Ask the Scholar

Document scope · 1 page
doc
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory. For page-specific OCR and visual context, open one of the page chats.

Scholar Source Context

Document identity
localId
122245006
label
“Empowerment Through Civic Participation: A Study of the Political Behavior of People with Disabilities” – Report to the Disability Research Consortium
core
doc
dtoType
document
pageCount
1
Source metadata
id
122245006
contentType
document
title
“Empowerment Through Civic Participation: A Study of the Political Behavior of People with Disabilities” – Report to the Disability Research Consortium
collections
Records of the Office of the Counsel to the President (Clinton Administration)
Jonathan Young's Files
imageCount
1
hasImages
yes
source
import
hasTranscription
no
Source extras
naId
122245006
levelOfDescription
fileUnit
otherTitles
42-t-7409251-20070143F-Seg3-015-010-2018
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
deb318e48029bb7b
ocrText
FOIA Number: 2007-0143-F FOIA MARKER This is not a textual record. This is used as an administrative marker by the William J. Clinton Presidential Library Staff. Collection/Record Group: Clinton Presidential Records Subgroup/Office of Origin: Counsel Office Series/Staff Member: Jonathan Young Subseries: OA/ID Number: 40202 FolderID: Folder Title: "Empowerment Through Civic Participation: A Study of the Political Behavior of People with Disabilities" - Report to the Disability Research Consortium Stack: Row: Section: Shelf: Position: S 112 1 8 2 EMPOWERMENT THROUGH CIVIC PARTICIPATION: A STUDY OF THE POLITICAL BEHAVIOR OF PEOPLE WITH DISABILITIES Final Report to the Disability Research Consortium Bureau of Economic Research Rutgers University and New Jersey Developmental Disabilities Council April 1999 Douglas L. Kruse, Ph.D. Kay Schriner, Ph.D. School of Management and Labor Relations Dept. of Rehabilitation Education and Research Rutgers University University of Arkansas 94 Rockafeller Road 346 N. West Avenue Piscataway, NJ 08854 Fayetteville, AR 72701 (732) 445-5991 (501) 575-6417 [email protected] [email protected] Lisa Schur, J.D., Ph.D. Todd Shields, Ph.D. School of Management and Labor Relations Department of Political Science Rutgers University University of Arkansas 50 Labor Center Way 434 Old Main New Brunswick, NJ 08903 Fayetteville, AR 72701 (732) 932-1743 (501) 575-6440 [email protected] [email protected] THE STATE UNIVERSITY OF NEW JERSEY RUTGERS Douglas L. Kruse Department of Human Resource Management School of Management and Labor Relations 94 Rockafeller Road Piscataway New Jersey 08854 732/445-5991 Fax: 732/445-2830 E-mail: [email protected] June 6, 1999 Jonathan Young Office of Public Liaison The White House 122 OEOB Washington, D.C. 20502 Dear Jonathan, Good to see you at SDS! Enclosed is our report, as promised. Feedback and questions are very welcome. Talk to you later! Day Douglas Kruse Sincerely yours, Associate Professor Executive Summary EMPOWERMENT THROUGH CIVIC PARTICIPATION: A STUDY OF THE POLITICAL BEHAVIOR OF PEOPLE WITH DISABILITIES Final Report to the Disability Research Consortium Bureau of Economic Research Rutgers University and New Jersey Developmental Disabilities Council April 1999 Douglas L. Kruse, Ph.D. Kay Schriner, Ph.D. School of Management and Labor Relations Dept. of Rehabilitation Education and Research Rutgers University University of Arkansas 94 Rockafeller Road 346 N. West Avenue Piscataway, NJ 08854 Fayetteville, AR 72701 (732) 445-5991 (501) 575-6417 [email protected] [email protected] Lisa Schur, J.D., Ph.D. Todd Shields, Ph.D. School of Management and Labor Relations Department of Political Science Rutgers University University of Arkansas 50 Labor Center Way 434 Old Main New Brunswick, NJ 08903 Fayetteville, AR 72701 (732) 932-1743 (501) 575-6440 [email protected] [email protected] EXECUTIVE SUMMARY Are people with disabilities as likely as those without disabilities to vote, and to engage in other forms of political and civic participation? If not, why not? Is there a role for policies to remove barriers or otherwise encourage greater participation by the 54 million Americans with disabilities? This study reports on a new national random-household telephone survey of 1,240 American citizens of voting age, conducted after the recent elections in November, 1998. The disability questions from the upcoming 2000 Census were used to identify people with disabilities, supplemented by additional questions about the nature, severity, and duration of the disability. The sample was stratified so that interviews were conducted with 700 people with disabilities and 540 people without disabilities. Survey respondents were asked standard questions about voting in 1996 and 1998, voter registration, voting mobilization by political parties and others, and perceptions of the political system. In addition, new questions were asked about actual and anticipated difficulties in voting at a polling place, transportation modes and difficulties, and personal competencies. Voter Turnout and Registration The major findings on voter turnout and registration in 1998 are: People with disabilities were on average about 20 percentage points less likely than those without disabilities to vote, and 10 points less likely to be registered to vote, after adjusting for differences in demographic characteristics (age, sex, race, education, and marital status). The voting gap between people with and without disabilities is especially wide among those who are age 65 or older, as shown in the figure below. Age Patterns in Voter Turnout, November 1998 90.0% 80.0% 70.0% Citizens without disabilities 60.0% Voter turnout 50.0% Citizens with disabilities 40.0% 30.0% 20.0% 10.0% 0.0% 18 go Age If people with disabilities voted at the same rate as those without disabilities, there would have been 4.6 million additional voters in 1998, raising the overall turnout rate by 2.5 percentage points. The one-third of people with disabilities who are employed were as likely as employed people without disabilities to vote, while the two-thirds who are not employed were 17 percentage points less likely to be registered, and 30 points less likely to vote, than non-employed people without disabilities. Political parties were less likely to contact people with disabilities in the 1998 campaigns. While people with disabilities are less likely to view the political system as responsive to "people like me," they are as likely as people without disabilities to follow government and public affairs most of the time, and their lower voter turnout is not explained by their perceptions of the political system or their perceived ability to participate. The overall disability voting gap is reduced only from 20 to 15 percentage points after accounting for other standard determinants of voting (such as mobilization, perceptions of the political system, and income). In addition to being concentrated among older and non- employed people with disabilities, low turnout is most likely among those who have difficulty going outside alone, severe visual impairments, and mental impairments. People with disabilities are more likely than those without disabilities to have encountered, or expect, difficulties in voting at a polling place. Of those voting in the past ten years, 8% of people with disabilities encountered such problems compared to less than 2% of people without disabilities. Among those not voting within the past ten years, 27% of people with disabilities would expect such problems compared to 4% of people without disabilities. The particularly low voter turnout among those who have difficulty going outside alone, despite the availability of absentee ballots, suggests that lack of access to polling places can dilute the value of voting and sense of civic duty and connection to one's fellow citizens. Problems with access to polling places as identified by respondent descriptions are split among general mobility, getting to the polling place, and being able to vote once at the polling place. Among all people with disabilities the percentages identifying such problems are as follows: Problems encountered if Problems expected if voted in past 10 years haven't voted in past 10 years Any difficulty in voting at polling place 8.2% 27.5% General mobility (walking, standing) 1.7% 10.3% Getting to polling place 1.6% 8.8% At polling place (getting inside, using booth/machine, long lines, seeing ballot) 3.9% 10.1% Other 1.4% 2.9% ii The National Voter Registration Act of 1993, also known as the "Motor Voter Act," was designed to increase voter registration by offering registration services at drivers' license agencies and other government service agencies. If this has had an effect, it appears similar for people with and without disabilities: close to one-fourth of respondents have registered since the Act took effect on January 1, 1995, close to one-tenth have registered at a driver's license agency, and close to 1% have registered at a disability agency or other public assistance agency, in both the disability and non-disability samples. Other Forms of Political and Civic Participation Apart from voting, are people with disabilities more or less likely to participate in other political and civic activities? The survey asked whether respondents had engaged in seven other forms of participation in the past 12 months: With disabilities Without disabilities 1. Contributed money to political party or candidate 13% 16% 2. Written or spoken to elected representative/official 30% 40% 3. Attended political meeting 13% 15% 4. Written letter to newspaper 7% 9% 5. Contributed money to org. trying to influence gov't policy or legislation 16% 22% 6. Otherwise worked with groups or on one's own to change gov't laws or policies 14% 19% 7. Worked with others on community problem 24% 28% The major findings with respect to other forms of political and civic participation are: People with disabilities were 10 percentage points less likely than those without disabilities to have engaged in any of these activities in the past twelve months, after accounting for demographic characteristics. The disability gap varies greatly, however, by activity. After accounting for demographic characteristics, people with disabilities were as likely to have engaged in three of these activities (attended political meeting, written letter to newspaper, otherwise worked with groups or on one's own to change government laws or policies), and were: 5 points less likely to have contributed to a campaign or candidate, 9 points less likely to have contacted an elected representative or public official, 5 points less likely to have contributed to an organization trying to influence government laws or policies, and 5.5 points more likely to have worked with others on a community problem. iii Lower participation in the above three activities, as with voting, is concentrated among those who are older and have difficulty going outside (even though the activities can be done in the home). Just under one-tenth (8%) of people with disabilities had engaged in any of the activities on a disability issue, while over two-fifths (43%) engaged in these activities only on non-disability issues. One-tenth (10%) of people with disabilities said that they had worked to change a private organization's policies regarding people with disabilities (such as through talking to business owners, or filing a lawsuit). Implications The overall results indicate that most people with disabilities feel well-qualified to participate in politics, and are as likely as otherwise-similar people without disabilities to participate in several non-electoral political and civic activities. They are, however, less likely on average to vote, make contributions to campaigns or political organizations, or contact elected representatives and public officials. The participation gaps are concentrated among people with disabilities who are non-employed, older, and have difficulty going outside alone-those who are employed and/or younger (18-44) are, in fact, about as likely as otherwise-similar people without disabilities to engage in these activities. The finding that participation is lowest among those who have difficulty going outside the home alone-despite the fact that these forms of participation can be done inside the home- indicates the importance of accessible transportation and interaction with mainstream society. Isolation and confinement to one's home can decrease "social capital"-the social skills, knowledge, connections, and identification developed from regular interaction with many other people. The loss of social capital, through isolationism or ostracism, can cause a self-perpetuating spiral into obscurity not only for the individual but also for the political ideas, experiences, and groups the individual represents. Feelings of isolation from mainstream society can only be heightened by difficulties in voting, since voting is a basic act of citizenship in a democracy. Actual or expected problems in voting at a polling place are highest among those who have difficulty going outside alone, and are high among those with severe vision impairments and mental impairments. The lack of accessible transportation and polling places marginalizes many people with disabilities, making them second- class citizens who cannot publicly join others in exercising the right to vote, and weakening their sense of connection to fellow citizens and mainstream society. Just as the low employment levels of people with disabilities represent a waste of economic resources, their low levels of voting and participation in several political activities represent a waste of political and social resources. Democracies work best with an active, engaged citizenry-only through full participation can individuals and groups expect to be adequately and correctly represented by political leaders. Participation can have important iv personal benefits as well: democratic theorists such as John Stuart Mill have argued that only through political involvement can one become truly empowered and perform as an enlightened democratic citizen. These findings support policies that seek to ensure fully accessible polling places and transportation, as well as the need to communicate the fact of greater accessibility to those who have not voted in the recent past. Among the policies that deserve serious scrutiny are: an information campaign for government service agencies to promote better implementation of the "Motor Voter Act," federal standards on polling place accessibility for all people with disabilities, technical assistance to state and local election officials to promote accessibility, and federal standards to ensure that all campaign activities (speeches, commercials, etc.) are accessible to people with various impairments. In addition, the fact that there is no voting gap between employed people with and without disabilities indicates that employment-in addition to its important economic and social benefits- can play an important role in increasing the social capital and political participation of people with disabilities. This lends further weight to the ADA's goal of increasing employment among people with disabilities. V TABLE OF CONTENTS Page i Executive Summary vi Acknowledgments Sections I. Introduction 1 Theory and Prior Literature 3 II. 10 III. Methodology IV. Demographic and Disability Characteristics 13 V. Voting and Other Forms of Participation: Basic Results 16 Demographic Characteristics and Participation 20 VI. Disability Characteristics and Participation 25 VII. Mobilization 28 VIII. IX. Political Interest, Efficacy, Competencies, and Participation 30 X. Employment and Participation 34 36 XI. Transportation and Participation XII. Group Involvement, Religious Attendance, and Participation 39 XIII. Income and Participation 41 XIV. Analyzing the Participation Gaps 44 47 XV. Implications 51 References Appendix A: Survey Instrument Appendix B: 2000 Census questions on disability Tables 1. Demographic Characteristics and Comparison to Other Samples 2. Disability Characteristics 3. Voter Turnout and Registration 4. Problems in Voting at Polling Place 5. Other Forms of Participation 6. Demographic Characteristics and Participation 7. Prediction of Voting with Demographic Characteristics 8. Prediction of Other Forms of Participation with Demographic Characteristics 9. Disability Characteristics, Participation, and Voting Difficulties 10. Prediction of Participation with Disability Characteristics 11. Mobilization in 1998 Elections 12. Political Efficacy, Competencies, and Participation 13. Employment and Participation 14. Transportation and Political Participation 15. Group Involvement, Religious Attendance, and Participation 16. Household Income and Political Participation 17. Prediction of Voting with All Characteristics Figure 1: Age Patterns in Voter Turnout, November 1998 EXECUTIVE SUMMARY Are people with disabilities as likely as those without disabilities to vote, and to engage in other forms of political and civic participation? If not, why not? Is there a role for policies to remove barriers or otherwise encourage greater participation by the 54 million Americans with disabilities? This study reports on a new national random-household telephone survey of 1,240 American citizens of voting age, conducted after the recent elections in November, 1998. The disability questions from the upcoming 2000 Census were used to identify people with disabilities, supplemented by additional questions about the nature, severity, and duration of the disability. The sample was stratified so that interviews were conducted with 700 people with disabilities and 540 people without disabilities. Survey respondents were asked standard questions about voting in 1996 and 1998, voter registration, voting mobilization by political parties and others, and perceptions of the political system. In addition, new questions were asked about actual and anticipated difficulties in voting at a polling place, transportation modes and difficulties, and personal competencies. Voter Turnout and Registration The major findings on voter turnout and registration in 1998 are: People with disabilities were on average about 20 percentage points less likely than those without disabilities to vote, and 10 points less likely to be registered to vote, after adjusting for differences in demographic characteristics (age, sex, race, education, and marital status). The voting gap between people with and without disabilities is especially wide among those who are age 65 or older, as shown in the figure below. Age Patterns in Voter Turnout, November 1998 90.0% 80.0% 70.0% Citizens without disabilities 60.0% Voter turnout 50.0% Citizens with disabilities 40.0% 30.0% 20.0% 10.0% 0.0% Age i If people with disabilities voted at the same rate as those without disabilities, there would have been 4.6 million additional voters in 1998, raising the overall turnout rate by 2.5 percentage points. The one-third of people with disabilities who are employed were as likely as employed people without disabilities to vote, while the two-thirds who are not employed were 17 percentage points less likely to be registered, and 30 points less likely to vote, than non-employed people without disabilities. Political parties were less likely to contact people with disabilities in the 1998 campaigns. While people with disabilities are less likely to view the political system as responsive to "people like me," they are as likely as people without disabilities to follow government and public affairs most of the time, and their lower voter turnout is not explained by their perceptions of the political system or their perceived ability to participate. The overall disability voting gap is reduced only from 20 to 15 percentage points after accounting for other standard determinants of voting (such as mobilization, perceptions of the political system, and income). In addition to being concentrated among older and non- employed people with disabilities, low turnout is most likely among those who have difficulty going outside alone, severe visual impairments, and mental impairments. People with disabilities are more likely than those without disabilities to have encountered, or expect, difficulties in voting at a polling place. Of those voting in the past ten years, 8% of people with disabilities encountered such problems compared to less than 2% of people without disabilities. Among those not voting within the past ten years, 27% of people with disabilities would expect such problems compared to 4% of people without disabilities. The particularly low voter turnout among those who have difficulty going outside alone, despite the availability of absentee ballots, suggests that lack of access to polling places can dilute the value of voting and sense of civic duty and connection to one's fellow citizens. Problems with access to polling places as identified by respondent descriptions are split among general mobility, getting to the polling place, and being able to vote once at the polling place. Among all people with disabilities the percentages identifying such problems are as follows: Problems encountered if Problems expected if voted in past 10 years haven't voted in past 10 years Any difficulty in voting at polling place 8.2% 27.5% General mobility (walking, standing) 1.7% 10.3% Getting to polling place 1.6% 8.8% At polling place (getting inside, using booth/machine, long lines, seeing ballot) 3.9% 10.1% Other 1.4% 2.9% ii