to voting faced by people with felony convictions. Further
information can be found at www.righttovote.org
The authors would like to acknowledge support from the JEHT Foundation and the Open Society
Institute, along with the following organizations and individuals who collaborated in the work
and provided invaluable resources: JoAnne Page, David Rothenberg, and Barry Campbell (The
Fortune Society); Joyce Hamilton Henry and Kathleen Miller (DemocracyWorks); Maureen
Price-Boreland (Community Partners in Action); and Molly Weiser (the Racial Fairness Project).
Ricardo Barreras was supported as a postdoctoral fellow in the Behavioral Sciences Training in
Drug Abuse Research program sponsored by Medical and Health Research Association of New
York City, Inc. (MHRA) and the National Development and Research Institutes (NDRI) with
funding from the National Institute on Drug Abuse (5T32 DA07233). Points of view, opinions,
and conclusions in this paper do not necessarily represent the official position of the U.S.
Government, Medical and Health Association of New York City, Inc. or National Development
and Research Institutes.
Copyright © 2005 by The Sentencing Project. Reproduction of this document in full or part in
print or electronic format only by permission of The Sentencing Project.
For further information:
The Sentencing Project
514 10th St. NW
Suite 1000
Washington, D.C. 20004
(202) 628-0871
www.sentencingproject.org

Introduction
This report outlines the findings of research examining key dimensions of felon
disenfranchisement – the laws (varying by state) that temporarily or permanently take away
voting rights following a felony conviction. These studies build on the growing body of literature
(Uggen & Manza, 2002; Fellner & Mauer, 1998) assessing the collective impact of laws on
people with felony convictions in the US – affecting over four million individuals in 2004. This
issue has received growing attention in the past decade as its large scale, the extent of racial
disparities, and its consequences for the electoral process have become more fully understood.
The study by Uggen and Manza has done much to bring attention to this issue by demonstrating
the potential impact of felony disenfranchisement laws on electoral outcomes in many states and
at the national level, most famously in the presidential race of 2000 where the Florida outcome
was decided by 537 votes while over 600,000 voters, disproportionately African American, were
disenfranchised.
Most of the previous research into felon disenfranchisement has examined the consequences of
de jure (legal) disenfranchisement. However, our research strongly suggests that the effect of
these laws extends beyond those who are legally disenfranchised – leading to a de facto
disenfranchisement of many eligible voters. A recent study published by the University of
Virginia Law School (2003) suggests that the effects of felony disenfranchisement laws may
spill over to populations of non-disenfranchised people, finding that general voting rates,
particularly in African American communities, in the states with more restrictive laws are lower
than in those with less severe laws.
Further, some recent reports have documented the lack of knowledge of felon
disenfranchisement laws among both persons with felony convictions and those who work with
them (parole and corrections officers, social workers), including election officials in Ohio,
Pennsylvania, and New York. And there are few mechanisms in place upon completion of their
period of disenfranchisement to inform persons who have lost their rights about what measures
they must take to restore them and vote again. Much anecdotal evidence from those who work
with this population has highlighted the level of misinformation about voting rights that exists.
Not everyone who is now, or has ever been, in the criminal justice system necessarily becomes
ineligible to vote and many people who have been convicted of crimes are currently eligible to
vote. In all but 13 states in which some or all persons remain prohibited from voting after the
completion of sentence people can vote once they complete their sentence to prison and parole,
and in two states (Vermont and Maine, as well as in Puerto Rico) prisoners may vote while still
incarcerated. In addition, in a number of states, persons on probation or parole, and those in jail
awaiting trial (or serving sentences for misdemeanors), are generally eligible to vote.
Finally these rules are being examined and change has occurred in many states. For example,
Iowa’s governor has recently issued an executive order restoring the right to vote to all persons
who have completed supervision, while Alabama legislators recently passed a bill permitting
expedited restoration for most offenders upon completion of sentence. Table 1 outlines the
voting laws for the three states in this study. The principal difference in felony
disenfranchisement laws across these three states is that people on parole are allowed to vote in

1

Ohio, but not in New York or in Connecticut (which restored voting rights to probationers in
2002).
Table 1: State Policies on Felony Disenfranchisement
State
New York
Connecticut
Ohio

Prison
FD
FD
FD

Parole
FD
FD
None

Probation
None
None (since 2002)
None

While previous research has demonstrated the collective impact of having large numbers of
voters (disproportionately Black and Latino) legally disenfranchised, little is known about how
populations involved with the criminal justice system (many of whom are legally eligible to
vote) understand the ways in which they are affected and what they know about restoring their
voting eligibility. The research reported here aims to build on these accounts and complement
the growing body of studies on felony disenfranchisement in three ways:
•

Most previous studies were concerned with enumerating the prevalence rates of legal (de
jure) disenfranchisement and examining this phenomenon in cross sectional studies. We have
studied the longitudinal impact of voting eligibility after disenfranchisement, measured over
an individual’s voting lifetime, and calculated the proportion of that person’s voting life lost
to felony disenfranchisement.

