SPECIAL REPORT
ON
THE IMPACT
OF INCARCERATION
AND REENTRY
from the

Rhode Island
Family Life Center

By Nina Keough
& Marshall Clement

841 Broad Street
Providence
Rhode Island
02907
phone

401.781.5808
fax

401.781.5361
www.ri-familylifecenter.org

POLITICAL
PUNISHMENT
The Consequences of Felon Disenfranchisement
for Rhode Island Communities

The 2000 Presidential election in
KEY FINDINGS
Florida, won by a mere 537 votes, was
one of the most hotly contested elections
in the history of the United States.1
• Rhode Island disenfranchises a
With its recounts, court challenges, and
greater share of its residents than
voter purges, the Florida tally raised a host
any other state in New England
of voting-rights issues. It also drew attention
• More than 15,500 Rhode Islanders
to the only legal means of denying the vote
cannot vote due to a felony
to U.S. citizens: felon disenfranchisement
conviction
laws.
Felon disenfranchisement is the legal
• 1 in 5 black men is barred from
restriction of voting rights due to a past or
voting statewide
current felony conviction, and it prevents
approximately 4.7 million US citizens from
• 1 in 11 Hispanic men is barred
voting nationwide.2 As national elections
from voting statewide
come down to the votes of a few city blocks,
policymakers are increasingly concerned
• The rate of disenfranchisement in
with the implications of felon
urban areas is 3.5 times the rate
disenfranchisement for both election
for the rest of the state
outcomes and political equity.
• More than 10 percent of South
Most media and scholarly scrutiny has
Providence residents are
focused on Southern states, states that
disenfranchised
permanently disenfranchise, and swing states
for the 2004 presidential election. But
• 86 percent of those disenfranchised
Rhode Island is home to some of the most
areCenter
not currently in prison
restrictive disenfranchisement laws
in the
The
Family Life
country. Rhode Island is the onlyi nstate
in
creases community
New England that restricts votings arights
f e t y by s u p p o r t i n g
not only while an individual is incarcerated
a n d a d v oFurthermore,
c a t i n g fRhode
o r Island’s urban residents,
for a felony conviction but for the
entire
and
the
state’s
black and Latino
ex-of fenders and their
duration of the sentence, including
time
communities,
make
up a disproportionate
families. share of those deprived
served in the community under parole and
of their voting
probation supervision.3
rights, reducing their political power
As a result, many Rhode Islanders are
statewide.
denied the right to vote, and the numbers
Whether in prison or under supervision,
of disenfranchised are only increasing. The
Rhode Islanders who have been convicted
state’s prison population has grown by 625
of a felony are shut out of the political
percent over the last 30 years.4 Rhode
process, to the detriment of themselves and
Island’s prison expansion is largely
their communities. In neighborhoods and
attributable to the war on drugs and tougher
among racial groups disproportionately
sentencing. In fact, 40 percent of Rhode
involved with the criminal justice system,
Island’s prison population is sentenced for
felon disenfranchisement deprives them of
nonviolent or drug offenses.5 As a result,
an effective political voice.
disenfranchisment is not a rare punishment,
This report answers the question: What
nor is it one reserved for Rhode Island’s
is the impact of Rhode Island’s felony
worst offenders. Rather, disenfranchisement
disenfranchisement law on the size and
affects a substantial portion of the state’s
nature of Rhode Island’s eligible voting
population in every community.
population?

2

POLITICAL PUNISHMENT: The Consequences of Felon Disenfranchisement

This report estimates the proportion of adult Rhode Islanders
prevented from voting due to a felony conviction. Our analysis
is unique in that it moves beyond a broad state-wide analysis
and compares the impact of felon disenfranchisement across
Rhode Island’s cities and towns, and even across neighborhoods
in Providence. This level of analysis is critical to understanding
the dilution of communities’ political voices at all levels of
government. Central to our analysis is the racial impact of
this law. This report illustrates how race, gender, and geography
together can have a profound affect on one’s political
opportunities.
Our results demonstrate that Rhode Island’s felon
disenfranchisement law is not an irrelevant relic of the past.
Felon disenfranchisement is a powerful feature of Rhode
Island’s political system, and one that has lasting consequences
for political representation and racial equality.