The National Voter Registration Act of 1993, also known as the "Motor Voter Act," was designed to increase voter registration by offering registration services at drivers' license agencies and other government service agencies. If this has had an effect, it appears similar for people with and without disabilities: close to one-fourth of respondents have registered since the Act took effect on January 1, 1995, close to one-tenth have registered at a driver's license agency, and close to 1% have registered at a disability agency or other public assistance agency, in both the disability and non-disability samples. Other Forms of Political and Civic Participation Apart from voting, are people with disabilities more or less likely to participate in other political and civic activities? The survey asked whether respondents had engaged in seven other forms of participation in the past 12 months: With disabilities Without disabilities 1. Contributed money to political party or candidate 13% 16% 2. Written or spoken to elected representative/official 30% 40% 3. Attended political meeting 13% 15% 4. Written letter to newspaper 7% 9% 5. Contributed money to org. trying to influence gov't policy or legislation 16% 22% 6. Otherwise worked with groups or on one's own to change gov't laws or policies 14% 19% 7. Worked with others on community problem 24% 28% The major findings with respect to other forms of political and civic participation are: People with disabilities were 10 percentage points less likely than those without disabilities to have engaged in any of these activities in the past twelve months, after accounting for demographic characteristics. The disability gap varies greatly, however, by activity. After accounting for demographic characteristics, people with disabilities were as likely to have engaged in three of these activities (attended political meeting, written letter to newspaper, otherwise worked with groups or on one's own to change government laws or policies), and were: 5 points less likely to have contributed to a campaign or candidate, 9 points less likely to have contacted an elected representative or public official, 5 points less likely to have contributed to an organization trying to influence government laws or policies, and 5.5 points more likely to have worked with others on a community problem. iii Lower participation in the above three activities, as with voting, is concentrated among those who are older and have difficulty going outside (even though the activities can be done in the home). Just under one-tenth (8%) of people with disabilities had engaged in any of the activities on a disability issue, while over two-fifths (43%) engaged in these activities only on non-disability issues. One-tenth (10%) of people with disabilities said that they had worked to change a private organization's policies regarding people with disabilities (such as through talking to business owners, or filing a lawsuit). Implications The overall results indicate that most people with disabilities feel well-qualified to participate in politics, and are as likely as otherwise-similar people without disabilities to participate in several non-electoral political and civic activities. They are, however, less likely on average to vote, make contributions to campaigns or political organizations, or contact elected representatives and public officials. The participation gaps are concentrated among people with disabilities who are non-employed, older, and have difficulty going outside alone-those who are employed and/or younger (18-44) are, in fact, about as likely as otherwise-similar people without disabilities to engage in these activities. The finding that participation is lowest among those who have difficulty going outside the home alone-despite the fact that these forms of participation can be done inside the home- indicates the importance of accessible transportation and interaction with mainstream society. Isolation and confinement to one's home can decrease "social capital"-the social skills, knowledge, connections, and identification developed from regular interaction with many other people. The loss of social capital, through isolationism or ostracism, can cause a self-perpetuating spiral into obscurity not only for the individual but also for the political ideas, experiences, and groups the individual represents. Feelings of isolation from mainstream society can only be heightened by difficulties in voting, since voting is a basic act of citizenship in a democracy. Actual or expected problems in voting at a polling place are highest among those who have difficulty going outside alone, and are high among those with severe vision impairments and mental impairments. The lack of accessible transportation and polling places marginalizes many people with disabilities, making them second- class citizens who cannot publicly join others in exercising the right to vote, and weakening their sense of connection to fellow citizens and mainstream society. Just as the low employment levels of people with disabilities represent a waste of economic resources, their low levels of voting and participation in several political activities represent a waste of political and social resources. Democracies work best with an active, engaged citizenry-only through full participation can individuals and groups expect to be adequately and correctly represented by political leaders. Participation can have important personal benefits as well: iv democratic theorists such as John Stuart Mill have argued that only through political involvement can one become truly empowered and perform as an enlightened democratic citizen. These findings support policies that seek to ensure fully accessible polling places and transportation, as well as the need to communicate the fact of greater accessibility to those who have not voted in the recent past. Among the policies that deserve serious scrutiny are: an information campaign for government service agencies to promote better implementation of the "Motor Voter Act," federal standards on polling place accessibility for all people with disabilities, technical assistance to state and local election officials to promote accessibility, and federal standards to ensure that all campaign activities (speeches, commercials, etc.) are accessible to people with various impairments. In addition, the fact that there is no voting gap between employed people with and without disabilities indicates that employment-in addition to its important economic and social benefits- can play an important role in increasing the social capital and political participation of people with disabilities. This lends further weight to the ADA's goal of increasing employment among people with disabilities. V ACKNOWLEDGMENTS This study received valuable help from a number of sources. Major funding for the survey was provided by the New Jersey Developmental Disabilities Council through the Disability Research Consortium of Rutgers University, administered by Monroe Berkowitz and Carol Harvey. Supplemental funding to increase the size of the disability sample was provided by the School of Management and Labor Relations of Rutgers University. The survey was carried out by the Rutgers Center for Public Interest Polling, directed by Janice Ballou, with interviewing subcontracted to the New York survey firm Schulman, Ronca, and Bucuvalas. Ongoing support and advice have been provided by members of the Coalition on Access to Political Elections (CAPE). Advice on constructing the disability questions for the survey was provided by Michele Adler, Monroe Berkowitz, Tom Hale, Mitch LaPlante, Jim Robinson, Gale Whiteneck, and Barbara Wilson, and advice on political questions was provided by Lloyd Leonard, Vicki Paris, and Jim Robinson. Ken Schriner provided very useful help in data analysis. The authors are responsible for any errors or omissions. vi I. INTRODUCTION Political participation is fundamental to individual and group empowerment in a democracy. Nowhere is this more evident than in the growth and success of the disability rights movement over the past 25 years. The participants in this movement have helped empower people with disabilities through changes in public policies-most notably the 1990 Americans with Disabilities Act-and the policies and practices of private institutions. In addition, their participation has helped to reshape societal attitudes and assumptions regarding disability, increasing the visibility of people with disabilities and combatting stereotypes of people with disabilities as passive objects of charity. While there has been some attention to the development of the disability rights movement (e.g., Shapiro, 1993), and efforts by politicians to attract the votes of people with disabilities (e.g., President Clinton's and Bob Dole's appeals for the disability vote in 1996), there has been very little study of the civic and political participation of people with disabilities in general. Recent research using existing datasets has produced some evidence that voter turnout is lower among people with disabilities (to be reviewed), but these data throw little light on the reasons for the lower turnout, or on differential participation rates by type or severity of disability. Is the lower turnout among people with disabilities due to physical barriers, such as inaccessible polling places or transportation problems? Is it instead due to social factors, such as decreased social contact that lessens recruitment opportunities and social pressure to vote? Or is it primarily due to psychological factors, such as alienation from society or a lower sense of efficacy? In addition to the recent studies on voter turnout, some research has provided insights into how individuals with disabilities become engaged in other forms of political participation (Schur, 1997, 1998), but there has been no broad-based evidence comparing the types, patterns, and determinants of political 1 participation between people with and without disabilities. This study reports on a nationwide stratified-random telephone survey following the November 1998 elections, covering 700 citizens with disabilities and 540 citizens without disabilities. The questions addressed are: 1. How do registration and voting rates compare between people with and without disabilities (before and after controlling for other personal characteristics), both overall and among different demographic segments? 2. How do other forms of political participation-both electoral (e.g., working for or contributing to campaigns) and non-electoral (e.g., lobbying, attending demonstrations)-compare between people with and without disabilities? 3. To what extent can differences in electoral and other forms of political participation be traced to physical, economic, social, and psychological factors? 4. What policies can increase political participation among people with disabilities? The next section briefly discusses theories of why people engage in political and civic participation, and relates this to disability concerns and prior studies of voter turnout, as background for the current study. 2 II. THEORY AND PRIOR LITERATURE Participation theorists have divided the factors influencing participation into several categories. Verba, Schlozman, and Brady (1995) divide the factors affecting political and civic participation into three categories: resources ("Are you able to participate?"), psychology ("Do you want to participate?"), and recruitment ("Did anyone ask you to participate?"). Resources include time, money, and civic skills; psychological factors include political interest, sense of political efficacy, civic values, group consciousness and commitment to specific policies; and recruitment may come through formal or informal networks in work, church, or political organizations. Rosenstone and Hansen (1993) take a different approach, dividing the factors that affect participation into "individual influences" and "political influences." Individual influences are represented by personal costs and resources (the demands that participation makes on one's "money, time, skill, knowledge, and self-confidence") which interact with personal rewards, interests, and beliefs regarding participation. While these help predict which people participate, political influences-social networks and strategic mobilization efforts-help predict when they participate. Many researchers have found that factors identified by these theorists exert strong influences on political participation. Voting has been found to be strongly and positively related to the resources of education and income, as well as to employment and union membership which can represent recruitment and mobilization networks (Cassel and Hill, 1981; Shaffer, 1981; Wolfinger and Rosenstone, 1980; Teixera, 1987; Leighley and Nagler, 1992; Miller and Shanks, 1996). Psychological factors are also clearly important; in particular, declines in feelings of political efficacy and partisan identification are linked to the decline in voter turnout (Teixera, 1987). Age and marital status also strongly predict voter turnout, which may reflect a 3 combination of resource, recruitment, and psychological factors (Wolfinger and Rosenstone, 1980; Teixera, 1987; Leighley and Nagler, 1992; Stoker and Jennings, 1995; Miller and Shanks, 1996). Age is strongly and positively related to voter turnout, which Miller and Shanks (1996) attribute to differences between generations, since voting rates within a generation rise only through one's third election (age 26-29). While most studies find a positive relation between marriage and voter turnout, Stoker and Jennings (1995) find that recent marriage appears to have an overall depressing effect on voter turnout among young adults, which they attribute to time and energy costs of recent major life transitions. Disability can affect many of the factors thought to influence participation. Most fundamentally, many disabilities limit people's physical or mental resources, often requiring that extra time, effort, and money be spent to compensate for the limitations. Securing adequate transportation can be difficult, time-consuming, and expensive, and architectural barriers can make it harder for people with disabilities to participate in public life. People with disabilities have lower employment levels, constraining financial resources and increasing the likelihood of living in poverty.¹ In addition, they tend to have lower educational levels, diminishing the resource of civic skills provided through education. 2 Disability is often accompanied by increased isolation. Verba et al. (1995) note that people who are isolated are unlikely to learn the skills necessary for effective political participation, to develop a sense of membership in a group with a common political purpose, or to be recruited by political activists. Some previous surveys have found that people with disabilities are less likely 1 Only 24.5% of people with severe disabilities are employed, compared to 74.8% of adults without disabilities, while adults with severe disabilities are more than three times as likely to live in poverty households (Kruse, 1997: Tables 1, 6). 2 Only 60.3% of adults with severe disabilities have completed 12 or more years of schooling, compared to 80.4% of adults without disabilities (Kruse, 1997: Table 3). 4 than non-disabled persons to be involved in community life, including participation in volunteer, religious, recreational, or other groups or activities (International Center for the Disabled, 1986: 41; Louis Harris and Associates, 1998). Isolation sometimes reflects the lack of financial resources, since it is often expensive to obtain renovations or modified vehicles that increase the ability to participate in outside activities. Clearly, the lower employment rates noted above contribute to increased isolation and fewer opportunities for political recruitment. Finally, disability can have psychological effects that influence participation. Many people with disabilities encounter stigma and discrimination (U.S. Commission on Civil Rights, 1983; Yuker, 1988), which several studies have concluded is responsible for at least some of the lower employment and earnings of people with disabilities (summarized in Johnson and Baldwin, 1993). Stigma and discrimination may combine with isolation and limits on resources to diminish feelings of personal efficacy and control, which may in turn affect political participation. Stigma may often inhibit politicization (as pointed out by Hahn, 1987, and Scotch, 1988), but it can also motivate some people to engage in political action to fight stigma and discrimination, as shown by the growth of the disability rights movement (Shapiro, 1993). In sum, constraints on resources and increased isolation would be expected to depress participation among people with disabilities, while psychological factors may decrease or increase participation. Existing evidence indicates that voter turnout, which may be the most basic form of political participation, is lower among people with disabilities.⁴ Results from five studies are 3 Anspach (1979) presents a theoretical model of "stratagems for stigma management" identifying activism as one response to disability, occurring when one has a positive self-concept and rejects society's devaluation of people with disabilities. Other possible responses in her model include "normalization" (trying to appear as "normal" as possible while minimizing or rationalizing differences associated with disability), "disassociation" (resentment and avoidance of non-disabled people) and "retreatism" (giving up and withdrawing from society). 4 The first row is drawn from Schur and Kruse (1997); the second row is drawn from Current Population 5 summarized below. Election Disability Disability Non-disability year sample turnout turnout Gap 1) 1992 People w/SCI's 56% 71% 15% 2) 1994 Non-employed 33% 54% 21% 3) 1992-96 Non-employed 57% 71% 14% 4) 1996 Non-employed 44% 65% 21% 4) 1996 Disability households 33% 49% 16% These five data sources use very different samples: the first is based on a survey of New Jersey residents with spinal cord injuries (SCI's); the next three are based on respondents to national surveys who report being out of the labor force due to a disability⁵; and the fifth is based on respondents to a national survey who said that someone in the household has a disability, and answered an intention-to-vote question just prior to the 1996 elections (although the voting question was asked of the respondent, who may not have been the person with a disability). Despite important differences in the samples and disability measures, the results are fairly consistent: the turnout of people with disabilities is 14-21 percentage points lower than that of people without disabilities. Absentee voting can be an attractive alternative for people with mobility impairments or other transportation difficulties. Two of these studies indicate that absentee voting is higher among people with disabilities, particularly for those with mobility impairments from an SCI. Voters with SCI's in 1992 were five times as likely as voters in the general population to vote by absentee ballot Survey (CPS) data analyzed by Shields, Schriner, and Schriner (1998a); the third row is drawn from National Election Studies data analyzed by LoBianca (1998); the fourth row is from CPS data analyzed by Shields and Schriner; and the fifth row is drawn from a National Organization on Disability press release, December 1996. 5 A few (3%) of those in the National Election Study dataset who said they were "permanently disabled" rather than employed later indicated that they were employed, but 97% of this disability group was not employed at survey time. 6 (35% compared to 5%), while voters with disabilities in 1994 were twice as likely as those without disabilities to vote by absentee ballot (13% compared to 7%)(Schur and Kruse, 1997; Shields, Schriner, and Schriner, 1998a). Three of these data sources contain additional variables that have been used to explore how voter turnout is related to characteristics of the disability population, and the factors that may help explain the lower overall turnout of people with disabilities. One noteworthy finding in both the 1992 and 1994 election studies is that turnout continues to be lower on average for people with disabilities after controlling for a variety of personal characteristics.⁶ The size of the voting gap, though, varies along several dimensions. One key finding is that the voting gap varies with age: While voter turnout rises strongly with age in the general population, it rises only weakly with age in the disability population, and begins decreasing at about age 55. The result is that voter turnout is only slightly depressed among young people with disabilities, and more strongly depressed among older people with disabilities. Employment also appears to be a major factor. Both the 1992 and 1994 election studies found lower turnout among non-employed people with disabilities, but the former study also found that turnout was almost identical between employed people in the general population and the one- third of people with SCI's who are employed. The voting gap still exists after controlling for income in these two studies, indicating that the importance of employment is not completely explained by the resources provided by income. Transportation problems may also contribute to the low turnout of people with disabilities. The 1992 election study found that 30% of people with SCI's were not able to drive, 6 Both studies control for age, gender, race, marriage, education, and employment status, while the 1992 study also controls for weekly pay and job characteristics, and the 1994 study also controls for region, urban area, and family income. The 1992-96 study of nonemployed persons controlled for education and income but no other personal characteristics. 