•

Examining the widely held belief that “these people don’t vote anyway,” our research
surveyed voter registration and voting behavior before and after periods of
disenfranchisement, as well as interest in voting during periods of eligibility.

•

In order to assess the extent of de facto disenfranchisement, we have analyzed how well the
applicable laws are understood (or misunderstood) by the affected population, as well as their
beliefs about how these rules affect their future right to vote.

To achieve these objectives the current study examines voting histories and individual
knowledge and attitudes about disenfranchisement among samples of persons currently or
formerly under correctional supervision in three states: New York, Connecticut, and Ohio. These
surveys were conducted in the summer and fall of 2004 in collaboration with community based
organizations in each state. Access to the samples included in our studies was gained through
these collaborations and their programs that provide services and support to persons with felony
convictions, many of whom are currently, or have been, affected by their state’s felony
disenfranchisement laws.

2

Specific Research Questions
1. What is this sample population’s felony disenfranchisement history?: Determine the
proportion of these individual’s adult lives in which they have been legally disenfranchised
(the cumulative impact of felony disenfranchisement on individuals over time) and how this
experience is distributed across different populations.
2. What are the registration and voting histories of this sample population prior to and after
their individual periods of disenfranchisement? What proportion ever registered and voted
before and after periods of ineligibility and, if they do re-register, is there a time gap between
the end of their period of de jure disenfranchisement and their resumption of
registration/voting?
3. What is this sample population’s knowledge of disenfranchisement laws, what are their
sources of information about those laws, and what are their current attitudes about voting?
To what extent is this population aware of applicable disenfranchisement laws and to what
extent are they misinformed?
4. What are this sample population’s intentions about future voting and political interest and
orientation? Do they plan to vote in the next election, do they have a political affiliation or
favor one party?

Methodology
Instruments and Procedures
For our research we have employed a new methodology to quantify an individual’s lifetime
exposure to disenfranchisement and assess patterns of ineligibility over the course of their adult
lives. The criminal justice calendar (CJCal) is an interview instrument that we have developed
and used to survey over 300 individuals with histories of criminal justice involvement. The
CJCal retrospectively collects detailed chronological data on an individual’s criminal justice
involvement. The main component of the instrument is a year-by-year calendar (beginning in
1950 and ending with the current year). For each year, there are five different forms of criminal
justice involvement – juvenile detention, probation, parole, jail, and prison. Participants are
asked to identify every year of their life in which they have been under some form of
correctional supervision, and to indicate that specific type for each year. For example, if they
were on probation from 1978 to 1982, they would check probation for each of those five years.
They are asked to do the same for each episode of criminal justice involvement during their
lifetime.
In the current study, we have used a brief self administered version of the CJCal to collect and
assemble data establishing the association between periods of disenfranchisement (prison, and/or
parole or probation, depending on state law) and voting and registration histories of the sample
population. This system allows us to calculate each individual’s disenfranchisement history (in
the calendar format) and link it to registration and voting both prior to disenfranchisement and
after legal entitlement to restoration of voting eligibility. This also allows calculation of each