Felon Disenfranchisement
Law in Rhode Island
According to Article II, Section 1 of the
Rhode Island State Constitution, all citizens
convicted of a felony lose the right to vote
for the entire duration of their sentence.
They may not vote while incarcerated or
while living in the community on parole or
probation. They are also excluded from
serving on a jury for the duration of their
sentence.

Statewide Impact
Felon disenfranchisement is not a new practice for the state
of Rhode Island. Felon disenfranchisement laws date back to
the very first state constitution, drafted in 1842.6 In the state
constitution, convicted felons are also barred from serving
on a jury.7
A felony conviction indicates a term of sentence greater
than one year or a fine exceeding $1,000, and a misdemeanor
conviction is reserved for those sentenced to less than 1 year
or those fined less than $1,000.8 Felonies may range from
writing a bad check to murder. Nearly all sentences include
a lengthy term of probation. Rhode Island has the second
highest percentage of people on probation (3.1 percent) in
the nation.9 Most Rhode Islanders in prison or under
community supervision have been convicted of felonies.
An estimated 15,500 Rhode Islanders are unable to vote
due to a felony conviction, representing nearly 2 percent of
the state’s voting-age population (see Table 1). Eighty-six
percent of those disenfranchised are not currently in prison.
While 2,188 disenfranchised men and women are incarcerated,
13,569 of them are out of prison and living in our communities.

86% of those disenfranchised
are not currently in prison
The male disenfranchised population is more than six times
the size of the female disenfranchised population.
Rhode Island’s black and Latino communities are
disproportionately disenfranchised. The rate of
disenfranchisement for black voters is over six times the
statewide rate. The rate of disenfranchisement for the adult
Hispanic population is 2.5 times the statewide average.
Even more striking, 20 percent of adult black men are
disenfranchised, as are 9 percent of adult Hispanic men. In
comparison, the rate of disenfranchisement for white males
is 2 percent (see Table 1).

TABLE 1: Statewide Population
Disenfranchised: Number and Percent

Number of ineligible voters. Percentage of given population,
over 18 years old, unable to vote.
NUMBER

PERCENT (%)

Total

15,758

1.96

Male
Female

13,405
2,352

3.54
.55

3,494
3,003
491
8,634
7,319
1,315
2,877
2,530
347
156
142
14
59
48
11

12.07
20.45
3.44
1.27
2.30
.36
5.12
9.30
1.20
.93
1.74
.16
2.16
3.79
.75

Black
Male
Female
White
Male
Female
Hispanic
Male
Female
Asian
Male
Female
American Indian
Male
Female

Source: See Technical Notes on Page 10

3

FIGURE 1: Disenfranchisment Across Rhode Island
Percentage of town residents, over 18 years old, unable to vote.
Woonsocket
3.0%

Burrillville
1.0%

N. Smithfield
0.6%

Glocester
0.8%

Cumberland
0.8%

Smithfield
0.6%

Lincoln
0.8%

Close to 80 percent of the
disenfranchised population is from
Rhode Island’s 10 most urban
areas.

Central Falls
3.9%

N. Providence
1.5%

Scituate
0.7%

Foster
0.7%

Johnston
1.4%

Providence
4.3%

Pawtucket
2.9%
E. Providence
1.4%
Barrington
0.5%

Cranston
1.3%

Coventry
1.0%

W. Greenwich
0.8%

West
Warwick
2.1%

Warren
1.1%

Warwick
1.3%

Bristol
0.7%

E. Greenwich
0.7%

Tiverton
0.6%

Exeter
1.3%

Portsmouth
0.7%

N.
Kingstown
1.1%

Hopkinton
1.0%
Richmond
1.1%

S. Kingstown
0.3%

Middletown
1.1%

Little
Compton
0.4%

Newport
2.3%
Jamestown
0.8%

Charlestown
1.2%

Block
Island
0.2%

Westerly
1.3%

Nearly 60 percent of the
disenfranchised population comes
from just five cities: Central Falls,
Pawtucket, Providence, Newport,
and Woonsocket.

Narrangansett
0.8%

Source: See Technical Notes on Page 10

INTERPRETING THIS REPORT
Research on correctional populations are accompanied by a number of
caveats with regard to interpreting and generalizing findings, and this study is
no different. The purpose of this report, Political Punishment, is to document
the impact of Rhode Island’s felon disenfranchisement laws by conducting the
closest estimation possible given the data available.
While the state bans felons serving sentences either in prison or under
supervision, there is no completely accurate database of existing felons. The two
data files analyzed for this report contained data on people indicted for felonies
in prison, and on probation or parole. The degree of error introduced by the
use of felony indictments rather than felony convictions results in a small degree
of overestimation. However, other aspects of the analysis significantly underestimate
the impact, resulting in an overall conservative estimate.
For a complete discussion of the methodology and its implications for the
analysis, see page 10 of this report.