7 and voter turnout was 20 percentage points lower among this group (after controlling for other personal characteristics). Voting clearly does not depend on driving ability (since one can vote by absentee ballot, or be taken to the polling place by others), suggesting that the greater ease of mobility provided by driving ability may have important social and psychological effects through increasing opportunities for recruitment or identification with mainstream society. Direct measures of recruitment are available in the 1992-96 election study, where people were asked whether they had been contacted by a political party or anyone else to encourage them to vote or influence their vote. While this is a strong predictor of voter turnout in general, the non-employed people with disabilities in this study had only slightly lower levels of such contacts, so this accounted for little of their lower turnout. Similarly, people with SCI's were found to have slightly lower levels of participation in several social activities in the 1992 election study, but the lower participation did not appear to account for the turnout gap. Psychological factors have been directly examined in two studies. One finding from a 1987 Harris poll of people with disabilities is that interest in politics is a strong predictor of turnout, as it is in the general population (Shields, Schriner, and Schriner, 1998b). Feelings of political efficacy-that one is qualified to participate in politics, and that the political system is responsive to people like oneself-are much higher among voters than non-voters in the general population. The one study with comparative information indicates that political efficacy is lower among non- employed people with disabilities than among people in the general population (although this may be due to lack of employment or other factors rather than to disability)(LoBianca, 1998), but it is not a strong predictor of voter turnout in the disability population (LoBianca, 1998; Shields, Schriner, and Schriner, 1998b). The perception of people with disabilities as a minority group was linked to lower voter turnout among people with disabilities in 1984 (perhaps reflecting alienation 8 from the political system, particularly before the ADA was passed), although not in 1986 (Shields, Schriner, and Schriner, 1998b). These results indicate that psychological factors are important in voter turnout of people with disabilities, but do not point to factors that account for the lower turnout among people with disabilities. In sum, prior research provides consistent evidence of a voting gap of 14-21 points between people with and without disabilities, and several clues but no firm answers as to what accounts for the lower turnout. There has been little research on other forms of political and civic participation among people with disabilities. Schur (1997, 1998) explores the characteristics of political and group involvement of 64 people with SCI's, using in-depth qualitative interviews supplemented with questionnaire data. She finds a wide variety of responses to traumatic SCI, some discouraging but others encouraging political involvement. Political participation is discouraged by isolation and decreased resources that contribute to fatalism, as well as by efforts to "normalize" and distance oneself from others with disabilities, or by a focus on self-help following SCI. The political activists have lived with their injuries for a longer time on average than other interviewees, providing more opportunity to come to terms with SCI, join disability groups, learn about disability issues, and go through a gradual process of politicization. While feelings of political and personal efficacy contributed to political work, such work in turn led to a greater sense of efficacy for several activists, and new definitions and perceptions of discrimination and stigma. The experiences of most activists indicate that social context and circumstances are very important in who becomes politically active. 9 III. METHODOLOGY This study uses a new nationwide survey to provide data on political and civic participation among people with and without disabilities. The Rutgers Center for Public Interest Polling conducted the household telephone survey using random-digit-dialing procedures, beginning one day after the national elections on November 4, 1998 and finishing on December 17, 1998. The sample was stratified so that people with disabilities were intentionally oversampled (as described below), resulting in a final sample of 540 adult citizens without disabilities and 700 adult citizens with disabilities. To ensure that deaf people would not be undercounted, responses to dialed calls were analyzed to identify the presence of TTY machines. The survey form is reproduced as Appendix A, providing the questions and skip patterns used in coding the survey into a Computer Aided Telephone Interviewing (CATI) software package. A number of the questions use standard wording from the Current Population Survey (CPS) and American National Election Survey (ANES), to ensure high-quality questions and enable later comparisons. The nine sections of the survey are: A. Screening questions for age, citizenship, and disability; B. Disability type and duration, and need for assistance; C. Voting, registration, and mobilization; D. Political efficacy and other psychological measures; E. Other forms of political and group participation; F. Transportation; G. Employment; H. Religion; I. Demographics. The screening questions used to identify individuals with disabilities (A3-A7) were built upon the six disability questions being used in the 2000 Census (reproduced in Appendix B). Apart from changing the wording to ask about citizens age 18 or older, only one Census question 10 was re-worded. In order to capture limitations in all major life activities and not just in work (in line with the broad ADA disability definition), the Census question regarding difficulty working at a job or business was replaced with: "Of those 18 or older, does anyone have a long-term health problem or impairment that limits the kind or amount of work, housework, or other activities he or she can do?" In addition, while maintaining the Census wording, the question to identify "blindness, deafness, or a severe vision or hearing impairment" was broken into four parts to identify each of those conditions separately (A4a-A4d), and the question to identify difficulty "learning, remembering, or concentrating" was broken into two parts to identify separately any difficulty "learning" and "remembering or concentrating" (A6a-A6b). Those who answered "no" to all of the above questions were asked whether anyone in the household considered himself or herself to have a disability (A7a), or was considered by others to have a disability (A7b), using question wording from the Harris/N.O.D. disability survey (Louis Harris and Associates, 1998). In total, twelve questions were used to identify disability (counting the parts of the two split-up Census questions as separate questions). If there were positive answers to any of these twelve questions, the household was considered to be a disability household (although it should be noted that the presence of an impairment or activity limitation does not mean the person considers himself or herself to have a disability-such self-perceptions were measured separately). If the respondent referred to more than one person in answering these questions, the interviewer asked about the characteristics of the person with the most recent birthday (to ensure random selection among those with disabilities). If the respondent gave negative answers to each of the twelve disability questions, the household was considered to be a non-disability household. Interviews were completed with respondents in only the first 540 non-disability 11 households, so that more time could be devoted to interviewing people with disabilities. Resources were available to locate and conduct interviews in 700 disability households, with an additional 1900 non-disability households screened out in the course of identifying the disability households. Therefore people with disabilities were intentionally oversampled, both to increase the confidence in conclusions drawn from the disability sample and to allow meaningful analysis of different types of disabilities. In the 700 disability households a total of 788 adult citizens were identified as having an impairment or activity limitation reflected in the twelve questions, along with 625 adult citizens who do not have such an impairment or activity limitation. The non-disability households in which interviews were completed have 1055 adult citizens, while the screened-out non-disability households have 3869 adult citizens. Therefore of all adult citizens in the houses contacted, 12.4% (788/6337) are identified by these questions as having a disability. This is higher than the 11.0% in the Current Population Survey who respond that they have a "work disability" (health condition limiting the kind or amount of work they can do), but lower than the 25.1% in the Survey of Income and Program Participation (SIPP) who are age 18 or over and identified as having a disability due to a significant impairment or activity limitation. The lower prevalence in this survey compared to the SIPP is likely due to: 1) the more detailed questions asked in SIPP (27 total) to identify impairments and activity limitations, increasing the likelihood of a "yes" answer to at least one question; 2) the fact that the questions in our survey were asked at the household level, rather than about each person as in SIPP-person-specific questions are more likely to yield positive answers as the respondent focuses on each individual; and 3) differences in survey design-the SIPP disability supplement is given to households that are already comfortable with answering questions and have built a rapport with the interviewer, while this survey's disability questions are 12 asked at the beginning of a survey in which there has been no prior contact or rapport with the interviewer, so that respondents are less likely to reveal possibly stigmatizing information about impairments and activity limitations. In any survey using randomization procedures there will be differences between the population and the sample. To take account of such differences and ensure that they do not skew the results, the cases were weighted to reflect the age and regional distribution of the disability and non-disability adult populations, using data on these distributions from the Survey of Income and Program Participation. Individuals in households with several adult citizens have a lower probability of being interviewed (since only one person per household is interviewed), so that cases were also weighted by number of adult citizens in the household in order to make the results as representative as possible.⁷ In all reported figures, the cases have been accordingly weighted. IV. DEMOGRAPHIC AND DISABILITY CHARACTERISTICS How does the current sample compare to other data on people with disabilities? The main source used for generating estimates of the disability population is the Survey of Income and Program Participation (SIPP). Table 1 compares demographic characteristics of the disability and non-disability samples from the current survey and SIPP, and Table 2 makes such comparisons with respect to the distribution of impairments and activity limitations of the disability samples. The distributions of demographic characteristics, for both the disability and non-disability samples, are quite similar between the two surveys (Table 1). Women are slightly more likely to 7 Weighting by household size is theoretically correct, but is done by only one of the major national polling organizations (Voss et al. 1995). The results in this study are affected little by household size weighting. Gelman and Little (1998) also explore how single-person households may be underrepresented due to a lower likelihood that someone is at home to answer the phone; in the current survey, however, the distributions of disability and non- disability household sizes are very close to those in the general population. 13 have disabilities, as revealed by their greater prevalence in the disability sample (54.6%) than in the non-disability sample (51.1%).⁸ People with disabilities are less likely to be currently married with a spouse present, and more likely to be divorced or widowed. The race and ethnic distributions are similar between the disability and non-disability samples in both surveys.⁹ The age distributions strongly indicate that people with disabilities are likely to be older: both surveys show that people with disabilities are an average 13-15 years older than people without disabilities, and more than one-third of people with disabilities are age 65 or older. They are also more likely than people without disabilities to live in rural areas. The disability characteristics of those in the current survey's and SIPP's disability samples are compared in Table 2. (Since the disability questions were built on the 2000 Census, a better comparison can be done once the Census data are available.) While most measures are not strictly comparable due to different question wording in the two surveys, comparisons of similar questions indicate that the prevalence of impairments and activity limitations is fairly similar across the surveys. Close to three-fifths of respondents in both surveys report a limitation in work, housework, or other activities (although SIPP asks just about work and housework), while about one-fifth have a severe vision impairment and one-fifth have a severe hearing impairment. The current survey records about three-fifths (62.5%) who have difficulty with walking, climbing stairs, reaching, lifting, or carrying; less than one-tenth (8.3%) who use wheelchairs; and almost one- fourth (23.0%) who use canes, crutches, or walkers-each of these is somewhat larger than the SIPP percentage (although SIPP does not ask about reaching or carrying). One-tenth of those in the current survey's disability sample report a learning disability 8 The gender figures are identical between the current survey and SIPP samples because of the weighting of the current survey to match the SIPP sample by gender and region. 9 Smaller percentages are identified as "white" in the current survey, but this difference large disappears if one distributes those saying "other" race proportionally among the categories. 14 (10.0%), while close to one-third report difficulty remembering or concentrating (30.0%); both figures are higher than the percentages in the SIPP disability sample reporting learning disabilities and other mental conditions (although "difficulty remembering or concentrating" may be a broader measure than the mental conditions asked about in SIPP). Close to one-sixth (17.4%) of the current survey's disability sample report difficulty with activities inside the home, one-fourth (25.2%) report difficulty going outside the home alone, and almost one-third (31.9%) report needing help with any daily activities-each of these percentages is somewhat higher than in the SIPP disability sample. Finally, 3.6% of the current survey's disability sample did not report any of the foregoing impairments or activity limitations, but nonetheless reported that they consider themselves, or are considered by others, to have a disability. Of all those in the disability sample, close to three-fifths (58.4%) said that they consider themselves to have a disability, and a similar number said that they are considered by others to have a disability. The current survey asked respondents for the major medical condition(s) causing an impairment or activity limitation, and for the person's age at which this medical condition began. As shown in Table 2, about one-fifth of those in the disability sample have disabilities from childhood-beginning before age 18-with 8.1% beginning at birth or in the first year. An additional one-fifth of the disabilities began during ages 18-34, one-fifth began during ages 34-49, one-fifth began during ages 50-64, and one-fifth began at age 65 or older. Therefore while there were differences in methods and question wordings between the current survey and SIPP, the best available comparisons indicate that the distributions of impairments and activity limitations are generally similar, so that the current survey's disability sample appears representative of the disability population. 15 V. VOTING AND OTHER FORMS OF PARTICIPATION: BASIC RESULTS a) Voting and Registration Levels How do the voting and registration levels of people with disabilities compare to those of people without disabilities? The basic results are presented in Table 3, where it can be seen that 53.4% of the disability sample reports voting in November, 1998 compared to 59.5% of the non- disability sample-a statistically significant difference of 6.1%. These reported voting rates are higher than the 38.9% rate calculated from actual votes cast, reflecting overreporting of voting on the survey.¹⁰ Such overreporting is typical of voting surveys, including the CPS and ANES (Silver et al., 1986; Leighley and Nagler, 1992; Rosenstone and Hansen, 1993: 58-59). Past research has found some links between overreporting, education, and race (Silver et al., 1986), but gives no reason to think that there will be a systematic difference in overreporting between people with and without disabilities, so that differential overreporting is unlikely to bias the comparisons made here. People with disabilities were more likely than those without disabilities to vote by absentee ballot, reflecting mobility limitations associated with many disabilities. The voting gap for the Presidential election in 1996 (5.6%) is similar to that for 1998. People with disabilities are more likely to report that they last voted more than 10 years ago, or that they do not know when or if they ever voted. People with disabilities were also somewhat less likely to be registered to vote in 1998, but the gap (2.6%) is not large enough to reject sampling difference as an explanation of the difference. The National Voter Registration Act of 1993, commonly called the "Motor Voter Bill," was designed to increase registration by requiring registration services at drivers' license agencies and a 10 The initial estimate is that 72.5 million people voted in November, 1998, which is 38.9% of the 186.9 million citizens age 18 or over (http://tap.epn.org/csae/cgans4.html). 16 variety of government service agencies. About one-fourth of both the disability and non-disability samples report registering to vote since the Act took effect on January 1, 1995. There is little difference by disability status in the method of registration: the largest number (one-tenth of the overall sample) were registered at driver's license agencies, while registration offices and registration drives were the next most common method. Only about 1% of each sample reported becoming registered at a disability agency or other public assistance agency since the beginning of 1995. This strongly suggests that many states and government agencies are not fully implementing the provisions of NVRA. In fact, many believe that the benefits of the NVRA will not be obtained until much more emphasis is placed on full compliance with all aspects of the bill (Knack, 1995). As will be seen in the next section, the disability gaps in voter turnout and registration become wider after accounting for demographic characteristics (comparing people with and without disabilities who are otherwise similar in age, education, and other factors). b) Problems in Voting To gauge problems with accessibility of polling places and other difficulties with the process of voting, respondents were asked whether they had encountered (if they had voted since 1988) or would expect (if they had not voted since 1988) any "difficulty in getting to the polling place or in using the ballot or voting machine."¹¹ As can be seen in Table 4, people with disabilities were more likely to report such difficulty: about one-twelfth (8.2%) of those voting since 1988 said they had experienced such difficulty, and over one-fourth (27.5%) of those not voting since 1988 said they would expect difficulty, compared to 1.7% and 3.6% (respectively) among people without disabilities. 11 This question was borrowed from the 1986 Harris poll commissioned by the National Organization on Disability. 