3

individual’s (and the entire population’s) percentage of “voting eligible life“ (years >18) that has
been lost to disenfranchisement.
The calendar methodology is not limited to assessing criminal justice histories. Most events in
people’s lives can be assessed through the calendar as well. In the current study, we assessed
voting histories in the same manner as criminal justice histories. Participants were asked if they
had ever been registered to vote, and if so, if they had voted. They were asked to locate the years
when they were both registered to vote and when they voted. In order to aid with the recall of the
years that they voted, the researchers had a list of all election winners in the presidential, mayoral
and gubernatorial races, along with the year in which the election occurred. In testing the
instrument, it became clear that recall of the actual races and the winners of the election was
much easier than of the year of the actual election. Therefore, having this information made it
possible to pinpoint the specific years (and, in some cases, elections) in which they voted.
In addition to the CJCal items, we surveyed demographic characteristics, attitudes toward voting,
political party preferences, intentions to vote in the future, and knowledge about voting laws.
Knowledge of disenfranchisement was assessed by a series of true/false questions, such as “you
are not eligible to vote if you: 1) are currently on parole; 2) are currently on probation; and 3)
have ever been arrested.”
Sample Participants
Data collection was conducted in collaboration with organizations that provide services and
advocacy for individuals under criminal justice supervision. These organizations include the
Fortune Society in New York City, Community Partners in Action, in collaboration with
DemocracyWorks, both in Connecticut, and the Racial Fairness Project, in Cleveland. The
sample consisted of 334 individuals under various forms of criminal justice supervision.
The Connecticut sample consisted of 118 clients receiving services from Community Partners in
Action (CPA), the longest-running service provider for offenders in the state. Data collection
was carried out by staff of both CPA and DemocracyWorks (an organization that has led a
coalition of groups dedicated to restoring voting rights to those disenfranchised because of
criminal convictions), and occurred in three different CPA sites across Connecticut-- Bridgeport,
Hartford and Waterbury. The data collection sessions ranged in size from three to nine
individuals.
The New York sample consisted of 138 surveys of individuals with current or past histories of
criminal justice involvement who were either clients or staff members at the Fortune Society in
New York City, one of the largest social service and advocacy agencies for persons with criminal
convictions in Manhattan. Data was collected from three different sites: two in midtown
Manhattan and the Fortune Academy, located in Harlem, which is a temporary housing and
transitional facility for persons reentering society after serving time in prison.
The Ohio sample consisted of 78 individuals currently in custody surveyed from three different
jails in Cleveland, two of which were in Cuyahoga County and the other in Huron County. The
Ohio surveys only included a subset of questions from the surveys that were conducted in New
York and Connecticut because of logistical reasons that limited the amount of time that inmates

4

in the jails could spend with interviewers. The surveys were conducted in collaboration with
workers from the Racial Fairness Project, along with staff and clients from the Fortune Society,
who were engaged in a voter registration and information campaign at the local jails. It is
important to note that the Ohio sample included many people who had voluntarily agreed to
participate in a previous workshop on voting. Therefore, this group is likely to have higher
voting rates and more interest than the general jail population.
Sample Differences
Some demographic differences between these samples need to be highlighted, since voting rates
vary significantly by such factors as race and ethnicity, age, and education level. One of the
most substantial differences by state is in the age variable, with the oldest average in New York
City, 36, as compared to an average age of 29 in Connecticut and 30 in Ohio. Furthermore, the
Connecticut sample has the largest number of young people in the 18 – 24 year range (see Table
1). Connecticut also had the highest percentage of males, and more people with less than a high
school degree. There were additional state-based racial and ethnic differences, but no clear
patterns emerged.
Table 2: Sample Characteristics (%)
State
CT
NY
OH

Gender
M
88.1
78.6
73.3

Race/Ethnicity
W
B
H
38.1 25.4
30.5
7.6
51.0
30.3
33.3 49.3
6.7

18-24
40.7
24.6
36.0

Age
25-44
49.2
42.1
44.0

45+
9.3
11.0
10.7

Education
<HS
HS+
39.8
60.2
37.2
62.8
28.0
72.0

Results
Percent of Voting Life Lost To Felony Disenfranchisement
The first set of analyses assessed the number of years that offenders were legally disenfranchised
according to applicable laws in their respective states. The survey data permitted the calculation
of estimates of the proportion of voting eligible life (when 18 years and older) that individuals
spent under correctional supervision, and as a result, disenfranchised.
Data were collected from the NY and Connecticut samples (but not from Ohio, where a different
version of the CJCal was used that did not include these data) to calculate the impact of those
states disenfranchisement laws on the sample population. Almost two-thirds (61.4%) of this
sample (71.8% in CT and 49.6% in NY) had lost some proportion of their voting-eligible lives
due to a felony conviction. The average number of years disenfranchised was 3.6 (CT = 2.8 and
NY= 3.9), with a range of 0 – 26 years. But these values are a function of the age mix of our
sample and differ for populations sampled at different points in their experience with the
criminal justice system. For example, in both NY and Connecticut many study participants were
first offenders on probation or in Alternative to Incarceration programs and were not subject to
disenfranchisement by these states’ laws.