The impact of felon disenfranchisement
varies dramatically across Rhode Island’s cities
and towns. But, Rhode Island’s urban areas
are disproportionately disenfranchised. 58
percent of the state’s disenfranchised population
comes from only five cities: Central Falls,
Pawtucket, Providence, Newport, and
Woonsocket. Nearly 4 out of every 5
disenfranchised residents come from just 10
municipalities.
Over 4 percent of Providence’s adult
population cannot vote as a result of felon
disenfranchisement. The second most
disenfranchised city is Central Falls, where 3.9
percent of the adult population cannot vote.

4

POLITICAL PUNISHMENT: The Consequences of Felon Disenfranchisement

Urban Impact
Table 2 provides a more complete picture of the impact in
Rhode Island’s urban areas by demonstrating the rate of
disenfranchisement for each racial group, by gender, for the
state’s eight most urban municipalities.
While the rates of disenfranchisement are greater in urban
areas, regardless of race, those for the black community are
disproportionately high. In every city and town, black residents
are disenfranchised at least 4 times as often as white ones. In
both Providence and Newport more than 1 in 4 black men
cannot vote.
Hispanics also have higher rates of disenfranchisement
than urban residents as a whole. In Woonsocket, home to the
most disenfranchised Hispanic population, more than one
out of every seven Hispanic men cannot vote.
White residents are disenfranchised less, on average, than
black and Hispanic residents. The most disenfranchised white
population is found in Central Falls, where just over 3 percent
of the adult population cannot vote.
In Newport, we find the greatest racial disparity in
disenfranchisement rates for white men versus black men in
the state. Newport’s rate of disenfranchisement for black men
is 11 times that of white men.
Across Rhode Island’s urban areas, an alarming proportion
of the state’s black and Hispanic residents are barred from
the polls. In denying voting rights to so many black and
Hispanic residents, felon disenfranchisement substantially
reduces the political power of entire communities, in both
local and statewide elections. Equal opportunity in political
representation is impossible without equal representation at
the polls.

In both Providence and
Newport, more than 1 in 4
black men cannot vote.
In Woonsocket, West
Warwick, East Providence,
and Pawtucket, more than
1 in 10 Hispanic men
cannot vote.

TABLE 2: Disenfranchisement in Urban Rhode Island: Race and Gender

Percentage of given population, over 18 years old, unable to vote in each town. (e.g. 5.5% of white males in Central Falls are disenfranchised.)