17 Those encountering or expecting difficulty in voting were asked an open-ended question concerning the type of difficulty, and the answers were coded into four main categories and ten subcategories. The most common category of difficulty among people with disabilities who had voted since 1988 includes problems inside the polling place, experienced by 2.9% of all voters with disabilities-particularly difficulty seeing the ballot. A smaller number (1.7% of all voters) reported difficulty with general mobility (standing or walking), while a similar number (1.6%) reported specific difficulty in getting to the polling place (finding the polling place, too great a distance, or not having necessary transportation). About 1% of all voters with disabilities reported difficulty getting into the polling place, due to stairs or steps. Expected difficulties among non-voters were fairly evenly split among problems with general mobility (10.3% of all non-voters), getting to the polling place (8.8%), and difficulties once inside the polling place (9.7%). The largest single reason was difficulty with transportation (7.8%), followed by walking (5.2%) or standing (5.1%), and understanding how to vote (4.5%). These actual or expected problems with voting at a polling place may significantly discourage voting among many people with disabilities. While it is clearly possible to vote by absentee ballot, this may not have the same psychological value as a public display of voting with fellow citizens. Apart from the practical aspects of any difficulties of voting at a polling place, the simple existence of difficulties in voting may be taken by many people as a sign that they are not fully welcome to participate in the rights and responsibilities of citizenship. The practical and psychological barriers stemming from difficulties in voting at a polling place will be further discussed in the final section. c) Other Forms of Participation Are people with disabilities more or less likely to engage in political and civic 18 participation other from voting? The survey asked all respondents about seven kinds of participation in the past 12 months, with results summarized in Table 5.¹² Overall, people with disabilities participated in fewer of these activities than did people without disabilities: about half (51.3%) did any activity and their average number of activities was 1.15, while about three- fifth (61.0%) of people without disabilities did any activity and their average number of activities was 1.47. People with disabilities were particularly less likely to have written or spoken to an elected representative or public official (less than one-third, or 30.3%, have done so compared to almost 40% of people without disabilities), but were almost as likely to have attended a political meeting or written a letter to a newspaper. Each of these activities is related to demographic characteristics, particularly age and education-the next section reports on differences by disability status after accounting for demographic characteristics. Those in the disability sample were also asked two questions about disability-related action in the past twelve months. Just under one-tenth (8.4%) of those with disabilities said that they had been involved in one of the political activities concerning a disability issue, while one- tenth (10.1%) said that they had worked to change a private organization's practices or policies regarding people with disabilities (such as by talking to business owners, or filing lawsuits). Combining these two forms of activity reveals that just under one-sixth (15.3%) had taken at least one action on a disability issue in the past twelve months. 12 The results for participation in the first five activities among people with disabilities are very similar to those found in the 1998 Harris/N.O.D. survey (Louis Harris and Associates, 1998: 92). 19 VI. DEMOGRAPHIC CHARACTERISTICS AND PARTICIPATION Are the lower participation rates described in the previous section due to characteristics other than disability, such as age and education? Are there demographic groups within the disability population that have particularly low participation? This section examines the role of demographic characteristics, first looking at how the overall gaps in participation are affected when controlling for differences in demographic characteristics, and then seeing how demographic characteristics may interact with disability in producing larger or smaller gaps for particular groups. Table 6 compares voter turnout and other forms of participation among demographic groups. The first important finding is that the participation gap widens with age: among the younger age groups there is only a small gap between people with and without disabilities in voter turnout and other forms of participation, but among the older age groups there is a much larger gap. The voting gap is only 4.8% in the 18-29 age group, but 29.6% among those age 65 or older, and almost 43% among those age 80 or older (column 5). This pattern is consistent with past research using the CPS (Shields, Schriner, and Schriner, 1998a). The disability voting gap is also widest among females and those who are widowed- possibly reflecting their greater likelihood of being age 65 or older-while the gap in other forms of participation is widest among men and those who are separated or divorced. Both types of gaps are largest among African-Americans, while the disability voting gap is concentrated among non-Hispanics. The education results produce an intriguing finding: lower voter turnout of people with disabilities only appears among those with 8th grade or less education, while gaps in other forms of participation appear both among this group and among those with a graduate degree. 20 Educational attainment is strongly linked to age (with lower rates of college attendance among those who are older), so the relationship of voting to education and disability is should be analyzed after controlling for age and other demographic characteristics, as done in the next section. a) Prediction of voting with demographic characteristics The simple differences presented in Table 6 could be accounted for by other factors, and are best evaluated when the effects of those other factors are held constant. Tables 7 and 8 present the results of regressions that predict voting and other forms of participation, allowing one to analyze the independent effect of each characteristic while holding constant the remaining characteristics. Table 7 uses probit regressions to predict the likelihood of voting, first for the disability and non-disability samples separately, and then for the full sample with disability variables to estimate the differences associated with disability. The reported figures are the estimated effects of a one-unit change in a variable (e.g., change from male to female, or increase in age by one year) on the probability of voting, which can be multiplied by 100 to obtain the effect in terms of percentage points. 13 Table 8 likewise uses probit regressions on the full sample to predict the likelihood of engaging in any other form of participation, and then uses standard (ordinary least squares) regressions to predict the number of other forms of participation. While people with disabilities were oversampled for this study, all of the regressions with the full sample are weighted to reflect the general population (so that the coefficients should reflect the effects of variables in a fully representative sample).¹⁴ 13 The figures reported in Tables 7 and 8 are derivatives (dF/dX) based upon probit coefficients, assuming independent variables are at their mean values (or zero, for effect of a dummy variable). 14 On top of the weighting by sex, region, and household size done separately for the disability and non- disability samples (described in the methodology section), the cases were weighted so that the disability sample represents 12.4% of the adult citizen population. The regression procedure calculates standard errors based on the unweighted number of observations, reflecting the probable distribution of values were the sample to be repeated. 21 Education and age are the strongest influences upon voting in both the disability and non- disability samples. Among people with disabilities, those with a high school degree are 32 percentage points more likely to vote than those who never attended high school, while those with a graduate degree are 43.5 percentage points more likely to do so. A similar pattern exists in the non-disability sample, with lower coefficients reflecting less of a gap between those not attending high school and all other groups. The strong effect of education is in line with past research showing the importance of education as a resource that enables and encourages political participation. Age has a strong positive effect on voting in both samples, but the effect is much larger in the non-disability sample where each year of age is associated with a 1.5 point increase in the likelihood of voting, compared to only a 0.6 point increase among those with disabilities. The positive effect of age is also in line with past research on the general population, where it may represent either increased likelihood of voting as one gets older (possibly due to resource effects) or differences between generations. In addition, these results reflect the simple comparisons in Table 6 showing that the voting gap between people with and without disabilities grows with age. The overall voting gap between otherwise-similar people with and without disabilities is 19.7 points, as shown in column 3 where the samples are combined and demographic characteristics have been held constant. The main reason this is bigger than the simple 6.1 point gap is that people with disabilities are disproportionately older, and older people are generally more likely to vote; the simple comparison is therefore muddied by the strong association between age and disability, while the regression adjusts for this and finds a much larger gap when comparing people of the same age. While the overall average voting gap is close to 20 22 points, the regression reported in column 4 interacts age and disability and confirms that the age effect is significantly lower among people with disabilities, so that the gap is close to zero among people in their 20's but grows rapidly with age. This is illustrated in Figure 1, which adjusts for overreporting so that average turnout reflects the 38.9% figure based on ballots cast. How many millions of Americans are represented by this voting gap? Combining the 19.7-point voting gap with this survey's disability prevalence indicates that if people with disabilities voted at the same rate as those without disabilities, there would have been an additional 4.6 million voters in November, 1998, which would have raised overall voter turnout by 2.5 percentage points.¹⁵ Voter turnout is lower among people with disabilities across all educational categories, as shown in column 5 which interacts disability with each education category. While this indicates that the simple comparisons in Table 6 were misleading in showing lower turnout only among the less-educated, it remains true that the voting gap is largest among those who have not attended high school (42.2 points), but is also substantial among those who have some college education (19 points), an AA degree (26 points), or a BA degree (19 points). The estimated gap in voter registration rates is also increased when accounting for demographic characteristics. In estimates using the variables in column 3 of Table 8 to predict voter registration, people with disabilities are estimated to be 10.2 percentage points less likely to have been registered to vote in 1998. 16 As with the voting gap, the registration gap increases strongly with age. Table 8 uses the variables in the last three regressions of Table 7 to predict any other 15 Of the estimated 186.9 million citizens in November, 1998 (http://tap.epn.org/ csae/cgans4.html), about 23.2 million had disabilities based on this survey's questions, so that the 19.7-point gap represents 4.6 million citizens with disabilities, which in turn represents 2.5% of all eligible citizens. 16 These estimates are not reported in the tables but are available. 23 form of participation, and the number of other forms of participation. In line with the voting results of Table 7, education and age have the strongest effects. Those who have high school degrees are about 30 points more likely, and those who have graduate degrees are about 47 points more likely, than those without high school education to engage in at least one of these other forms of participation (column 1). Similarly, those with graduate degrees engage in an average of 2.3 more forms of participation than those without high school education (column 4). People with disabilities are 8.2 points less likely than those without disabilities to have engaged in any form of participation (column 1), and have engaged in an .17 fewer forms of participation on average (column 4). As with voting, however, the disability participation gap widens with age: each year of age is associated with a 0.6 point increase in the likelihood of any participation in the non-disability population, but no increase in the disability population (column 2). The participation gap does not, however, show a clear pattern with respect to education (columns 3 and 6), indicating that while education raises the likelihood and number of these forms of participation, people with disabilities are less likely to engage in them at each education level. Which particular activities are people with disabilities less likely to engage in? While the simple differences indicated that people with disabilities were less likely to engage in each of the activities (as reported in the last section and Table 5), some of these differences can be attributed to demographic characteristics apart from disability. When estimates are made separately for each of the seven activities, the overall participation gap is statistically indistinguishable from zero for three of the activities after accounting for demographic characteristics: having attended a political meeting, written a letter to a newspaper, and otherwise worked with groups or on one's own to change government laws or policies. People with disabilities are 5.5 points more 24 likely to have worked with others on a community problem, but 5.1 points less likely to have contributed money to a political party or candidate, 9.3 points less likely to have written or spoken to an elected representative or public official, and 4.8 points less likely to have contributed money to organization trying to influence government policy or legislation. As with voting, for these last three activities the gap between those with and without disabilities grows with age, so that it is greater among older people with disabilities. V. DISABILITY CHARACTERISTICS AND PARTICIPATION Are the voting and participation gaps greater among those with certain impairments or activity limitations? Disability characteristics can provide insights into the types of problems that decrease voter turnout and other types of participation among people with disabilities. Table 9 presents simple comparisons of voting, registration, voting difficulties, and other forms of participation by disability characteristics. As can be seen in column 2, voter turnout is especially low among those who use wheelchairs, have difficulty learning, and have difficulty with activities inside or outside the home. Those who have difficulty going outside the home alone are also most likely to report actual difficulties encountered in voting (23.9%), followed by those who need help with any daily activities (19.2%), who are blind (17.6%) and who have a learning disability (17.1%). (There is overlap among these categories; in particular, many of those having difficulty outside the home also need help with daily activities.) Finally, the number of other forms of participation is lowest among those who are unable to walk ¼ mile, who have difficulty going outside, and who have difficulty learning, remembering, or concentrating. The independent effects of these characteristics are explored in regressions in Table 10, 25 which include the non-disability sample so that the figures represent estimated differences from someone in the general population without that characteristic. Controls for demographic characteristics are included in all regressions. When only functional limitations and demographic characteristics are used to predict voting, voter turnout is particularly depressed among those with severe mobility restrictions (-34.5 percentage points for wheelchair users, - 19.8 points for those who use no mobility aid but cannot walk ¼ mile, and -17.2 points for those using a cane, crutches, or walker), a severe vision impairment (-18.2 points), or difficulty remembering or concentrating (-15.9 points). These effects are decreased but not driven to zero when controlling for activity limitations-particularly difficulty in going outside home alone which decreases turnout by about 30 points (column 2). The continuing negative effects from wheelchair use and the inability to walk ¼ mile-on top of the negative effects from difficulty in going outside one's home alone-indicate that particularly severe mobility problems discourage voting. Difficulty in going outside home alone is also a strong negative predictor of the likelihood and number of other forms of participation (columns 4 and 6). As with voting, the inability to walk ¼ mile, and having difficulty remembering or concentrating, also exert a strong negative influence on participation. In estimates not reported here, these same patterns appear when looking just at the three forms of participation in which people with disabilities lag (contributing money to a campaign or candidate, contributing to an organization trying to influence government policy or legislation, and writing or speaking to an elected representative or public official). Is there a difference in participation according to how long one has lived with a disability? Recently-occurring disabilities may be more likely to depress participation, as people 26 spend time and energy trying to adjust to the new condition. The influence of time since disability was assessed in two ways. First, a variable representing the number of years since a disability occurred was entered as a predictor of participation (along with other demographic variables) in regressions limited to the disability sample. This showed that more recent disabilities have a more negative effect on voter turnout, but not on other forms of participation: each year of living with a disability adds 0.35 points to the likelihood of voting (on top of the general 0.55 point age effect for people with disabilities). Second, categories representing the age at which one's disability began were entered (along with other demographic variables) into full- sample regressions predicting participation separately for those age 65 or older, and those age 45 to 64. The effects of the age at onset are as follows (with asterisks next to the estimates for which a zero effect can be confidently rejected): Voting (percentage point effect) Number of other forms Current age Current age Age when 65+ 45-64 18-44 65+ 45-64 18-44 disability began 0-17 -35* -20 13 -.88* .02 .04 17-44 -29* -10 -6 -.12 .30 .06 45-64 -35* -24* -.53* .11 65+ -50* -.50* These results support the idea that more recent onset of disability discourages voter turnout, given that the turnout is most depressed (50 points lower) among those who had the disabling condition begin when they were 65 or older, while the effect is about two-thirds that size (35 points lower) for those age 65 or older who had an onset of disability before age 18. Those with recent disabilities were also least likely to vote in the 45-64 age group (24 points lower) and 18- 44 age group (6 points lower). There is no clear relation, however, between other forms of 27 political participation and the age at which the disability began. In sum, the analysis of disability characteristics points to an especially strong effect of independent mobility-being able to go outside home alone. This is consistent with the finding that many people with disabilities encounter or expect mobility problems in voting at a polling place (Table 3). Given that people are able to vote by absentee ballot, the ability to go outside home alone is clearly not required for voter turnout, or for many other forms of participation. This result therefore suggests that increased mobility is one of the factors that can raise the likelihood of voting. Mobility is not, however, the only factor lowering participation among people with disabilities. Voter turnout and other participation are also low among people who have difficulty remembering or concentrating. Some people with such difficulty can have problems understanding how to vote, as reflected in Table 3's finding that 4.5% of those with disabilities who haven't voted since 1988 would expect such problems (although 3.2% of the non-disability sample also report that they expect such problems). People with severe vision impairments are less likely to vote, although not to engage in other forms of participation, suggesting that the vision impairments may add to mobility difficulties and/or require aid in reading the ballot which makes the process of voting more troublesome (also reflected in voting difficulties listed in Table 3). Finally, those with recent disabilities appear less likely to vote (although not to engage in other participation) than those with longer-term disabilities, indicating that there may be general problems in adjusting to life with a disability that depress voter turnout (consistent with Stoker and Jennings, 1995, who claim that major life transitions depress turnout). 