5

Among those who had been subject to disenfranchisement, the proportion of their adult life
affected by felony disenfranchisement is an indicator of the cumulative impact of
disenfranchisement on their potential election participation. Table 3 shows the proportion of their
adult life (when they would be eligible to vote) that has been affected by disenfranchisement.
One of every six (16.8%) had been disenfranchised for half of their adult lives or more, and one
of every three (34.6%) had been disenfranchised for at least one quarter of their lives.
Table 3: Percent of Voting Life Lost (VLL)
Percent
0%
25% or Less
25% - 50%
50% or More
% of Sample with some VLL

CT
28.2
30.8
21.8
19.2
71.8

NY
50.4
19.3
15.1
15.1
49.6

Avg.
41.6
24.0
17.8
16.8
61.4

Voting History Pre- and Post-Incarceration and Felony Disenfranchisement
Voting rates - One of the key questions to be probed with this survey was the voting patterns of
persons with a felony conviction before and after being disenfranchised. It has been widely
believed that participation is quite low. As seen in Table 4, our findings indicate that the
registration and voting rates of this population (in those periods when they are eligible to vote)
are substantial. The percentage of participants who reported they had registered to vote
sometime in their lives was 33.9% (CT), 61.9% (NY) and 57.3% (OH), with an average of 50%.
The percentage that reported voting at least once in their lives was 26% (CT), 54.7% (NY) and
38.4% (OH), with an average of 39.7%.
Table 4: Registration and Voting History
Question
Have you ever been registered to vote?
Have you ever voted?

CT
33.9
26.0

NY
61.9
54.7

OH
57.3
38.4

Total
50.0
39.7

Table 5 illustrates the differences in voting rates found among subpopulations, which echo the
patterns seen in the general population, i.e. voting rates are higher among those who are older,
more educated, and white.

Table 5: Voting Rates by Age, Race, and Education
Race/Ethnicity
W
B
Study Sample
44% 36%

H
30%

18-24
11%

6

Age
25-45
40%

45+
85%

Education
<HS HS+
29% 47%

Future Intentions to Vote and Identification with a Political Party – Other findings from the
survey similarly suggest that interest in voting and, more generally, a concern for political issues,
is shared by many in the sample. Table 6 shows the survey respondents’ expectations of voting
in the future, and having a political party affiliation. These data were collected in the summer
and fall before the 2004 general election and include those who could not vote (or did not intend
to vote) in that election. When these participants were asked if they were planning on registering
once eligible to vote, 66.4% (CT = 50%, NY = 68.3%) of the sample said yes, and 53.4% (CT =
28.3%, NY = 57.2%, OH = 70.7%) said they were planning on voting in the upcoming
presidential election. When questioned about partisan identification, 75.5% of the sample
responded that they did have some party affiliation.

Table 6: Political Affiliation and Future Intentions to Vote (% of those reporting)
Question
Intention to register once eligible to vote
Intention to vote in upcoming presidential election
Party affiliation

CT
50.0
28.3
77.5

NY
68.3
57.2
91.4

OH*
-70.7
--

Total
66.4
53.4
75.5

* some items were not surveyed in Ohio.
Change in voter registration after incarceration – Many observational accounts and published
reports expose significant barriers to registering after disenfranchised persons become eligible to
vote again. Among those who had been incarcerated, either in jail or prison, we compared the pre
and post incarceration rates of “ever registered” to “currently registered” in order to examine
changes in voting eligibility and behavior. The results, shown in Table 7, indicate a substantial
decrease in the current probability of electoral participation for the total sample. There is a 40%
drop from “ever registered”(48.4%) and “currently registered” (29.2%).

Table 7: Registration History and Current Registration Rates1
Question
Have you ever been registered to vote?
Are you currently registered to vote?
Percent Reduction in Registration Rates

CT
33.9
16.8
50%

NY
61.9
40.0
33%

Total
48.4
29.2
40%

Misinformation regarding felony disenfranchisement laws – We posed a series of true-false
questions that assessed participants’ understanding and beliefs about disenfranchisement laws.
1

These rates do not include OH data where participants had recently been through a voter education program and
registration drive, thereby inflating registration rates.