TOTAL

TOTAL
MALE

FEMALE

TOTAL

%

%

%

Urban Core
Central Falls
Pawtucket
Providence
Newport
Woonsocket

3.9
3.1
4.3
2.3
2.9

6.9
5.6
7.7
4.1
5.2

Urban Ring
Cranston
East Providence
North Providence
Warwick
West Warwick

1.3
1.3
1.5
1.3
2.0

2.4
2.6
2.8
2.4
1.2

Source: See Technical Notes on Page 10

WHITE
MALE

FEMALE

TOTAL

%

%

%

1.1
.9
1.3
.6
.9

3.1
2.2
2.1
1.4
2.1

5.5
4.0
3.7
2.4
3.8

.4
.3
.5
.3
2.7

1.2
1.1
1.4
1.2
1.8

2.2
2.0
2.5
2.3
3.3

BLACK

HISPANIC

MALE

FEMALE

TOTAL

%

%

%

%

MALE

FEMALE

%

%

1.2
.6
.7
.4
.7

13.8
10.5
14.5
14.8
12.5

20.8
17.5
26.3
26.5
19.9

3.2
2.8
4.3
3.2
3.8

4.1
6.5
5.3
2.9
7.9

7.4
11.8
9.6
5.1
15.9

.7
1.6
1.3
.5
1.6

.3
.3
.4
.3
.5

7.0
6.6
5.8
5.1
18.3

10.3
13.1
9.4
8.0
15.0

3.0
1.0
2.3
1.8
3.3

2.3
5.3
2.8
3.1
15.6

4.0
12.3
5.8
5.3
14.1

.9
.3
.2
1.1
1.6

5

Providence may be Rhode Island’s economic, cultural, and
political hub, but it is also marked by profound economic
inequality.10 The median family income in the College Hill
neighborhood is $120,783 per year, whereas the median family
income in Olneyville is less than $20,000.11 24 percent of
Providence families live in poverty.12 Providence is the most
incarcerated city in the state. While Providence residents make
up only 17 percent of the state’s population, 38 percent of
the prison population is from Providence.
Over 5,000 Providence residents are unable to vote as a
result of a felony conviction. 85 percent of those disenfranchised
are out of prison, living in the city.
While Providence is the most disenfranchised city in the
state, there is considerable variation in the distribution of that
population. In some neighborhoods, felon disenfranchisement
restricts voting rights for less than one percent of the population,
while in others, more than one in ten residents is restricted.

Impact on Providence

FIGURE 2: Disenfranchisement in Providence

More than half of the disenfranchised population lives in
just seven neighborhoods. Upper and Lower South Providence,
the most disenfranchised neighborhoods in the city, lose more
than 10 percent of their voting population. By way of
comparison, the neighborhoods of Blackstone, College Hill,
and Wayland lose less than one percent of their voting
population. The rate of disenfranchisement in Upper South
Providence is over 35 times the rate in Blackstone.
This analysis demonstrates that felon disenfranchisement
does not only punish the felon, but the entire neighborhood.
In statewide and citywide elections, the neighborhoods of
Upper and Lower South Providence cannot represent their
interests in the way East Side neighborhoods can, because so
many of their residents are barred from casting a vote. If crime
affects entire communities, so too does the loss of voting
rights for large numbers of residents. It’s not only felons who
suffer a loss of political voice, but their neighbors as well.

Percentage of residents, over 18 years old, unable to vote in each neighborhood.

TABLE 3: Disenfranchised in
Providence: by Gender

Number of ineligible voters of each gender per
neighborhood. Percentage of given population, over
18 years old, unable to vote.
#

Blackstone
Charles
College Hill
Downtown
Elmhurst
Elmwood
Federal Hill
Fox Point
Hartford
Hope
Lower South Prov.
Manton
Mount Hope
Mount Pleasant
Olneyville
Reservoir
Silver Lake
Smith Hill
South Elmwood
Upper South Prov.
Valley
Wanskuck
Washington Park
Wayland
West End
Source: See Technical Notes on Page 10

MALE

16
147
23
59
121
419
205
56
132
25
300
78
137
174
245
57
296
202
61
317
122
273
213
11
624

%

.6
6.3
.6
4.8
2.8
12.5
6.8
2.7
7.2
1.9
19.0
8.6
8.0
4.9
11.9
6.5
8.5
11.4
8.5
19.8
8.2
7.9
8.6
.8
13.0

Source: See Technical Notes on Page 10

FEMALE
%
#

3
35
2
10
31
79
42
6
20
5
77
18
26
31
50
15
83
36
7
62
25
53
31
112

.1
1.4
.8
.6
1.9
1.4
.3
.9
.3
3.9
1.6
1.1
.7
2.3
1.4
2.1
1.8
.8
3.5
1.6
1.3
1.2
2.1

6

POLITICAL PUNISHMENT: The Consequences of Felon Disenfranchisement

FIGURE 3: Black Men
Disenfranchised in Providence

Percentage of black male population, over 18 years old, unable to
vote in each census tract.

Providence
Citywide:
26.3%

Source: See Technical Notes on Page 10

FIGURE 4: Hispanic Men
Disenfranchised in Providence

Percentage of Hispanic male population, over 18 years old, unable
to vote in each census tract.

Providence
Citywide:
9.6%

Racial Impact
Providence is one of the most diverse cities in the state.
While Providence claims 17 percent of the state’s population,
it is home to 50 percent of the state’s nonwhite population.13
12.7 percent of Providence residents are black or African
American, compared to 5 percent for the state as a whole.
30 percent of its residents are Hispanic, compared to 8.6
percent statewide. Furthermore, the Hispanic population
in Providence has doubled in the last ten years.14
Black men are the most disenfranchised group in the
city. Rates of disenfranchisement for adult black men
exceed the state average in more than half of all Providence
neighborhoods. More than 40 percent of black men
cannot vote in some parts of South Providence, the West
End, Silver Lake and Olneyville.
Hispanic men have the second highest rate of
disenfranchisement. Nearly 1 in 10 Hispanic men in
Providence have lost their voting rights. More than 15
percent cannot vote in parts of the West End.
Table 4 indicates that men ages 18-35 are the most
disenfranchised population in the state. Residents of the
Southside of Providence are the most disenfranchised,
where 2 out of every 5 black men, ages 18-34, cannot vote.
The rates of disenfranchisement for Asian and white men
on the Southside is more than four times their rates
statewide.