28 VIII. MOBILIZATION Recruitment by others is one of the important means by which people become politically active. In an election year the political parties spend a large amount of resources on mobilization efforts-trying to contact potential voters and convince them to vote for the party's candidates. Other organizations also become involved in such efforts on behalf on candidates or issues. Apart from efforts for specific candidates, a variety of organizations and individuals often talk to potential voters about the importance of registering and voting, in both formal efforts and informal conversations. To measure the importance of such mobilization efforts, survey respondents were asked three ANES questions concerning whether they had been contacted by parties or others about supporting specific candidates in 1998, and whether someone had talked to them about registering or getting out to vote. For those who reported that they were contacted, follow-up questions were asked to determine who had contacted them. The results are summarized in Table 11. People with disabilities were less likely than people without disabilities to be contacted during the 1998 election campaigns by either major political party: slightly less than one-third (31.0%) were contacted, which is 6.4 points lower than for people without disabilities. This gap widens to 9.0 points when holding constant demographic characteristics; unlike the voting gap, it does not change much by age. Party contact appears to make a difference in voter turnout, given that turnout in both the disability and non-disability samples is about 20 percentage points higher among those contacted than among those not contacted. There continues to be a voting gap between people with and without disabilities, however, whether people were contacted or not, and controlling for party contact in voting regressions makes only a small difference in the 29 estimated gap, indicating that little of the disability voting gap is due to different levels of party mobilization for people with and without disabilities. About one-tenth of both samples were contacted by someone apart from the major parties about supporting candidates, and someone spoke to one-fourth of respondents in both samples about registering or getting out to vote. There were not significant differences between the disability and non-disability samples in these measures (either before or after controlling for demographic characteristics), and the disability voting gaps persist among the contacted and non- contacted. Apart from political parties, other organizations and friends were the most likely to speak to respondents about supporting specific candidates (as shown in Table 11). Similarly, parties, other organizations, and friends were also the most likely to talk to respondents about registering or getting out to vote. The only significant differences by disability status are that people with disabilities were less likely to be talked to by family members or co-workers-reflecting their greater likelihood of living alone and lower likelihood of being employed. In sum, while people with disabilities were less likely to be contacted by political parties in the 1998 campaigns, they were as likely to be contacted by someone else about supporting candidates and registering or getting out to vote, and mobilization appears to play little role in explaining why people with disabilities were less likely to vote. IX. POLITICAL INTEREST, EFFICACY, COMPETENCIES, AND PARTICIPATION Having an interest in politics, and a belief that one is able to understand and make a difference by participating, are clearly basic elements in the decision to participate. This survey measured such factors in a variety of ways. 30 First, political interest was measured with an ANES question concerning the extent to which respondents follow what's going on in government and public affairs. Next, political efficacy was measured using five items from the ANES. Political scientists distinguish between two types of political efficacy: "internal" efficacy is the belief that one is well-qualified to participate in politics, while "external" efficacy is the belief that the political system is responsive to individuals like oneself. This survey uses two external efficacy and three internal efficacy measures.¹⁷ Schneider and Ingram (1993a, 1993b, 1995) have theorized that participation in politics is strongly affected by how the group of which one is a member is perceived by politicians and public officials. The likelihood of participation may be enhanced when one is a member of a group that is perceived by politicians to be powerful, and that is treated with respect, such as senior citizens. Participation may be depressed, however, when one is a member of a group that is not seen to have political power, and whose members are not treated with respect and dignity. This survey sought to measure the perceived status of people with disabilities by asking all respondents how much they agree or disagree with the statements "People with disabilities have as much influence as other groups in government and public affairs," and "Government officials treat people with disabilities with as much respect and dignity as they treat members of other groups." Finally, a set of questions was used to measure perceived competency in several skills that are important in political and civic participation. This survey supplemented the internal efficacy questions with five measures of specific competencies, including the abilities to: work with others, 17 The ANES has used three of these measures (the first three in Table 12) in every survey since 1948. The other two measures were used experimentally in 1992, and are included here because they were strong predictors of voting in that year. 31 speak in public, lead a group, compose an effective letter to an elected official, and communicate ideas to others.¹⁸ These are likely to be especially important for forms of participation other than voting, such as communicating with elected representatives and working with others on community problems. Table 12 contains the results from the questions on political efficacy and competencies. As can be seen, over half the respondents said they follow what's going on in government and public affairs "most of the time," which does not differ by disability status. People with disabilities, however, are more likely to say that they follow government and public affairs "now and then" or "hardly at all" (19.0%) compared to people without disabilities (13.9%). They also have lower average levels of political efficacy according to the first three efficacy measures, being more likely than people without disabilities to agree that "public officials don't care much about people like me," "people like me don't have any say about what the government does," and "sometimes politics and government seem so complicated that a person like me can't really understand what's going on." While people with disabilities continue to report lower average efficacy on these measures after controlling for demographic characteristics, they are just as likely as people without disabilities to agree that they are "well-qualified to participate in politics" and "better-informed about politics and government than most people."¹⁹ People without disabilities have more positive views of the influence and treatment of people with disabilities in government and public affairs. Just over two-thirds (69.0%) agree that people with disabilities have as much influence as other groups, and three-fifths (59.2%) agree that people with disabilities are treated with as much respect as other groups, while only 62.2% and 18 These were adapted from a previous survey done by the Rutgers Center for Public Interest Polling. 19 Each regression (not reported but available) uses the 5-point scale for the efficacy measure (questions D2a- e in Appendix A) as the dependent variable, and controls for the demographic variables used in Table 7. 32 49.3% (respectively) of people with disabilities agree with these statements. People with disabilities also rate themselves less highly on each of the five specific competency measures than do people without disabilities. While two-thirds (67.7%) of people without disabilities say they are better than most people at working with others, only half (49.8%) of people with disabilities say so. Similarly, while half (52.4%) of people without disabilities say they are better than most people in leading a group, only one-third (35.9%) of people with disabilities say so. These competencies may be strongly related to demographic characteristics, particularly to education that can increase speaking and writing skills. Estimates that take demographic characteristics into account show that education is in fact a strong predictor of these skills; once these characteristics are held constant, people with disabilities report lower average scores only in ability to work with others and lead a group (possibly reflecting greater isolation).²⁰ How are these perceptions related to voting and other forms of participation, and can they account for the participation gaps between people with and without disabilities? As shown in column 6 of Table 12, people with disabilities continue to have lower voting rates no matter how they answered most of these efficacy and competency questions. The voting gap is highest among those who say they follow what's going on in government and public affairs most of the time, and close to zero among those who do so only now and then or hardly at all, strongly suggesting that the lower turnout is not due to less interest in politics. When the political efficacy measures are used to predict voting, the estimated disability voting gap is lowered only slightly, suggesting that lower turnout is not primarily due to lower feelings of political efficacy. Similarly, the disability voting gap is changed little when accounting for perceptions of the influence and treatment of 20 Each regression (not reported but available) uses the 4-point scale for the competency measure (questions D3a-e in Appendix A) as the dependent variable, and controls for the demographic variables used in Table 7. 33 people with disabilities, and the five competencies. It is noteworthy, however, that the disability voting gap is small among those who say they are better than most people in composing an effective letter to an elected official, and in communicating ideas to others. These skills appear to help people with disabilities participate as much as those without disabilities, while there is a large disability voting gap among those who report having fewer of these skills. Lower levels in other forms of participation by people with disabilities also do not appear to be explained by these efficacy and competency measures. As shown in column 9 of Table 12, the number of other forms of participation in the last year is generally lower no matter how people responded to these questions. As with the voting results, the gap is highest among those who follow government and public affairs most of the time, again suggesting that lower participation is due to barriers other than lack of interest. X. EMPLOYMENT AND PARTICIPATION Employment can affect political and civic participation in a number of ways. One way is that it provides income that can be an important resource for participation; research on the general population typically finds that higher-income people are more likely to vote and engage in other forms of participation (Rosenstone and Hansen, 1993; Verba et al., 1995). Employment can also provide a wider range of day-to-day contacts, increasing the chance that one will be explicitly recruited, or subject to implicit social pressure, to vote or engage in other forms of participation. Finally, both income and a wider range of contacts can have psychological effects that encourage participation, through increasing one's sense of citizenship and stake in mainstream society. The survey asked several employment questions, using standard wording from the Current Population Survey that serves as the main source of employment information for the general population. 34 Overall, just under one-third (31.9%) of people with disabilities and over two-thirds (70.7%) of people without disabilities were employed in the week prior to the survey. Since the low employment rate of people with disabilities partly reflects their greater likelihood of being over age 65 and retired, Table 13 breaks out employment rates separately for those of working age (18- 64) and those age 65 or older. As can be seen, just under half (46.7%) of the working-age people with disabilities were employed, compared to four-fifths (79.5%) of working-age people without disabilities. These employment rates are very close to those found for working-age people with and without disabilities in the Survey of Income and Program Participation.² Of those who were employed, people with disabilities were more likely to work part-time, and slightly more likely to be union members. Employment appears to make a substantial difference in the likelihood of voting. As shown in column 5 of Table 13, among people age 18 to 64 the voting rate is almost identical between employed people with and without disabilities, while non-employed people with disabilities are about 13 points less likely to vote than non-employed people without disabilities. Similarly, the disability voting gap among those age 65 and older is concentrated among those who are not employed (32.5 points, compared to a gap of 11.6 points among the employed). Among those who are employed, the disability voting gap does not vary substantially between the full-time and part- time workers, or between union and non-union workers. These patterns are maintained when estimates are made holding demographic characteristics constant: voter turnout is similar among employed people with and without disabilities, but is 29.3 percentage points lower overall among 21 From the SIPP survey, an estimated 52.4% of working-age people with disabilities had a job in the previous month, compared to 82.1% of people without disabilities. As noted earlier, the SIPP asked more disability questions and had a higher incidence of disability; in addition, it measured employment as any job within the previous month rather than the week-based CPS measure used here (http://www.census.gov/hhes/www/ disable/sipp/disab9495/asc9495.html) 35 non-employed people with disabilities compared to non-employed people without disabilities. The results are similar for other forms of participation: employed people with disabilities engage in a slightly higher average number of political activities than do employed people without disabilities. In contrast, non-employed people with disabilities engage in fewer types of participation than do non-employed people without disabilities, particularly if they are age 65 or older. These results raise the twin questions of why employment appears to have such a strong positive effect on participation for people with disabilities, and why there is such a participation gap among people who are not employed. Is it income, mobility, social contact, something else, or a combination of factors that is responsible for these patterns? The following sections on transportation, group involvement, and income can shed light on these possible explanations. XI. TRANSPORTATION AND PARTICIPATION Convenient and reliable transportation is important for many kinds of political and civic participation. It is clearly necessary when such participation requires attending meetings outside of one's neighborhood. Many forms of participation can take place within one's home, such as voting by absentee ballot, writing or calling elected representatives, and making financial contributions to campaigns or organizations. Transportation may nonetheless play an indirect role in encouraging these activities-voting at a polling place and having face-to-face contact with elected representatives and colleagues may enhance the value of political action. In addition, convenient and reliable transportation increases general mobility, which will affect recruitment opportunities and a sense of connection with mainstream society. To explore the importance of transportation, the survey asked several questions to 36 determine respondents' general mobility, usual transportation modes, ability to drive, and frequency of transportation difficulties. People with disabilities make fewer trips for work, shopping, or any other reason in a typical week. As shown in Table 14 (columns 1-2), the average numbers of days they take such trips is 3.34, compared to 5.18 for people without disabilities; in addition, only two-fifths (40.9%) of people with disabilities take such trips on five or more days a week, compared to three-fourths (75.9%) of people without disabilities. These differences partly reflect employment patterns: among employed people the percentage taking trips on five or more days is very similar between people with and without disabilities (about 85%), while among non-employed people the percentage is much lower among people with disabilities (about 20% compared to 50% among people without disabilities, both for those age 18-64 and for those age 65 or older). While the large majority of people with disabilities use their own or a family vehicle for transportation needs, they are less likely to do so than are people without disabilities. About one- fifth use some other means of transportation, such as someone else's vehicle or public transportation. They are also less likely to have a vehicle they can drive: just over two-thirds (68.1%) can drive their own vehicle, compared to almost all (93.0%) of people without disabilities. Non-employed people with disabilities are much less likely than those who are employed to be able to drive their own vehicles, while this figure is about equal between employed and non- employed people without disabilities. Somewhat surprisingly, people with disabilities are no more likely than those without disabilities to report transportation difficulties: close to two-thirds of both samples say that they never or rarely face delays or other problems in transportation. Transportation is clearly related to levels of political and civic participation. As shown in column 5 of Table 14, the disability voting gap is concentrated among those who travel less than 37 five days in the average week, and close to zero among those who travel five days or more. When broken out by employment status, the disability voting gap is close to zero among those who are employed (no matter how many days they travel in an average week), but continues to exist among the non-employed whether they travel fewer than five, or five or more, days in an average week. It is noteworthy, however, that the gap is about 10 points smaller among those who travel five or more days per week, indicating that lower turnout is concentrated among non-employed people with disabilities who travel fewer than five days per week. Further evidence that transportation is an important factor for people with disabilities is provided by the finding that the disability voting gap is narrowest among those who use their own or a family vehicle as the usual mode of transportation. In addition, the voting gap is close to zero among those who do not have a vehicle they can drive-the low voting rate among this group indicates that this condition appears to discourage voter turnout in general, and contributes to the lower overall turnout of people with disabilities given that they are much more likely than people without disabilities to be in this group. Simply having a vehicle one can drive does not close the voting gap, as indicated by the finding that voter turnout remains lower among non-employed people with disabilities who can drive their own vehicle (9.0 percentage points lower in the 18-64 age category, and 19.6 points lower in the age 65 or older category). The disability voting gap is smallest among those who say they never or rarely encounter transportation problems; after accounting for demographic characteristics, however, the turnout of people with disabilities is below that of people without disabilities across all levels of transportation problems.²² Other forms of political and civic participation are, not surprisingly, lowest among those 22 Predictions of voting were done with demographic characteristics and dummy variables for each travel difficulty category, plus the interactions of these dummies with disability. The estimated disability voting gap was largest among those who "occasionally" have problems (30 points), but also large and significant among those who say they never or rarely face problems (18 points). 38 who take trips in no more than two days in an average week (column 8 of Table 14). The gap is especially large for non-employed people with disabilities who are age 65 or older and take trips for no more than four days in an average week. The gap remains large even among those age 65 or older who have a vehicle they can drive, suggesting that lack of independent transportation is not the key barrier limiting participation in this group. In sum, transportation is clearly affects political and civic participation. Voter turnout and other forms of participation are higher among those who are more mobile (taking more trips during the week) and who have a vehicle they can drive. People with disabilities are less likely to take as many trips or have vehicles they can drive, which partly accounts for their lower levels of participation. This is not a full explanation, however, given that non-employed people with disabilities-particularly those age 65 or over-are less likely to participate even if they have a vehicle they can drive and take trips as often as people without disabilities. XII. GROUP INVOLVEMENT, RELIGIOUS ATTENDANCE, AND PARTICIPATION Political and civic participation often takes place through involvement in groups and organizations explicitly focused on political or civic projects. Even groups and organizations that are not explicitly focused on political or civic projects can provide social contacts and exposure to issues that increase members' political and civic participation. Involvement with religious institutions in particular has been linked to higher voter turnout in the general population (e.g., Rosenstone and Hansen, 1993). To assess the role that group involvement plays in political and civic participation, the survey asked several questions about the groups that respondents meet with regularly, how often 39 such meetings take place, and whether and how often respondents attend religious services.