7

Questions focused on the level of information regarding the categories of criminal justice
supervision that would render an individual ineligible to vote. Questions included: “you are not
eligible to vote if you have ever been arrested” or “if you have ever been in jail.”
Table 8 illustrates the percentage of the sample that either answered the questions incorrectly as
true or chose “don’t know.” Half of the sample were grossly misinformed about the manner in
which felony disenfranchisement laws applied to them. For example, half the respondents did not
know that one can vote while currently on probation (in all three states), and slightly over half
(52.8%) did not know if ever having a previous felony conviction makes someone ineligible to
vote in these states (it does not). Many did not know if a past prison sentence (32.8%), probation
sentence (28.4%) or parole sentence (38.9%) made them ineligible to vote. And 31.1% either
believed that a previous arrest disenfranchised them, or were not sure of the correct response.

Table 8: Degree of Misinformation Regarding Disenfranchisement Laws:
Percent of Incorrect Answers and “Don’t Know” Responses
“True or False: You are not eligible to vote if…”
You have ever been convicted of a felony
You have ever been arrested
You have ever been on parole
You have ever been in jail
You are currently on probation
You have ever been on probation
You have ever been in prison

CT
66.4
37.0
44.9
33.5
41.4
30.2
37.4

NY
41.3
28.8
34.0
18.9
57.8
27.0
29.0

Total
52.8
31.1
38.9
25.9
50.0
28.4
32.8

Two additional questions highlight the level of misinformation among this sample. Participants
were asked if they were currently eligible to vote and if they would ever be eligible to vote in the
future. Table 9 shows the results of these questions, broken down by two groups: those who had
been disenfranchised at some point in their lives – the “Ever Disenfranchised” (ED) group – (i.e.,
those that had been in prison or on parole, n = 122) and those who had never been
disenfranchised (n = 119) 2 – the “Never Disenfranchised” (ND) group
Among the ED group, two-thirds either believed they were not currently eligible (41.3%) or did
not know if they were eligible (26.4%). In the ND group an even greater percentage did not
know if they were eligible (33.1%), and 10.2% believed they were not eligible. Similar patterns
emerged for the question asking about future eligibility. While almost all will be eligible to vote,
about one quarter of both groups (27.9% for the ED group and 24.7% for the ND group) either
believed they would not be eligible in the future or did not know if they would be eligible. In
reality (with the exception of non-U.S. citizens and a very small percentage that may have
received life time parole and disenfranchisement), almost all respondents will become eligible at
some point in their lives.

2

This analysis only includes the New York and Connecticut Samples

8

Table 9: Beliefs about Voting Eligibility Among Those who had been Disenfranchised at
Some Point in their Lives (ED Group) and Among those who had never been
Disenfranchised (ND Group)
Question
Are you currently eligible to vote?
Will you ever be eligible to vote?

Ever Disenfranchised
No
DK
Total
41.3
26.4
67.7
7.4
20.5
27.9

Never Disenfranchised
No
DK
Total
10.2
33.1
44.3
8.6
18.1
24.7

Sources of Information of Felony Disenfranchisement Laws – The survey also asked about
where, and from whom, participants had received information regarding felony
disenfranchisement laws. Table 10 illustrates that a majority (61%) report that they were never
told about disenfranchisement rules by officials in the criminal justice system (judges or
lawyers), or from anyone else.
Table 10: Sources of Information Regarding Disenfranchisement Laws*
Percent who said each of the following provided them
with information regarding disenfranchisement laws
Nobody
Judge
Lawyer
Corrections official, police, probation, or parole
Friend, family, or acquaintance

CT

NY

OH

Total

62.1
15.5
12.1
8.6
13.8

63.6
9.1
11.2
11.9
16.8

53.3
5.3
5.3
12.0
14.7

60.8
10.5
10.2
10.8
15.3

Some respondents in Connecticut and New York indicated they had received information from more than one source
(producing a column total larger than 100%), while in Ohio some respondents did not answer the question
(producing a column total lower than 100%).

Summary of the Main Findings
Lifetime Loss of Voting Rights
There is a significant individual and aggregate impact of felony disenfranchisement on lifetime
voting eligibility: 33%–40% of this sample’s adult voting life has been lost to de jure
disenfranchisement, and another significant portion is lost to de facto disenfranchisement. This
is largely because many are unaware of when the period of disenfranchisement is complete.
Voting Participation Rates
Prior to incarceration, voting and registration rates of this sample are comparable to patterns in
similar demographic groups in the general population. But there is a significant drop in
registration (and, consequently, voting) following legal disenfranchisement, and this de facto
disenfranchisement may account for election non-participation rates equal to those of legal
disenfranchisement.