TABLE 4:
Disenfranchised Men: 18-34 years old
Percentage of given male population, ages 18-34,
unable to vote.
RI

Providence

Southside*

%

%

%

Total

5.0

8.0

15.7

White

3.3

3.3

10.4

Black

24.8

32.1

40.0

Hispanic

10.0

10.4

10.9

Asian

2.3

2.8

7.1

AIAN**

4.5

2.9

8.6

* Southside refers to four Providence neighborhoods: West End, Elmwood,
Upper South Providence, and Lower South Providence.
** AIAN is a U.S. Census category abbreviation for American Indian or
Alaskan Native.

Source: See Technical Notes on Page 10

Source: See Technical Notes on Page 10

7

Providence has a smaller white population (45.8%) than the
rest of the state (81%). Although the Asian and Asian American
population in Providence is small (5.9%) it amounts to 44
percent of the Asian population statewide. Less than 1 percent
of Providence residents are American Indian or Alaskan
Natives.15
White men are less disenfranchised than black or Hispanic
men in every neighborhood in Providence. In the most
disenfranchised neighborhoods, on the Southside, 8 percent
to 28 percent of white men cannot vote.
Asian men are the least disenfranchised (2.2%) group in
the city, but are disproportionately concentrated in certain
neighborhoods. While in most of the city less than 5% are
disenfranchised, in parts of two neighborhoods (Lower South
Providence and the West End) more than 14 percent of Asian
men cannot vote.
There is substantial variation in the disenfranchisement
of American Indian and Alaskan Native men. In most of the
city, no American Indian or Alaskan Native men are
disenfranchised, reflecting the small size of their populations.
Where present, though, American Indian and Alaskan Native
men have some of the highest rates of disenfranchisement in
the city. In parts of Smith Hill and the West End, 20 to 30
percent of American Indian or Alaskan Native men are
disenfranchised.

FIGURE 5: White Men
Disenfranchised in Providence

Percentage of white male population, over 18 years old,
unable to vote in each census tract.

Providence
Citywide:
3.7%

Source: See Technical Notes on Page 10

8

POLITICAL PUNISHMENT: The Consequences of Felon Disenfranchisement

Conclusion: Political Consequences of
Felon Disenfranchisement

Felon
disenfranchisement
dramatically diminishes the
political power of cities and
neighborhoods.
With this loss of electoral
power, the neighbors of
felons are subjected to some
of the same punishments as
felons themselves.