²³ People with disabilities are less likely to meet regularly with any groups or organizations: only two-fifths (41.3%) do so compared to over half (54.6%) of people without disabilities, as shown in Table 15. The percentage who attend such a meeting an average of once a month is similar between people with and without disabilities, but people with disabilities are less likely to average two or more meetings a month with groups or organizations. In addition, only three-fifths of people with disabilities (59.0%) ever attend religious services, compared to over two-thirds (70.0%) of people without disabilities. It is noteworthy, however, that among those who do not attend religious services, people with disabilities are much more likely to say that they think of themselves as part of a church or denomination, suggesting that mobility problems or other barriers keep them from attending. In fact, the samples are similar in the portion of respondents who say that they never attend services and do not think of themselves as part of a church or denomination (about one-sixth of each sample). How are group involvement and religious attendance related to political and civic participation? Those who regularly meet with any groups or organizations are much more likely to vote and engage in other forms of participation, so that part of the lower overall participation of people with disabilities is linked to their lower levels group involvement. The disability voting gap is smaller among those who regularly meet with any groups or organizations, and who attend religious services every week or almost every week, indicating that such activity may help to close the voting gap.24 The gap continues to exist, however, whether people do or do not attend such 23 The questions on religion were taken from the ANES, while the questions on group involvement were created for this survey but based on the questions asked in the study by Verba et al. (1995). 24 This is based on probit regressions predicting voting with demographic variables, group and religion variables, and group and religion variables interacted with disability. 40 meetings, or do or do not attend religious services. Similarly, people with disabilities who are active in groups and/or attend religious services have average participation levels that are closer to (but still lower than) those of people without disabilities. In sum, those who are involved in groups or organizations, and those who attend religious services, are more likely to vote and engage in other forms of political and civic participation. While people with disabilities are less likely to be involved in groups or attend religious services, this difference accounts for little of the political and civic participation gaps between people with and without disabilities. XIII. INCOME AND PARTICIPATION Income can have a direct effect on political and civic participation, most obviously through the ability to make financial contributions to campaigns, causes, or projects. While the process of earning an income may absorb much of one's time and leave little time for political and civic work, high income levels can provide the resources and free time for household members to become involved in such work. In addition, high income may be associated with both an increased economic stake in governmental policies and psychological stake in mainstream society, leading to a greater likelihood of voting, contact with representatives, and other forms of participation. Past research has consistently found a positive link between income and political participation. To assess the role of income, the survey asked respondents to indicate the category in which their total household income falls, using categories from the monthly CPS. People with disabilities are more likely to live in low-income households. As shown in Table 12, over two-fifths (43.7%) reporting household incomes less than $25,000 per year, which is about three times the rate for people without disabilities (14.8%). Correspondingly, only one-sixth 41 (16.4%) of people with disabilities report household incomes of $50,000 per year or more, compared to over two-fifths (42.6%) of people without disabilities. 25 The patterns are very consistent with data from SIPP showing that the mean household income of people with disabilities is only three-fourths that of people without disabilities, and their poverty rate is twice as high (Kruse, 1997). Consistent with past research, there is greater voter turnout among high-income households. People with disabilities who live in households with more than $25,000 in income appear to vote at the same rate as people without disabilities, but there is a disability voting gap among the low- income households and those not revealing household income. The same pattern holds for number of other forms of participation. Earnings of the respondent obviously increase total household income, and it may be that the disappearance of the disability voting gap among high-income households reflects the disappearance of the voting gap among employed people found earlier. Table 16 therefore breaks out income levels and participation rates for the employed, and for those who are not employed above and below the age of 65. When this is done, there is little difference in voter turnout by income status between employed people with and without disabilities. Among non-employed people under age 65 there is lower voter turnout among people with disabilities in low-income households, while among non-employed people age 65 or older there is lower voter turnout across each of the income categories. These same patterns generally hold for other forms of participation as well, with especially lower levels of participation among those who are in low-income households or who did not reveal household income. 25 Each of these percentages is slightly higher in actuality, since they do not include those who said they don't know or refused to give household income data. 42 When used in combination with demographic variables to predict voting, income is a strong positive predictor for people with disabilities, for whom an additional $10,000 in household income raises predicted voter turnout by 1.6 percentage points.²⁶ This effect is stronger than for people without disabilities, for whom an additional $10,000 in income raises predicted voter turnout by only 0.5 percentage points (and a true effect of zero cannot be confidently ruled out). The strong effect of income for people with disabilities reflects the particularly low voter turnout of people with disabilities who have low household income levels, while high-income people with disabilities voted at close to the same rate as high-income people without disabilities. Income is also an important factor in the three other forms of participation in which people with disabilities are less likely to engage. It is a factor not only in the decision to contribute to a candidate or to an organization trying to influence government policy or legislation, but also in whether one contacted an elected representative in the past twelve months. As with voting, controlling for household income levels makes only a small difference in the overall participation gaps for these activities, but the gaps are found to be concentrated among those who have low incomes or did not report their incomes. In sum, these results accord with past studies by finding a positive influence of income on political participation. The lower average income levels of people with disabilities, however, can account for only small portions of the overall gaps between people with and without disabilities in several areas of participation, so that income differences are not the main factor in explaining these gaps. The participation gaps are found to be widest among those who have low household incomes (less than $25,000/year) or do not report household income, indicating that these people may face 26 To create a continuous income variable, the average dollar values for those in each income category were calculated using data from the March 1998 CPS. Those who declined to provide any income data were identified by a separate dummy variable. 43 special problems or barriers to participation. XIV. ANALYZING THE PARTICIPATION GAPS Are people with disabilities more or less likely to engage in voting and other forms of participation once all of the factors discussed in previous sections are taken into account? A variety of these characteristics have been shown to be associated with voting and other forms of participation, as well as with disability, and they may combine to explain the basic differences in participation linked to disability. For example, lower income levels, lower average political efficacy, lower likelihood of mobilization, and limited mobility of people with disabilities may work together to account for their lower likelihood of voting. The combined influence of these factors is explored in regressions predicting voting in 1998, for which results are presented in Table 17. Column 1 reproduces the result from Table 7 that the probability of voting among people with disabilities is 19.7 points lower among people with disabilities after accounting for demographic characteristics. The gap drops, but only to a 14.6 point difference, in column 2 after additionally accounting for the general effects of political efficacy, mobilization, transportation, employment, social activities, and income levels.²⁷ Among these measures, three are clearly important in predicting voter turnout: one external political efficacy measure (disagreement with the statement that "public officials don't care much about people like me"), one mobilization measure (having been contacted by someone in one of the major political parties in the 1998 campaign), and one transportation measure (having a vehicle one can drive). 27 Income is measured as broad categorical variables; continuous measures of household income were also tried with little difference in the results. 44 Only about one-fourth of the disability voting gap is therefore explained by the lower scores of people with disabilities on most of these characteristics. To further explore the voting gap, columns 3 and 4 present the results when disability is broken into six categories. Column 3 shows that after accounting for demographic characteristics, turnout is most depressed among people who have difficulty going outside alone (-32.2 percentage points), and next most depressed among those who cannot walk ¼ mile and do not use mobility devices (-15.8 points) and those with a severe vision impairment (-15.0 points). One might expect that the lower turnout of those with difficulty going outside would be explained by their fewer days of traveling per week, or their lower likelihood of having a vehicle they can drive. When the full set of explanatory variables is used in column 4, however, the gap for those with such difficulty closes only to 28.6 points. The gaps associated with the other five disability categories also close slightly, but not sufficiently to indicate that differential levels of efficacy, mobilization, transportation, employment, social activities, and income can explain the lower turnout for any of the categories. The strong negative effect of difficulty in going outside alone is explored by interacting it with several measures in column 5. These results show that the relationship of such difficulty to voting is linked to participation in several other activities. Those with difficulty going outside alone have lower turnout if they: do not meet regularly with any groups or organizations, or do not have a vehicle they can drive, or take trips less than five days a week. Turnout is even lower if they meet more than one of these conditions, and lowest if they meet all three.²⁸ Those who meet these conditions probably have particular difficulties with mobility. The importance of mobility is reinforced by the negative effects of being unable to walk one-fourth mile, or having difficulty in doing so. Those with mobility problems often have problems in voting at polling places, as 28 This is confirmed by the use of additional interactions. 45 confirmed by the higher level of actual or expected difficulties among those who have difficulty going outside alone: over one-third (35.5%) have recently encountered or would expect difficulties in voting at a polling place, which is more than five times the rate for people with disabilities who have no trouble in going outside alone (6.4%). Mobility problems therefore appear to play a strong role in the lower turnout of people with disabilities. While everyone is able to vote by absentee ballot, lower mobility somehow discourages voting in any format. This may reflect a lower value put on absentee voting, which is an isolated activity that does not have the symbolic and demonstrative value of voting at a polling place with fellow citizens. Turnout also appears to be lower among people with severe vision impairments and mental impairments (difficulty remembering or concentrating), even after controlling for the full set of characteristics. As with those who have difficulty going outside alone, people with these impairments are more likely to have encountered or to expect difficulties in voting at a polling place: over one-fourth of people with severe vision impairments (27.6%), and over one-fifth of people with mental impairments (22.0%), have encountered or expect difficulties in voting at a polling place, compared to less than one-tenth (8.5%) of those who have neither type of impairment. Therefore the low voting rates among those with mental or severe vision impairments may partly reflect difficulties with voting at a polling place combined with a lower value put on absentee voting. What role do the variety of characteristics considered here play in explaining the lower likelihood that people with disabilities engage in other forms of political and civic participation? As noted in Section VI, after accounting for demographic characteristics, people with disabilities were 8.2 percentage points less likely to have engaged in any of seven other forms of participation, and engaged in an average .173 fewer activities over the past twelve months. When looked at 46 separately, however, lower participation occurred in only three of the activities: people with disabilities were 5.3 points less likely to have contributed to a campaign or candidate, 4.8 points less likely to have contributed to an organization trying to influence government laws or policies, and 9.4 points less likely to have written or spoken to an elected representative or public official. Unlike the voting gap, the gaps in other forms of participation almost fully disappear after accounting for the political efficacy, transportation, employment, social activity, and income measures used in Table 17. Further analysis shows that the gaps are primarily due to lower levels of income, less travel throughout the week, lower likelihood of having a vehicle one can drive, and lower perceptions of political efficacy. These measures are highly correlated with each other- people with lower incomes tend to have less travel, lower likelihood of driving, and lower perceptions of political efficacy-making it difficult to cleanly sort out which specific measures account for the disability gaps. XV. IMPLICATIONS The above results show that voter turnout is lower on average among people with disabilities-particularly those who are older and non-employed-than among people without disabilities who have the same demographic characteristics. People with disabilities are as likely as those without disabilities to participate in several other forms of political and civic activities, but are less likely to engage in three activities (contributing to a campaign, contributing to a political organization, and contacting an elected representative). The analysis revealed that lower participation in these latter three activities is linked to lower incomes, decreased mobility, and weaker feelings of political efficacy among some people with disabilities. The lower voter turnout of people with disabilities, however, is only partly 47 explained by these and other standard predictors of voter turnout. After controlling for many factors, voter turnout is found to be still much lower than expected among those who have difficulty going outside alone, and is also lower than expected among those with severe vision impairments, mental impairments, and difficulty walking one-fourth of a mile. It is particularly low among those who, along with difficulty going outside alone, have little mobility or social interaction (no regular meetings with groups or organizations, no vehicle one can drive, and trips on fewer than five days in an average week). The groups for which turnout is especially low have high levels of actual or expected problems in voting at a polling place. These results indicate that actual or expected difficulties in voting at a polling place may be a culprit in the low turnout of many people with disabilities. While absentee ballots are easily available, those who feel forced to use them because of inaccessible polling places may not put as much value on voting as do those who can go to a polling place without difficulty. People may value the demonstrative aspects of publicly congregating with fellow citizens at a polling place, which can reinforce the social nature and value of democratic participation. Also, most elections have a declared winner before the absentee ballots are counted, so that those who feel forced to vote by absentee ballot may be especially likely to feel that their votes do not matter. Difficulties in voting at a polling place may, however, be only one factor that represents and contributes to a more basic problem. To explain why people engage in community-oriented activities in which there is little individual self-interest to do so (like voting, given that the chance of one vote deciding an election is almost always negligible), social scientists have recently developed the concept of "social capital" which is the "web of cooperative relationships between citizens that facilitates resolution of collective action problems" (Brehm and Rahn, 1997: 999). Social capital is likely to be lower among those who have difficulty going outside and interacting 48 with the larger community, because they have fewer opportunities to build and maintain relationships that engender a sense of identification and cooperation with fellow citizens. The depreciation of social capital can lead to a self-perpetuating spiral in which individuals become more and more detached from the community and less likely to engage in political and civic activities. This kind of isolation may result not just from physical difficulties in going outside, but from stigma and discrimination that leads one to avoid social contact. In this interpretation, low levels of social capital may be an underlying reason for low levels of turnout and participation in several political activities among some segments of the disability population. An overarching goal of the Americans with Disabilities Act is to integrate people with disabilities more fully into mainstream life, by combatting stigma and discrimination and increasing the accessibility of many areas of public life. As such, the ADA may play an important role in developing the social capital of people with disabilities. Just as the ADA can have economic benefits by promoting the full use of our country's human resources, it can have important social and civic benefits by developing social capital that leads people with disabilities to share their skills and experiences in addressing community problems. The role of employment in developing social capital may help explain the positive effect that employment has on voting among people with disabilities, giving further weight to the ADA's goal of increasing employment among people with disabilities. Since voting is the most basic act of citizenship in a democracy, difficulty in voting at a polling place may contribute to a sense of being a second-class citizen, which can weaken connections to one's fellow citizens and decrease social capital among people with disabilities. Many of the difficulties faced in voting can be lessened or eliminated through efforts to ensure the full accessibility of polling places. Following are several policies to increase accessibility of 49 political elections that deserve careful consideration: federal standards establishing national standards for polling place accessibility for all people with disabilities, technical assistance to state and local election officials to promote accessibility, an information campaign for government service agencies to promote better implementation of the "Motor Voter Act," and federal standards to ensure that all campaign activities (speeches, commercials, etc.) are accessible to people with various impairments. Many people with disabilities see themselves as second-class citizens, as indicated by our finding that a substantial number feel that public officials treat them with less respect and dignity than other groups receive. Aggressive efforts to incorporate people with disabilities into political and civic activities, including but not limited to the policies listed above, can fight these perceptions and help ensure that this country fully benefits from the ideas, skills, and active citizenship of the 54 million Americans with disabilities. 