9

Misinformation About Eligibility
There is widespread misinformation about voting laws among the affected population. Many
wrongly believe they cannot vote because they are unaware of when the period of
disenfranchisement is complete and are uncertain of their rights to restore their voter eligibility.
Nonetheless the sample was politically interested, with over 75% responding with a preferred
political party affiliation and more than half reporting a desire to vote in future elections.

Conclusions and Implications
These results present some of the first detailed data on voting patterns and beliefs among a
sample of respondents who have been disenfranchised by felony convictions. In diverse criminal
justice samples from three states, several important and consistent patterns develop: a predisenfranchisement history of registration and voting that is similar to comparable populations in
the US; a persistence of de facto disenfranchisement after legal ineligibility has ended; and
widespread misinformation about the severity of specific disenfranchisement rules that affect the
right to vote. Because of this widespread misinformation about who is actually eligible to vote
following criminal justice involvement, this de facto disenfranchisement serves as a multiplier of
de jure disenfranchisement and the loss of voting rights under current laws. If there are more
than 4 million people who are legally disenfranchised, there may be an equal number who
believe they are ineligible to vote, thus depressing registration and voting participation. There are
now substantial populations of eligible voters in all states with large disenfranchised populations
and many may be interested in voting but are uncertain of their eligibility. The primary task is
both to inform and assist eligible voters about their state laws and enable them to participate.
Most people with a current of former felony conviction are ignorant of their right to vote. While
90% of the sample knew the basic eligibility requirements for voting in the US (to be a U.S.
citizen and 18 or older), they often believed that their criminal justice involvement still
disenfranchised them, even though the majority of those persons we interviewed were eligible to
vote at the time. On all questions regarding the extent of specific felony disenfranchisement laws
affecting them, about half of respondents did not know their rights. Even among those persons
that were legally able to vote, they generally believed that their own circumstances prohibited
them from registering. Our preliminary indication of the effects of de facto disenfranchisement
on suppressing rates of registration and voting after release confirms the now widely recognized
need for more effective educational programs targeted toward reentry populations. Formal
programs about voting rights and the implementation of registration efforts should be instituted
in all states.
Our experience working with disenfranchised populations suggests a special role for peer
education in voter registration and voting participation programs. In New York and Ohio
parolees played a major role in the administration of the survey and in explaining applicable
voting laws and the procedure for registration to the participants. More generally, there should
be practical education about voting rights and the impact of disenfranchisement laws on the
specific communities from which these populations come. These educational programs (and
access to registration forms) should be offered in close proximity to release and probation/parole

10

programs, as well as in the many programs that work with re-entering populations in the
community at the time of discharge.

Directions for Further Research
There is a clear need for additional research to document registration rates and voting behavior in
the periods after release from supervision and the end of de jure disenfranchisement. This
research is feasible in many localities because registration and voting lists can be cross-matched
to the lists of released populations that can be obtained or are publicly available through most
state departments of correction. Officials and community service organizations need to develop
and test new models for educational and outreach programs designed to provide valid
information about applicable disenfranchisement rules at the time of release. Efforts that can
accomplish these goals dovetail well with the evaluation of programs to register persons with
felony convictions and facilitate their electoral participation. All such programs should include
rigorous assessment of the outcomes in actual registration and voting rates by re-enfranchised
populations.

11

References
Drucker, E. M., & Barreras, R. E. (2004). The criminal justice calendar: An epidemiological
method for assessing the population effects of incarceration. Paper presented at the American
Society of Criminology Conference, Nashville, TN., November, 2004.
Fellner, J. & Mauer, M. (1998). Losing the Right to Vote: Felon Disenfranchisement Laws in the
United States, Washington DC: Human Rights Watch and The Sentencing Project.
McLeod, A., White, I.K., and Gavin, A.R. (2003). “The Locked Ballot Box: The Impact of
State Criminal Disenfranchisement Laws on African American Voting Behavior and
Implications for Reform.” Journal of Social Policy and the Law, 11, (1), 66-88.
Uggen, C. & Manza, J. (2002). Democratic Contraction? Political Consequences of Felon
Disenfranchisement in the United States. American Sociological Review, 67, 777-893.

12