In this report, we examine the impact of Rhode Island’s felony disenfranchisement
law on the size and nature of eligible voting populations across the state. Our
study finds that urban areas lose the largest share of their voting population under
the law. Providence loses more of its voters than any other town. Furthermore,
within the city of Providence, we found enormous disparity in the rates of
disenfranchisement across neighborhoods, with some neighborhoods losing the
votes of more than 10 percent of their residents.
The racial impact of felon disenfranchisement is troubling. A full 12 percent
of African-Americans are disqualified from voting. More than 20 percent of black
men cannot vote, 1 out of 5 statewide. 5 percent of Rhode Island’s Hispanic
population cannot vote, and over 9 percent of Hispanic men. At the municipal
level, rates of disenfranchisement for black and Hispanic residents are even higher.
Up to 25 percent of black men are disenfranchised in some cities, and, in others,
18 percent of Hispanic men cannot vote. In some neighborhoods in Providence,
over 40 percent of the young black male population is barred from the political
process.
Comparisons by race, gender, and region show that felon disenfranchisement
is more than a personal punishment for the offender. Felon disenfranchisement
dramatically diminishes the political power of cities and neighborhoods. In statewide
and citywide elections, communities with high incarceration rates lose a
disproportionate share of their voting population, and with it, electoral clout. With
this loss of electoral power, the neighbors of felons are consequently subjected to
some of the same punishments as felons themselves.
Felon disenfranchisement has significant consequences for racial equality in
political representation in Rhode Island. Community advocates have long fought
to redraw district boundaries to ensure majority status for a particular racial group.
Racial disparities in felony disenfranchisement, and the dilution of political power
in Rhode Island’s communities of color, reverse this important effort and undermine
the fairness of our political system.
Felon disenfranchisement does nothing to improve public safety in Rhode
Island. Restricting voting rights does not prevent felons from committing crimes.
It does not provide compensation to victims, nor does it rehabilitate the person
convicted of a felony.16
Expanding the franchise for people on probation or parole may actually increase
public safety. 99 percent of those who enter Rhode Island prisons will eventually
be released. Encouraging felons to engage in society, through their behavior, but
also through political expression, may in fact help them commit to a positive life.
Participation in the electoral process may help people with felony records develop
a sense of social responsibility, community, and civic engagement.17 Re-enfranchisement
could be a vital step in the reentry process.
Most Americans support extending voting rights to ex-felons. A recent survey
found that 80 percent of respondents believe that ex-felons who have served their
entire sentence, regardless of type of offense or length of sentence, should have
the right to vote. Additionally, 64 percent maintain that people on probation and
living in their communities should have the right to vote.18
Many states have recently reconsidered and revised their felony disenfranchisement
laws. The states of Nevada, Pennsylvania, Delaware, New Hampshire, Wyoming,
Alabama, Texas, Virginia, New Mexico, and Maryland have already enacted legislation
to expand the voting rights of formerly incarcerated individuals. In 2001, Connecticut’s
legislature voted to remove the ban for probationers, leaving Rhode Island as the
only New England state with such restrictive practices.19
Rhode Island has received national attention on the issue of prisoner reentry,
rehabilitation, and public safety. In order to maintain a reputation for national
excellence, Rhode Island must revisit the issue of felon disenfranchisement.

9

Note on Race Definitions
Definitions of race vary from study to study. This
study requires a comparison between two data
sources: the 2000 United States Census and the
Rhode Island Department of Corrections (RIDOC)
INFACTS data system. Population totals reflect
the 2000 United States Census definitions of race.
Data used to estimate disenfranchisement totals
reflect racial definitions used by RIDOC. Estimating
the proportion of a given racial population
disenfranchised thus required a comparison of
incongruous definitions of race.
When the study refers to white individuals, 2000
Census estimates refer to all individuals who
reported an ethnicity of non-Hispanic and only
one race: white. The RIDOC data estimates refer
to all individuals who self-reported the race white,
and may include Hispanic whites as well as
multiracial individuals. As a result of a more limited
Census category, white disenfranchisement estimates
are slighly increased.
When the study refers to black individuals, 2000
Census estimates refer to all individuals who
reported an ethnicity of non-Hispanic and only
one race: black or African-American. The RIDOC
data estimates refer to all individuals who selfreported the race black and may include Hispanic
blacks as well as multiracial individuals. As a result
of a more limited Census definition, black
disenfranchisement estimates are slighly higher.
When the study refers to Hispanic, 2000 Census
estimates refer to all individuals who reported the
ethnicity Hispanic and any other race. The RIDOC
data estimates refer to all individuals who selfreported the race Hispanic. This comparison slightly
underestimates the disenfranchised Hispanic
population.
When the study refers to Asian individuals, 2000
Census estimates refer to all individuals who
reported an ethnicity of non-Hispanic and only
one race: Asian or Pacific Islander. The RIDOC
data estimates refer to all individuals who selfreported the race Asian and may include Hispanic
Asians as well as multiracial individuals. This
comparison likely had little impact on Asian or
Pacific Islander disenfranchisement estimates.
When the study refers to American Indian or
Alaskan Native, 2000 Census estimates refer to all
individuals who reported an ethnicity of nonHispanic and only one race: American Indian or
Alaskan Native. The RIDOC data estimates refer
to all individuals who self-reported the race American
Indian and may include Hispanic American Indians
as well as multiracial individuals. This comparison
likely had little impact on Asian or Pacific Islander
disenfranchisement rates, although Alaskan Natives
did not likely report American Indian to RIDOC.