50 REFERENCES Anspach, Renee (1979). "From stigma to identity politics: political activism among the physically disabled and former mental patients," Social Science and Medicine 13A: 765-773. Brehm, John, and Wendy Rahn (1997). "Individual-Level Evidence for the Causes and Consequences of Social Capital," American Journal of Political Science, Vol. 41, No. 3, pp. 999-1024. Cassel, Carol A., and David B. Hill (1981). "Explanations of turnout decline: a multivariate test," American Politics Quarterly 9: 181-195. Conway, M. Margaret (1991). Political Participation in the United States. Washington, DC: Congressional Quarterly. Gelman, Andrew, and Thomas C. Little (1998). "Improving on Probability Weighting for Household Size," Public Opinion Quarterly, 62, pp. 398-404. Hahn, Harlan (1985). "Toward a politics of disability: definitions, disciplines, and policies," The Social Science Journal 22: 87-105. (1987). "Civil rights for disabled Americans: The foundation of a political agenda." In Alan Gartner and Tom Joe (eds.), Images of the Disabled, Disabling Images. New York: Praeger Publishers. International Center for the Disabled (1986). The ICD Survey of Disabled Americans: Bringing Disabled Americans into the Mainstream. New York: International Center for the Disabled. Johnson, William G., and Marjorie Baldwin (1993). "The Americans with Disabilities Act: Will it make a difference?" Policy Studies Journal 21(4): 775-788. Kruse, Douglas; Alan Krueger; and Susan Drastal (1995). Disability, employment, and earnings in the dawn of the computer age, Report to the Rutgers Disability Research Consortium. School of Management and Labor Relations, Rutgers University. Kruse, Douglas (1997). Employment and Disability: Characteristics of Employed and Non- employed People with Disabilities, Report to the U.S. Department of Labor. School of Management and Labor Relations, Rutgers University. Knack, Stephen (1995). "Does 'motor voter' work? Evidence from state-level data," Journal of Politics, Vol. 57, August, pp. 796-811. Leighley, Jan E., and Nagler, Jonathan (1992). "Individual and systemic influences on voter turnout: who votes? 1984," Journal of Politics 54(3): 718-740. 51 LoBianca, Anthony F. (1998). "Are the Costs Too High? Political Participation Among Disabled Americans." Draft, Department of Political Science, University of Kentucky Louis Harris and Associates. 1998. 1998 N.O.D./Harris Survey of Americans with Disabilities. New York: Louis Harris and Associates. McNeil, John M. (1993). Americans with Disabilities: 1991-92, Current Population Reports, Household Economic Studies, P70-33. Washington, D.C.: Government Printing Office. Miller, Warren E., and J. Merrill Shanks (1996). The New American Voter. Cambridge, MA: Harvard University Press. Nardulli, Peter F.; Jon K. Dalager; and Donald E. Greco (1996). "Voter turnout in U.S. presidential elections: an historical view and some speculation," PS: Political Science and Politics 29: 480-490. Rosenstone, Steven, and John Hansen (1993). Mobilization, Participation and Democracy in America. New York: Macmillan Publishing Company. Schneider, Anne and Helen Ingram (1993a). "Social Constructions and Target Populations: Implications for Politics and Policy," American Political Science Review, 87:334-47. (1993b). "Constructing Citizenship: The Subtle Messages of Policy Design." In Helen Ingram and Steven Rathgeb Smith, eds., Public Policy for Democracy. Washington: Brookings Institution. (1995). "Social Constructions and Target Populations: Implications for Politics and Policy (Continued)," American Political Science Review, pp. 443-447. Schur, Lisa (1997). Disability and Political Participation. Ph.D. Dissertation, Department of Political Science, University of California-Berkeley. (1998). "Disability and the Psychology of Political Participation," Journal of Disability Policy Studies, 9(2), December. Schur, Lisa, and Douglas Kruse (1997). "What determines voter turnout? Lessons from citizens with disabilities." Draft, School of Management and Labor Relations, Rutgers University. Scotch, Richard (1988). "Disability as the basis for a social movement: advocacy and the politics of definition," Journal of Social Issues, 44(1). Shields, Todd; Kay Schriner; and Ken Schriner (1998a). "The disability voice in American politics: Political participation of people with disabilities in the 1994 election." Journal of Disability Policy Studies, 9(2), pp. 33-52. 52 (1998b). "Influences on the political participation of people with disabilities: The role of individual and elite factors in 1984 and 1986," Journal of Disability Policy Studies, 9(2), pp. 77-91. Shaffer, Stephen D. (1981). "A multivariate explanation of decreasing turnout in presidential elections, 1960-1976," American Journal of Political Science, 25(1): 68-95. Shapiro, Joseph (1993). No Pity: People With Disabilities Forging a New Civil Rights Movement. New York: Random House, Inc. Stoker, Laura, and M. Kent Jennings (1995). "Life-cycle transitions and political participation: the case of marriage," American Political Science Review, 89(2): 421-433. Stone, William, and Paul Schaffner (1988). The Psychology of Politics. New York: Springer- Verlag. Teixera, Ruy (1987). Why Americans Don't Vote: Turnout Decline in the United States, 1960- 1984. Westport, CN: Greenwood Press. U.S. Commission on Civil Rights (1983). Accommodating the Spectrum of Individual Abilities, Clearinghouse Publication 81. Washington, D.C.: U.S. Commission on Civil Rights. Verba, Sidney (1996). "The citizen as respondent: sample surveys and American democracy," American Political Science Review, 90, pp. 1-7. ; Kay Lehman Schlozman; and Henry E. Brady (1995). Voice and Equality: Civic Voluntarism in American Life. Cambridge, MA: Harvard University Press. Voss, D. Stephen, Andrew Gelman, and Gary King (1995). "Preelection Survey Methodology: Details from Eight Polling Organizations, 1988 and 1992," Public Opinion Quarterly, 59, pp. 98-132. Yuker, Harold E., ed. (1988). Attitudes Toward Persons with Disabilities. New York: Springer Publishing. 53 APPENDIX A: Survey Instrument (prior to CATI coding) 1 Questions for 1998 Disability/Election Participation 97-076 FINAL REVISED 10/28/98 RESPONDENT ID: SAMPLE STATE: SAMPLE COUNTY: AREA CODE: DAY: TIME INTERVIEW BEGAN: TIME INTERVIEW ENDED: LENGTH OF INTERVIEW: RESPONDENT SEX: 1. MALE 2. FEMALE ADD "Don't Know" as 8, 88 or 888 and "Refused" as 9, 99 or 999 for all questions Hello, my name is (first and last name). I'm on the staff of the Eagleton Poll at Rutgers University. We're conducting a study about people's health and elections. The survey will only take about 15 minutes of your time. A. Screening for disability A-1. To begin, including yourself, how many people regularly live in this household? / / / A-2. How many of these people are age 18 or older? / / / A-2a. Of those 18 or older, how many are citizens of the United States? / / / If A2a matches A2, then go to A3 If zero, thank and terminate If >0 but <A2 say "For our study we're focusing on citizens, so for the following questions I'd like to know just about those who are citizens." 2 A-3. Of those 18 or older, does anyone have a long-term health problem or impairment that limits the kind or amount of work, housework, or other activities he or she can do? INTERVIEWER NOTE: Other activities can include major activities such as school, parenting, or recreation 1. Yes 2. No 8. Don't know 9. Refused A-4. Does anyone age 18 or older have any of the following long-lasting conditions: INTERVIEWER NOTE: Does not include impairment corrected by ordinary eyeglasses or a hearing aid; "Long-lasting" = 6 months or more Yes, more than Don't Yes one (volun.) No Know Refused A-4a. Blindness? 1 2 3 8 9 A-4b. Deafness? 1 2 3 8 9 A-4c. Severe vision impairment? 1 2 3 8 9 A-4d. Severe hearing impairment? 1 2 3 8 9 A-5. Does anyone age 18 or older have a long-lasting condition that substantially limits one or more basic activities such as walking, climbing stairs, reaching, lifting, or carrying? 1. Yes 2. No 8. Don't know 9. Refused 3 A-6. Because of a physical, mental, or emotional condition lasting 6 months or more, does anyone age 18 or older have any difficulty in doing any of the following activities: Don't Yes No Know Refused A-6a. Learning? 1 2 8 9 A-6b. Remembering or concentrating? 1 2 8 9 A-6c. Dressing, bathing, or getting around inside the home? 1 2 8 9 A-6d. Going outside the home alone to shop or visit a doctor's office? 1 2 8 9 IF YES TO ANY QUESTION IN A-3 TO A-6, SKIP TO A-8 OTHERS ASK A-7 A-7. Is there anyone age 18 or older in your household, including yourself, who a) Considers himself or herself a person with a disability? 1. Yes Skip to question A-8 2. No 8. Don't know 9. Refused b) Most other people would consider a person with a disability? 1. Yes Skip to question A-8 2. No Skip to question C-1 8. Don't know Skip to question C-1 9. Refused Skip to question C-1 4 If one person in household AND A7a=1 or A7b=1 skip to B2 Otherwise, if one person in household skip to A-10 A-8. Is it just one person, or more than one person, who is age 18 or older and has one of the impairments or conditions I just asked about? / / / 88. Don't know 99. Refused .IF A8=1: If A7a=1 or A7b=1 skip to B2 If A-5=1, skip to A-10, Else skip to B-1 IF A8>1: READ A-9sel A-9sel. [TEXT ONLY] For the following questions, I'd like to ask you about the situation of just one person with the impairments or conditions I've mentioned - - the person with the most recent birthday. If A7a=1 or A7b=1 skip to B2 A-9. Is it this person who [ASK EACH ALTERNATIVE THAT FOLLOWS ONLY IF GAVE "YES" ANSWER IN A-3 TO A-6 THAT SOMEBODY IN HOUSEHOLD HAS THIS CONDITION]: Don't Yes No Know Refused A-9a. Is limited in the kind or amount of work, housework, or other activities he or she can do? 1 2 8 9 A-9bi. Is blind? 1 2 8 9 A-9bii. Is deaf? 1 2 8 9 A-9biii. Has a severe vision impairment? 1 2 8 9 A-9biv. Has a severe hearing impairment? 1 2 8 9 A-9c. Is limited in walking, climbing stairs, reaching, lifting, or carrying? 1 2 8 9 A-9di Has difficulty learning? 1 2 8 9 A-9dii Has difficulty remembering, or concentrating? 1 2 8 9 A-9e Has difficulty dressing, bathing, or getting around inside the home? 1 2 8 9 A-9f Has difficulty going outside the home alone to shop or visit a doctor's office? 1 2 8 9 5 IF ["YES" TO A-9c] OR ["YES" IN QA-5 AND ONE PERSON IN HOUSEHOLD], ASK A-10, OTHERS SKIP TO B-1 Don't A-10. Yes No Know Refused A-10a. Does this person (Do you) have difficulty walking 1 2 8 9 a quarter of a mile--about three city blocks? IF NO, DON'T KNOW OR REFUSED TO A-10a, SKIP TO A-10c A-10b. Is this person (Are you) able to walk a quarter of a mile 1 2 8 9 at all? A-10c. Does this person (Do you) use a wheelchair? 1 2 8 9 A-10d. Does this person (Do you) use a cane, crutches, or walker? 1 2 8 9 6 B. Disability questions once disability has been established ASK IF ANY TYPE OF DISABILITY Bla.B1. What medical condition or conditions cause this impairment or difficulty with activities? ASK B1b IF MORE THAN 1 MEDICAL CONDITION NAMED IN B1a: B1b. Which of these conditions would you say is the MAIN cause of any difficulty with activities this person has? [PROBE FOR CLEAR ANSWERS-RECORD UP TO 4] B1b-CIRCLE BELOW IF MAIN 1. 1 2. 2 3. 3 4. 4 8. Don't know 8 9. .Refused 9 ALL RESPONSES: SKIP TO B-3 ASK B-2 ONLY IF YES TO A-7a OR A-7b B2a. What medical condition or conditions cause this disability? [PROBE FOR CLEAR ANSWERS-RECORD UP TO 4] ASK B2b IF MORE THAN 1 MEDICAL CONDITION NAMED IN B2a: B2b. Which of these conditions would you say is the MAIN cause of any difficulty with activities this person has? B2b-CIRCLE BELOW IF MAIN 1. 1 2. 2 3. 3 4. 4 8. Don't know 8 9. Refused 9 7 ASK EVERYBODY WHO HAS BEEN IDENTIFIED AS HAVING A DISABILITY OR HAVING SOMEONE IN THE HOUSEHOLD WITH A DISABILITY B3. How old were you (was this person) when this medical condition began? IF MORE THAN ONE CONDITION NAMED IN Bla OR B2a, ASK ABOUT CONDITION IDENTIFIED AS "MAIN" CONDITION IN B1b OR B2b / / / Years old [IF NEEDED:" Your best estimate is fine"] 888. Don't know 999. Refused B4. Because of this medical condition, does this person (do you) need the help of another person with any daily activities either inside the home or getting around outside the home? 1. Yes 2. No 8. Don't know 9. Refused ASK TO SPEAK TO PERSON WITH DISABILITY, AND CONSIDER THEM AS PART OF "DISABILITY SAMPLE" IN FOLLOWING QUESTIONS. IF THE PERSON WITH THE DISABILITY IS NOT ABLE TO TALK ON THE TELEPHONE USE THE PERSON YOU ARE TALKING WITH AS A PROXY AND ASK THE QUESTIONS THAT FOLLOW ABOUT THE PERSON WITH THE DISABILITY ADD SECOND INTRO AND TRANSITION C. Voting and mobilization Now, I'd like to ask you some questions about elections. C-1. In talking to people about elections, we often find that a lot of people were not able to vote because they weren't registered, they were sick, or they just didn't have time. How about you-did you vote in the elections on November 3rd? 1. Yes 2. No Skip to question C-2. 8. Don't know Skip to question C-2. 9. Refused Skip to question C-2. 8 C-1a. Did you vote in person or by absentee ballot? 1. In person Skip to question C3 2. By absentee ballot 8. Don't know 9. Refused C-1b. Have you voted in person in any previous election? 1. Yes 2. No Skip to question C-4. 8. Don't know Skip to question C-4. 9. Refused Skip to question C-4. C-1c. What year was the last election in which you voted in person? / / / / / IF LATER THAN 1988- Skip to question C-3 IF 1988 OR EARLIER - Skip to question C-4 IF DON'T KNOW OR REFUSED - Skip to question C-4 ASK C-2 ONLY IF 2,8,OR 9 IN C-1.. C-2. Have you voted in any previous national election? 1. Yes 2. No Skip to question C-4 8. Don't know Skip to question C-4 9. Refused Skip to question C-4 C-2a. What year was the last election in which you voted? / / / / / IF LATER THAN 1988 - Ask question C-3 IF 1988 OR EARLIER - Skip to question C-4 IF DON'T KNOW OR REFUSED - Skip to question C-4 9 ASK IF VOTED IN-PERSON IN C-1a or VOTED SINCE 1988 IN C-2a C-3. Did you experience any difficulty in getting to the polling place or in using the ballot or voting machine? 1. Yes Ask question C-3a 2. No If voted this year 1998 (yes to C1), skip to C-6, else skip to C-5. 8. Don't know If voted this year 1998 (yes to C1), skip to C-6, else skip to C-5. 9. Refused If voted this year 1998 (yes to C1), skip to C-6, else skip to C-5. C-3a. What type of difficulty did you experience? [coded verbatim] All responses: If voted this year 1998 (yes to C1), skip to C-6, else skip to C-5 ASK ONLY IF 2,8, OR 9 IN C-2 C-4. If you wanted to vote in person inside the polling place, do you think you would experience any difficulty in getting to the polling place or in using the ballot or voting machine? 1. Yes Ask question C-4a 2. No If voted this year (yes to C1), skip to C-6, else skip to C-5. 8. Don't know If voted this year (yes to C1), skip to C-6, else skip to C-5. 9. Refused If voted this year (yes to C1), skip to C-6, else skip to C-5. C-4a. What type of difficulty do you think you would experience? [coded verbatim] All responses: If voted this year (yes to C1), skip to C-6, else skip to C-5 ASK IF DID NOT VOTE THIS YEAR C-5. Were you registered to vote in this election? 1. Yes Ask question C-6 2. No Skip to question C-7 3. (Vol) Registration not required to vote Skip to question C-7 8. Don't know Skip to question C-7 9. Refused Skip to question C-7 10 C-6. Did you register to vote after January 1, 1995? 1. Yes Ask C-6a 2. No Skip to question C-7 8. Don't know Skip to question C-7 9. Refused Skip to question C-7 C6a. There are various ways people can register to vote. Did you register to vote when you obtained or renewed your driver's license, or did you register in some other way? 1. When driver's license was obtained/renewed Skip to question C-7. 2. Some other way Ask C-6b 8. Don't know Skip to question C-7 9. Refused Skip to question C-7 C6b. How did you register to vote? DO NOT READ PRECODED OPEN-ENDED RESPONSES At an agency serving people with disabilities (for example, vocational rehabilitation agency, Easter seals, community mental health center) At some other public assistance agency (for example, Medicaid, AFDC, Food Stamp, or Unemployment office) Mailed in form to election office At a school, hospital, or on campus Went to a county or government voter registration office Filled out form at a registration drive (for example, political rally, someone came to your door, registration drive at mall, market, fair, post office, library, store, church, etc.) Other place/way Registered at the polls on election day 88 Don't know 99 Refused C-7. As you know, the political parties try to talk to as many people as they can to get them to vote for their candidate. Did anyone from one of the political parties call you up or come around and talk to you about the campaign this year? 1. Yes 2. No 8. Don't know 9. Refused C-8a. Other than someone from the two major parties, did anyone (else) call you up or come around and talk to you about supporting specific candidates in this election? 11 1. Yes Ask C-8b 2. No Go to C-9a 8. Don't know Go to C-9a 9. Refused Go to C-9a C-8b. Who talked to you-friends, family members, co-workers, representatives from some organizations, or someone else? CIRCLE ALL THAT APPLY..PROBE: Are there any others? 1. Friends 2. Family members 3. Co-workers 4. Representatives from political parties 5. Representatives from other organizations Ask question C-8c. 6. Someone else Ask question C-8d. 8. Don't know 9. Refused IF ORGANIZATIONAL REPRESENTATIVES--C-86 #5: C-8c. What organizations did they represent? IF SOMEONE ELSE-C-8b #6: C-8d. Who was that? C-9a. During the campaign this fall, did anyone talk to you about registering to vote or getting out to vote? 1. Yes 2. No Skip to question C-10 8. Don't know Skip to question C-10 9. Refused Skip to question C-10 12 C-9b. Who talked to you-friends, family members, co-workers, representatives from some organizations, or someone else? CIRCLE ALL THAT APPLY PROBE: Are there any others? 1. Friends 2. Family members 3. Co-workers 4. Representatives from political parties 5. Representatives from other organizations Ask question C-9c. 6. Someone else Ask question C-9d. 8. Don't know 9. Refused IF ORGANIZATIONAL REPRESENTATIVES--C-96 #5: C-9c. What organizations did they represent? IF SOMEONE ELSE-C-9b #6: C-9d. Who was that? C-10. In the Presidential election two years ago, a lot of people were not able to vote because they weren't registered, they were sick, or they just didn't have time. How about you-- did you vote in the national elections two years ago, in 1996? 1. Yes 2. No 8. Don't know 9. Refused 13 D. Political efficacy D-1. Some people seem to follow what's going on in government and public affairs most of the time, whether there's an election going on or not. Others aren't that interested. Would you say you follow what's going on in government and public affairs most of the time, some of the time, only now and then, or hardly at all? 1. Most of the time 2. Some of the time 3. Only now and then 4. Hardly at all 9. Don't know D-2. Please tell me how much you agree or disagree with these statements. Neither Agree Agree Agree Nor Disagree Disagree Don't Strongly Somewhat Disagree Somewhat Strongly Know The first is: DO NOT ROTATE ITEMS a. Public officials don't care much what people like me think. 1 2 3 4 5 9 b. People like me don't have any say about what the government does. 1 2 3 4 5 9 c. Sometimes politics and government seem so complicated that a person like me can't really understand what's going on. 1 2 3 4 5 9 d. I consider myself well-qualified to participate in politics. 1 2 3 4 5 9 e. I think that I am better informed about politics and government than most people. 1 2 3 4 5 9 f.. People with disabilities have as much influence as 1 2 3 4 5 9 other groups in government and public affairs. g. Government officials treat people with disabilities 1 2 3 4 5 9 with as much respect and dignity as they treat members of other groups. 