End Notes
1 David Gonzales, “The 2000 Election; Protesters; Jesse Jackson Demands
Inquiry on Florida Vote,” The New York Times, 10 November 2000, Section
A, p. 27.
2 Alec Ewald, Demos: A Network for Ideas and Action, Punishing at the Polls:
The Case Against Disenfranchising Citizens With Felony Convictions (24
November 2003), 13. < http://www.demos-usa.org/pub109.cfm>
3 Patricia Allard & Marc Mauer, The Sentencing Project, Regaining the Right
to Vote: An Assessment of Activity Relating to Felon Disenfranchisement
Laws (January 2002) <http://sentencingproject.org/pubs_05.cfm>
4 Leo Carroll, Lawful Order: A Case Study of Correctional Crisis and Reform
(New York: Garland Publishing, Inc., 1999), 329-330; Rhode Island Department
of Corrections. 2003 Annual Report. Cranston, RI: Rhode Island Department
of Corrections, 2003.
5 Carroll, 213-16; Rhode Island Department of Corrections, 2003 Annual
Report, 14.
6 Chistopher Uggen & Jeff Manza, The Sentencing Project, “Summary of
Changes to State Felon Disfranchisement Law 1865-2003,” (April 2003)
<http://sentencingproject.org/pubs_05.cfm>
7 Rhode Island State Constitution, Article II, Section 1.
8 Rhode Island General Laws, Section 11-1-2.
9 Lauren Glaze & Seri Palla, “Probation and Parole in the United States, 2003,”
Bureau of Justice Statistics Bulletin, Office of Justice Programs, July 2004.
10 The Providence Plan, “Presentation to Mayor Cicilline's Transition Working
Groups,” Powerpoint Presentation, November 26, 2002. Available at
<http://www.provplan.com/download/CTT_Final.ppt>
11 The Providence Plan,“2000 Median Family Income Summary Table.” Available
at <http://provplan.org/nprof/cw_mrecon.html>
12 The Providence Plan, “Census Quick Fact Sheet”. Available at
<http://www.provplan.com/nprof/censummary.html>
13 The Providence Plan, “Presentation to Mayor Cicilline's Transition Working
Groups.”
14 Ibid.
15 The Providence Plan, “2000 Racial and Ethnic Breakdown”, Available at
<http://www.provplan.com/nprof/charts/combpie.gif> & U.S. Census Bureau,
2000 Census of Population and Housing, Summary Population and Housing
Characteristics, PHC-1-41, Rhode Island (Washington, DC, 2002) p. 19.
Available at <http://www.census.gov/prod/cen2000/phc-1-41.pdf>
16 Ewald, 23-24.
17 Christopher Uggen, "Barriers to Democratic Participation ," paper presented
at the third Reentry Roundtable on Prisoner Reentry and the Institutions
of Civil Society: Bridges and Barriers to Successful Reintegration. (Washington,
D.C.: Urban Institute, March 2002).
18 Jeff Manza, Clem Brooks, and Christopher Uggen,“Civil Deaths or Civil
Rights? Public Attitudes Towards Felon Disenfranchisement in the United
States,” upcoming in Public Opinion Quarterly. (Volume 68), 2004
19 Allard & Mauer.

10

POLITICAL PUNISHMENT: The Consequences of Felon Disenfranchisement

Technical Notes
Data Sources
1. Rhode Island Department of Corrections Data Files of Sentenced and Supervised Offenders. As of September 30, 2003.
2. Census 2000. SF1: PCT12 Sex by Age; PCT12H Sex by Age (Hispanic or Latino); PCT12I Sex by Age (White Alone, Hispanic or
Latino); PCT12J Sex by Age (Black Alone, Hispanic or Latino); PCT12K Sex by Age (AIAN Alone, Hispanic or Latino); PCT12L Sex
by Age (Asian Alone, Hispanic or Latino).
Methodology

Given the absence of a completely accurate database of
individuals currently serving felony sentences in prison, on
parole, or probation, this study used RIDOC data to best
estimate the impact of felon disenfranchisement laws.
Throughout this study we designated individuals as
disenfranchised if they had a felony designation in either the
sentenced data file or probation and parole data file, they
were 18 years or older, and (based on their address) were a
Rhode Island resident.
A felony designation, however, indicates an indictment
rather than conviction in the sentenced data file and in the
probation and parole data file. Since no accessible database
exists with conviction data for people currently in prison or
on probation or parole, this study uses indictment data in
order to estimate rates of disenfranchisement, resulting in a
small degree of overestimation. In order to gauge the degree
of overestimation caused by the use of indictment data, a
randomly selected sample was examined. From those people
on probation or parole predicted to be disenfranchised based
on indictment data, a random sample of 143 cases were crossexamined with information publicly available from the Rhode
Island Judiciary. Of these, 119 case identification numbers
matched data from the courts and of these, 95.8 percent were
reported as having also been convicted of a felony, while only
4.2 percent had been convicted of a misdemeanor. Given this
sample and using a 95 percent confidence level, this estimation
has a confidence interval of plus/minus 3.9 percent. Reliance
on indictment data to estimate felon disenfranchisement
complicates this research and similar studies in other states
as well. In order to arrive at the closest estimate possible we
use indictment data while taking measures that underestimate
the disenfranchised population in later parts of the analysis.
Since people in Rhode Island's criminal justice system are
not all residents of this state, this study determined residency
in the following manner. Using the last known self-reported
address in RIDOC data files, only records without an out of
state address were considered Rhode Island residents. Records
were then geocoded to census tract and Providence
neighborhood levels. 688 records (4.4 percent) did not contain
sufficient address information to geocode by municipality.
An additional 411 records (7.4 percent) were without sufficient
information to geocode to Providence neighborhoods. The
removal of these records from municipal and Providence
analyses results in an underestimation of the disenfranchised
populations.
Due to sentencing laws in Rhode Island, some individuals
appear in both the sentenced and the probation and parole
data files. Prisoners serving probation for a felony conviction
while incarcerated for a misdemeanor conviction were removed

from the analysis, resulting in underestimation. Both RIDOC
databases collect multiple offenses for each offender. All but
one most serious charge was considered for each offender for
the purposes of this study.
Citizenship and its impact on voting eligibility were not
taken into account for the purpose of this analysis. Both the
census data and the estimated disenfranchised population were
not scanned for citizenship.
Missing values in the data file for race, gender, or address
resulted in those records being removed from the analysis.
This further underestimates the disenfranchised population.
The population against which the estimated disenfranchised
population is compared for rates and percentages comes from
2000 Census estimates of population over 18, or ages 18-34
in Table 4.
Population estimates were also adjusted to correct for the
Census miscounting prisoners as residents where the prison
is located. To account for incarcerated residents, denoted as
Cranston residents in the 2000 Census (location of the Adult
Correctional Institutions (ACI)), population estimates reflect
the 2000 population total, over 18 years old, for the given
location with the addition of current inmates living in those
localities directly prior to conviction (which remains their
address according to Rhode Island's election statute section
17-1-3.1). Estimates of the Cranston population denote Cranston
residents in the 2000 Census, with the addition of inmates
living in Cranston prior to conviction, and the subtraction
of the total ACI population. This methodology was used
consistently across estimates for all racial and gender breakdowns,
reducing both the underestimation that otherwise would have
resulted from using mistakenly higher Census population
estimates in Cranston and the overestimation that would have
resulted from using lower Census population estimates in all
other localities. Rates of disenfranchisement per locality
represent the total estimated disenfranchised divided by the
total adjusted population, 18 and over, using the above changes
to the Census.
Despite the imperfections in both the RIDOC data files
and the 2000 Census as data sources for this study, they are
the best and only sources available. Throughout this study,
we have tried to arrive at a conservative estimate of Rhode
Island's disenfranchised population. The lack of conviction
data for individuals on probation and parole presents an initial
and unknown degree of overestimation. Throughout the rest
of the analysis, however, we have taken measures that
underestimate to both known and unknown degrees. On the
balance, we feel as though the methodology employed and
described above provides the best possible estimation of Rhode
Island's disenfranchised population to date.

11

Acknowledgments
This study would not have been possible without the
support of a number of organizations: The Providence
Plan, Rhode Island Department of Corrections Planning
& Research Unit, Urban Institute’s Reentry Mapping
Network, and the Annie E. Casey Foundation. The
findings of this report do not represent the conclusions
or opinions of the Rhode Island Department of
Corrections.
Special thanks to: Jim Vandermillen, Ethan Horowitz,
Holly Harriel, Jen Olivelli, Ray Perault, Pat McGuigan,
Ross Cheit, Nancy LaVigne, Marc Mauer, and Laleh
Ispahani.
The Right to Vote Campaign provided generous funding
to support the printing and distribution of this report.

RI Family Life Center
841 Broad Street
Providence, Rhode Island 02907

www.ri-familylifecenter.org