14 D-3. I'm going to read you a list of several skills and activities. Please rate yourself with respect to how well you do on each of these as compared to most people. Compared to most people are you not as good, about the same, better, or much better at the (READ ITEM)? Not As About Much Don't Good The Same Better Better Know DO NOT ROTATE ITEMS a. Ability to work with others 1 2 3 4 9 b. Ability to speak in public 1 2 3 4 9 c. Ability to lead a group 1 2 3 4 9 d. Ability to compose an effective letter to an elected official 1 2 3 4 9 e. Communicating your ideas to others 1 2 3 4 9 E. Other forms of political/group participation E-1. In the last year, have you (READ EACH ITEM) or not? Don't Yes No Know a. Contributed money to a political party or candidate 1 2 9 b. Written or spoken to an elected representative or public official 1 2 9 c. Attended a political meeting 1 2 9 d. Written a letter to a newspaper 1 2 9 e. Contributed money to an organization which is trying to influence government policy or legislation 1 2 9 f. Otherwise worked with groups or on your own to change government laws or policies 1 2 9 IF NON-DISABILITY SAMPLE SKIP TO E-3 IF IN DISABILITY SAMPLE, AND RESPONED "YES" TO ANY QUESTION IN E-1a TO E-1f..ASK E-1g E-1g. Was any of this work on disability issues? 1. Yes 2. No 8. Don't know 9. Refused 15 IF IN DISABILITY SAMPLE: E-1h. (Have you) Worked to change a private organization's policies or practices affecting people with disabilities, such as through talking to business owners, or filing lawsuits? 1. Yes 2. No 8. Don't know 9. Refused E-2. DELETED E-3. In the last twelve months, have you worked with others or joined an organization in your community to do something about some community problem? 1. Yes 2. No 8. Don't know 9. Refused E-4a. Do you regularly meet with other people in any groups or organizations-for example, groups organized around service, hobbies, sports, social or political issues, or other interests? 1. Yes Ask question E-4b. 2. No Skip to question F-1 8. Don't know Skip to question F-1 9. Refused Skip to question F-1 E-4b. What types of groups or organizations are these? E-4c. How many times in an average month do you meet with these groups or organizations? / / / 16 IF IN DISABILITY SAMPLE, IF NOT GO TO F-1: E-5a. Do you regularly participate in any disability groups or organizations? 1. Yes 2. No Skip to question F-1 8. Don't know Skip to question F-1 9. Refused Skip to question F-1 IF YES: E-5b. What organizations are these? E-5c. What do you do in these organizations? F. Transportation F-1. How many days in an average week do you commute to work or go on trips for work, shopping, recreation, or any other reason? /Days F-2a. What do you use most often for basic transportation? DO NOT READ PRE-CODED OPEN-ENDED RESPONSES 1. Own car, van, or truck Ask question F-2b. 2. Family's car, van, or truck Ask question F-2b. 3. Someone else's car, van, or truck Ask question F-2b. 4. Public transportation Ask question F-2c 5. Para-transit Ask question F-2c. 6. Other Ask question F-2c IF CAR, VAN, OR TRUCK IN F-2a ASK: F-2b. Is that a vehicle that you can drive, or does someone else do all the driving? 1. You can drive Go to question F-3. 2. Someone else does all the driving Go to question F-3. 6. Other Go to question F-3. IF PUBLIC TRANSPORTATION OR OTHER #4 in F-2a ASK: F-2c. Do you have a car, van, or truck that you can drive? 17 1. Yes 2. No 8. Don't know 9. Refused F-3. Sometimes people face delays or other problems in transportation. When you want to go somewhere, how often do you face transportation delays or problems in getting there-- never, rarely, occasionally, often, very often, or always. 1. Never 2. Rarely 3. Occasionally 4. Often 5. Very often 6. Always 9. Other G. Employment G-1. Last week, did you have a job in which you do work for pay or profit? [Note: even if didn't actually work at the job last week] 1. Yes Ask question G-2. 2. No Skip to question H-1. 3. Don't know Skip to question H-1. G-2. How many hours per week do you usually work at your job? / / / / RANGE 1-168 777 (Vol)Hours vary 888 Don't know 999 Refused 18 G-3. What kind of work do you do, that is, what is your occupation (for example, plumber, typist, farmer)? [coded verbatim] IF NECESSARY, PROBE WITH: What are your usual duties or activities at this job? (For example, type, keep account books, file, sell cars, operate printing press, lay brick) / / / / G-4. On this job, are you a member of a labor union or of an employee association similar to a union? 1. Yes 2. No 9. Don't know H. Religion H-1a. Lots of things come up that keep people from attending religious services even if they want to. Thinking about your life these days, do you ever attend religious services, apart from occasional weddings, baptisms or funerals? 1. Yes Skip to question H-1c. 2. No Ask question H-1b. 9. Don't know Ask question H-1b. H-1b. Regardless of whether you now attend any religious services, do you ever think of yourself as part of a particular church or denomination? 1. Yes Skip to I-1 2. No Skip to I-1 9. Don't know Skip to I-1 IF YES: H-1c. Do you go to religious services every week, almost every week, once or twice a month, a few times a year, or never? 1. Every week 2. Almost every week 3. Once or twice a month 4. A few times a year 5. Never 9. Don't know I. Demographics 19 Now, I have a few questions so we can group your answers. I-1. Are you currently married, widowed, divorced, separated or never married? 1 Married 2 Widowed 3 Divorced 4 Separated 5 Never married I-2. Could you please tell me what year you were born? / / / / / Refused. I-3. How much school have you completed? a. Less than first grade b. 1st, 2nd, 3rd, or 4th grade c. 5th or 6th grade d. 7th or 8th grade e. 9th grade f. 10th grade g. 11th grade h. 12th grade, no diploma i. high school graduate j. GED k. Some college, but no degree 1. Associate degree in college-academic program m. Occupational/vocational certification program n. Bachelor's degree (BA, AB, BS) O. Master's degree (MA, MS, MENG, MED, MSW, MBA) p. Professional school degree (MD, DDS, DVM, LLB, JD) a Doctorate degree (PHD, EDD) I-4. Are you of Hispanic origin or descent? 1. Yes 2. No 8. Don't know 9. Refused 20 I-5. Do you consider yourself white, Black or African-American, Asian/Pacific Islander, Native American, or some other race? 1. White 2. Black or African-American 3. Asian/Pacific Islander 4. Native American, or 5. Some other race ASK I-6a and I-6b ONLY IF IN THE DISABILITY SAMPLE..OTHERS GO TO I-7a I-6a. Do you consider yourself to have a disability? 1. Yes 2. No 8. Don't Know 9. Refused I-6b. Would other people consider you to have a disability? 1. Yes 2. No 8. Don't Know 9. Refused ASK EVERYONE IF LIVING ALONE (A1=1), USE "YOUR TOTAL INCOME" IN NEXT QUESTION. I-7a. Considering all sources of income, would you say (your total income/the total combined income of all family members living in your household) in the past 12 months was--under $25,000, between $25,000 to just under $50,000, or more than $50,000? 1. Under $25,000 Ask I-7b 2. $25,000-$50,000 Ask I-7c 3. More than $50,000 Ask I-7d 8. Don't Know Skip to I-7e 9. Refused Skip to I-7e I-7b. Is that under $10,000, $10,000 to $15,000, $15,000 to $20,000, or $20,000 to $25,000? 1. Under $10,000 2. $10,000 to just under $15,000 3. $15,000 to just under $20,000 4. $20,000 to just under $25,000 8. Don't Know 9. Refused I-7c. Is that $25,000 to $30,000, $30,000 to $35,000,$35,000 to $40,000 or $40,000 to $50,000? 21 1. $25,000 to just under $30,000 2 $30,000 to just under $35,000 3 $35,000 to just under $40,000 4. $40,000 to just under $50,000 I-7d. Is that $50,000 to $60,000, $60,000 to $75,000,$75,000 or more? 1. $50,000 to just under $60,000 2. $60,000 to just under $75,000 3. $75,000 or more? 8. Don't know 9. Refused I-7e. IF DON'T KNOW OR REFUSED IN I-7a: Would you say it is greater or less than $30,000? 1. Greater than $30,000 10 Less than $30,000 10 Don't know 9. Refused I-8. BY OBSERVATION (ASK ONLY IF NECESSARY) I am required to ask whether you are male or female. 1. Male 2. Female You've been very helpful. Sometimes we need to re-contact people to ask some follow-up questions. Would it be okay for us to re-contact you? (IF YES, ASK: Could I please just have your first name? 1. Yes RECORD NAME 2. No Sex of Interviewer: 1. Male 2. Female THANK YOU VERY MUCH, YOU'VE BEEN VERY HELPFUL. APPENDIX B 2000 Census questions regarding disability 1. Does this person have any of the following long-lasting conditions: blindness, deafness, or a severe vision or hearing impairment? 2. Does this person have a long-lasting condition that substantially limits one or more basic activities such as walking, climbing stairs, reaching, lifting, or carrying? 3. Because of a physical, mental, or emotional condition lasting 6 months or more, does this person have any difficulty in doing any of the following activities: a. Learning, remembering, or concentrating? b. Dressing, bathing, or getting around inside the home? C. Going outside the home alone to shop or visit a doctor's office? d. Working at a job or business? TABLE 1: Demographic Characteristics and Comparison to Other Samples Current Survey SIPP^ With Without With Without Disability Disability Disability Disability (1) (2) (3) (4) Overall 100.0% 100.0% 100.0% 100.0% Gender^^ Female 54.6% 51.1% 54.6% 51.1% Marital Status Married, not separated 48.4% 61.5% 52.3% 59.9% Separated 3.2% 5.2% 3.3% 2.5% Divorced 13.7% 9.8% 11.9% 8.6% Widowed 18.7% 1.6% 17.2% 3.6% Never married 16.0% 21.9% 15.5% 25.4% Race White 80.6% 82.9% 84.9% 84.9% African-American 12.6% 9.1% 12.2% 11.1% Asian/Pacific Islander 0.6% 1.8% 3.4% 2.0% Native American 1.9% 1.0% 0.8% 0.6% "Other" 4.2% 5.2% Ethnicity Hispanic origin 5.8% 6.4% 7.6% 9.1% Age Mean 55.9 42.6 55.1 40.2 18-24 5.5% 14.7% 5.2% 16.3% 25-34 9.8% 22.0% 11.4% 25.8% 35-44 15.6% 21.6% 15.1% 24.1% 45-54 16.8% 19.4% 15.3% 15.4% 55-64 14.3% 9.1% 16.3% 9.0% 65+ 38.0% 13.2% 36.8% 9.4% Region^^ Northeast 18.5% 21.0% 18.5% 21.0% Midwest 25.3% 25.5% 25.3% 25.5% South 35.6% 33.5% 35.6% 33.5% West 20.6% 20.0% 20.6% 20.0% City status Urban 26.7% 27.3% Suburban 46.0% 53.0% Rural 27.3% 19.7% Sample size (unweighted^^) 700 540 18,060 53,163 ^ Survey of Income and Program Participation data from 1993-94 for those age 18+. The SIPP disability definition includes impairments, activity limitations, and specific health conditions as described in McNeil (1993) and Kruse (1997). AA All figures in the current survey have been weighted by gender and region to reflect national distributions, and by household size to account for differential probabilities of selection (see text). This accounts for the identical gender and region breakdowns for the SIPP sample and current sample. TABLE 2: Disability Characteristics Figures pertain to those in disability samples. Current SIPP survey (age 18+) (1) (2) Limitation in work, housework, or other activities 59.8% Limitation in work or housework 61.7% Severe vision impairment 18.7% 18.1% Blindness 6.0% 3.8% Severe hearing impairment 17.4% 22.7% Deafness 2.3% 2.3% Difficulty with walking, climbing stairs, reaching, lifting, or carrying 62.8% 54.5% Difficulty walking 1/4 mile 45.9% 38.8% Unable to walk 1/4 mile 22.6% 20.3% Use wheelchair 8.3% 3.2% Use cane, crutches, walker 23.0% 11.0% Learning disability 10.0% 4.9% Difficulty remembering or concentrating 30.0% Mental cndn. except learning dis. 11.7% Difficulty dressing, bathing, or getting around inside home 17.4% 13.5% Difficulty going outside the home alone to shop or visit doctor's office 25.2% 17.5% Need help with any daily activities 31.9% 20.0% Consider self to have disability If none of above impairments or limitations 2.8% Among all in disability sample 58.4% Others consider one to have disability If none of above impairments or limitations 0.8% Among all in disability sample 59.5% Age at which impairment/limitation began At birth or in first year 8.1% 1-17 11.1% 18-34 19.4% 34-49 20.1% 50-64 20.5% 65+ 19.9% SIPP data are from Survey of Income and Program Participation, disability supplement, 1993-94. See McNeil (1993) and Kruse (1997) for survey questions and definitions, and further results. TABLE 3: Voter Turnout and Registration Disability Non-disability Sample Sample Difference (1) (2) Voted in November 1998 53.4% 59.5% -6.1% ** At polling place 46.0% 54.7% -8.7% ** By absentee ballot 7.4% 4.8% 2.6% * Voted in November 1996 67.6% 73.2% -5.6% ** Last voted: In November 1998 53.4% 59.5% -6.1% ** Not in November 1998 but within past 10 years 21.1% 19.1% 2.0% More than 10 years ago 2.1% 0.6% 1.5% ** Never voted 17.2% 18.3% -1.1% Don't know when last voted 5.4% 2.5% 2.9% ** Don't know if ever voted 1.6% 0.0% 1.6% ** Registered to vote for 1998 elections 78.5% 81.1% -2.6% Registered to vote since 1/1/95 24.9% 26.7% -1.8% Driver's license agency 9.3% 11.2% -1.9% Disability agency 0.5% 0.3% 0.2% Other public assistance agency 0.7% 0.7% 0.0% Mailed form to election office 1.6% 0.6% 1.0% * At school, hospital, or on campus 1.1% 2.1% -1.0% County or gov't registration office 2.6% 2.8% -0.2% Registration drive 2.0% 3.0% -1.0% At polls on election day 1.4% 1.9% -0.5% Other way 1.5% 1.5% 0.0% Don't know how registered 4.1% 2.6% 1.5% Sample size 700 540 * Statistically significant difference at p<.10 ** p<.05 TABLE 4: Problems in Voting at Polling Place Encountered problems, if Would expect problems, if voted at polling place since 1988 haven't voted at polling place since 1988 Disability Non-disability Disability Non-disability Sample Sample Difference Sample Sample Difference (1) (2) (3) (4) (5) (6) Any problems encountered/expected in voting 8.2% 1.7% 6.5% ** 27.5% 3.6% 23.9% ** If voted in 1998 5.8% 1.9% 3.9% ** If voted in 1989-97 12.2% 1.1% 11.1% ** Don't know if would have problems 4.4% 0.6% 3.8% ** Types of problems encountered or expected (as percent of all voters or potential voters) General mobility 1.7% 0.0% 1.7% 10.3% 0.0% 10.3% Difficulty standing 0.9% 0.0% 0.9% 5.1% 0.0% 5.1% Difficulty walking 0.8% 0.0% 0.8% 5.2% 0.0% 5.2% Getting to polling place 1.6% 0.9% 0.7% 8.8% 0.0% 8.8% Finding polling place 0.5% 0.6% -0.1% 0.0% 0.0% 0.0% Too far away 0.2% 0.3% -0.1% 1.0% 0.0% 1.0% Transportation problem 0.9% 0.0% 0.9% 7.8% 0.0% 7.8% Getting inside polling place (stairs/steps) 1.0% 0.0% 1.0% 0.4% 0.0% 0.4% Inside polling place 2.9% 0.1% 2.8% 9.7% 3.2% 6.5% Long lines 0.6% 0.0% 0.6% 0.0% 0.0% 0.0% Difficulty with booth 0.5% 0.1% 0.4% 0.0% 0.0% 0.0% Can't read 0.0% 0.0% 0.0% 2.5% 0.0% 2.5% Difficulty seeing 1.3% 0.0% 1.3% 2.7% 0.0% 2.7% Difficulty understanding how to vote/use machine 0.5% 0.0% 0.5% 4.5% 3.2% 1.3% Other problem 1.4% 0.5% 0.9% 2.9% 0.5% 2.4% Sample size 515 425 185 115 * Statistically significant difference at p<.10 ** p<.05 TABLE 5: Other Forms of Participation Figures refer to those who have done activity within past 12 months. With Without Disability Disability Difference (1) (2) (3) Total number of first seven activities (average) 1.15 1.47 -0.32 ** Done any of first seven activities 51.3% 61.0% -9.7% ** Any of first six on disability issues 8.4% 1) Contributed money to political party or candidate 12.7% 15.6% -2.9% 2) Written or spoken to elected representative or 30.3% 39.8% -9.5% ** public official 3) Attended political meeting 12.5% 14.6% -2.1% 4) Written letter to newspaper 7.3% 9.0% -1.7% 5) Contributed money to org. trying to influence 16.4% 21.9% -5.5% ** gov't. policy or legislation 6) Otherwise worked with groups or on one's own 14.3% 19.1% -4.8% ** to change gov't laws or policies 7) Worked with others on community problem 24.2% 27.6% -3.4% 8) Worked to change private org.'s policies 10.1% regarding people with disabilities * Statistically significant difference at p<.10 ** p<.05 TABLE 6: Demographic Characteristics and Participation Distribution within Political participation if in group at left samples (proportions) Voter turnout Other Forms (avg. number) With Without With Without With Without Disability Disability Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (5) (6) (6) Overall 1.000 1.000 53.4% 59.5% -6.1% ** 1.15 1.47 -0.32 ** Age 18-29 0.095 0.246 28.8% 33.6% -4.8% 0.88 1.09 -0.21 30-44 0.211 0.336 45.7% 55.5% -9.8% * 1.18 1.37 -0.19 45-64 0.311 0.285 53.7% 72.9% -19.2% ** 1.42 1.86 -0.44 ** 65+ 0.383 0.132 61.9% 91.5% -29.6% ** 1.01 1.76 -0.75 ** 18-30 0.104 0.267 28.9% 34.3% -5.4% 0.89 1.07 -0.18 31-40 0.132 0.247 42.2% 51.8% -9.6% 1.12 1.31 -0.19 41-50 0.170 0.198 57.4% 63.4% -6.0% 1.38 1.62 -0.24 51-60 0.161 0.127 46.9% 85.3% -38.4% ** 1.40 2.07 -0.67 ** 61-70 0.148 0.098 68.8% 85.5% -16.7% ** 1.26 1.92 -0.66 * 71-80 0.193 0.050 59.6% 95.1% -35.5% ** 1.11 1.79 -0.68 * 80+ 0.092 0.014 57.1% 100.0% -42.9% ** 0.60 1.64 -1.04 Gender Female 0.544 0.511 51.3% 60.5% -9.2% ** 1.04 1.27 -0.23 ** Male 0.456 0.489 55.9% 58.4% -2.5% 1.29 1.68 -0.39 ** Marital Status Married, spouse present 0.484 0.615 57.9% 65.1% -7.2% * 1.42 1.60 -0.18 Separated/divorced 0.168 0.114 45.6% 57.9% -12.3% 0.98 1.40 -0.42 * Widowed 0.187 0.052 57.3% 90.0% -32.7% ** 0.76 1.11 -0.35 Never married 0.161 0.219 42.5% 36.9% 5.6% 1.01 1.23 -0.22 Race White 0.806 0.829 54.4% 60.4% -6.0% * 1.17 1.44 -0.27 ** African-American 0.126 0.091 49.7% 56.3% -6.6% 0.85 1.50 -0.65 ** All others 0.068 0.080 49.4% 49.1% 0.3% 1.40 1.78 -0.38 Ethnicity Hispanic origin 0.058 0.064 40.5% 37.3% 3.2% 1.05 1.47 -0.42 Non-hispanic origin 0.942 0.936 54.2% 61.0% -6.8% ** 1.16 1.48 -0.32 ** Education 8th grade or less 0.091 0.030 28.1% 57.6% -29.5% ** 0.38 0.55 -0.17 Some H.S., no degree 0.177 0.078 40.7% 40.6% 0.1% 0.75 0.75 0.00 H.S. degree or GED 0.293 0.193 52.2% 50.7% 1.5% 0.88 0.92 -0.04 More than H.S. 0.440 0.700 64.0% 64.0% 0.0% 1.70 1.75 -0.05 B.A. degree 0.094 0.191 66,5% 64.8% 1.7% 2.03 1.87 0.16 Graduate degree 0.057 0.140 77.0% 69.2% 7.8% 2.36 2.65 -0.29 * Statistically significant difference at p<.10 ** p<.05 TABLE 7: Prediction of Voting with Demographic Characteristics Prediction of voting in 1998 using probit regressions.^ Sample: Disability sample Non-disability sample Full sample Full sample Full sample Indep. variables (1) (2) (3) (4) (5) Female -0.064 (1.42) -0.011 (0.22) -0.012 (0.27) -0.016 (0.35) -0.014 (0.31) Married 0.067 (0.92) 0.079 (1.12) 0.083 (1.31) 0.075 (1.16) 0.086 (1.35) Separated/divorced -0.012 (0.15) -0.004 (0.04) 0.006 (0.07) -0.007 (0.08) 0.010 (0.12) Widowed 0.012 (0.13) 0.202 (1.50) 0.084 (0.91) 0.122 (1.33) 0.082 (0.87) African-American 0.048 (0.72) 0.085 (1.01) 0.071 (1.00) 0.076 (1.08) 0.076 (1.07) Asian/Pacific Islander -0.325 (1.25) -0.333 (1.31) -0.301 (1.24) -0.330 (1.35) -0.307 (1.28) Native American -0.187 (1.04) -0.089 (0.36) -0.108 (0.52) -0.112 (0.54) -0.111 (0.53) Other race 0.051 (0.53) 0.031 (0.28) 0.027 (0.27) 0.032 (0.33) 0.030 (0.30) Some HS (9-12) 0.210 (2.34) ** 0.080 (0.47) 0.110 (0.98) 0.107 (0.93) 0.025 (0.15) HS degree 0.320 (3.85) ** 0.188 (1.16) 0.207 (1.94) 0.206 (1.84) * 0.126 (0.81) Some college 0.406 (4.96) ** 0.356 (2.35) ** 0.360 (3.52) ** 0.362 (3.40) ** 0.301 (2.09) ** AA degree 0.328 (3.62) ** 0.285 (1.88) ** 0.291 (2.77) ** 0.291 (2.68) ** 0.239 (1.62) BA degree 0.406 (4.73) ** 0.354 (2.42) ** 0.359 (3.56) ** 0.360 (3.46) ** 0.304 (2.16) ** Graduate degree 0.435 (4.78) ** 0.343 (2.40) ** 0.355 (3.52) : 0.352 (3.37) ** 0.303 (2.17) ** Age 0.006 (4.13) ** 0.015 (7.64) ** 0.014 (8.58) ** 0.016 (8.23) ** 0.014 (8.57) ** Disability -0.197 (4.93) ** 0.307 (3.58) ** Disability*age -0.011 (5.15) ** Disability*No HS -0.422 (2.63) ** Disability*Some HS -0.175 (1.87) * Disability*HS degree -0.142 (2.00) ** Disability*Some college -0.193 (2.33) ** Disability*AA degree -0.262 (2.41) ** Disability*BA degree -0.187 (1.79) * Disability*Grad degree -0.086 (0.66) Log-likelihood -408.19 -287.57 -667.09 -658.79 -665.58 n 650 521 1171 1171 1171 * p<.10 ** p<.05 ^ Reported numbers are derivatives (dF/dX), represented estimated change in probability for one-unit change in independent variable, assuming other variables are at mean values. TABLE 8: Prediction of Other Forms of Participation with Demographic Characteristics Any Other Participation (probits^) Number of Forms of Other Participation (OLS) Indep. variables (1) (2) (3) (4) (5) (6) Female -0.085 (1.87) * -0.088 (1.94) * -0.085 (1.87) * -0.313 (2.45) ** -0.317 (2.48) ** -0.315 (2.45) Married 0.084 (1.31) 0.081 (1.26) 0.084 (1.31) 0.100 (0.54) 0.092 (0.50) 0.101 (0.55) Separated/divorced 0.023 (0.28) 0.018 (0.22) 0.023 (0.27) -0.164 (0.74) -0.175 (0.79) -0.165 (0.74) Widowed 0.037 (0.34) 0.055 (0.50) 0.037 (0.34) -0.465 (1.73) * -0.413 (1.52) -0.460 (1.71) African-American 0.118 (1.61) 0.121 (1.66) 0.117 (1.59) 0.364 (1.74) * 0.368 (1.77) 0.364 (1.73) Asian/Pacific Islander -0.356 (1.69) * -0.373 (1.78) * -0.357 (1.70) * 0.402 (0.37) 0.362 (0.34) 0.398 (0.37) Native American 0.023 (0.11) 0.021 (0.10) 0.025 (0.12) -0.097 (0.19) -0.099 (0.20) -0.083 (0.16) Other race 0.085 (0.93) 0.088 (0.96) 0.084 (0.92) 0.343 (1.17) 0.350 (1.21) 0.338 (1.15) Some HS (9-12) 0.226 (2.18) : 0.230 (2.23) ** 0.228 (1.64) 0.476 (1.95) * 0.471 (1.94) * 0.418 (1.26) HS degree 0.303 (3.09) ** 0.305 (3.11) ** 0.300 (2.28) ** 0.667 (3.05) ** 0.654 (2.97) ** 0.608 (2.05) ** Some college 0.365 (3.80) ** 0.369 (3.86) ** 0.360 (2.84) ** 1.219 (5.00) ** 1.216 (4.97) ** 1.163 (3.68) ** AA degree 0.361 (3.85) ** 0.363 (3.87) ** 0.356 (2.93) ** 1.099 (4.40) ** 1.094 (4.36) ** 1.002 (3.11) ** BA degree 0.446 (5.09) ** 0.449 (5.16) ** 0.444 (3.93) ** 1.663 (6.57) ** 1.666 (6.57) ** 1.593 (4.97) ** Graduate degree 0.471 (6.01) ** 0.472 (6.06) ** 0.471 (4.82) : 2.325 (8.09) ** 2.317 (8.04) ** 2.279 (6.51) ** Age 0.005 (3.40) ** 0.006 (3.63) ** 0.005 (3.39) ** 0.021 (4.58) ** 0.024 (4.72) ** 0.021 (4.53) ** Disability -0.082 (2.06) ** 0.263 (3.01) ** -0.173 (1.57) 0.876 (3.28) ** Disability*age -0.007 (3.58) ** -0.020 (3.70) ** Disability*No HS -0.096 (0.58) -0.374 (1.25) Disability*Some HS -0.111 (1.14) -0.185 (0.73) Disability*HS degree -0.086 (1.25) -0.185 (1.15) Disability*Some college -0.056 (0.79) -0.239 (1.10) Disability*AA degree -0.033 (0.33) 0.151 (0.47) Disability*BA degree -0.073 (0.75) -0.035 (0.11) Disability*Grad degree -0.180 (1.46) -0.439 (1.07) Constant -0.593 (1.90) * -0.705 (2.17) ** -0.525 (1.43) Log-likelihood -711.42 -706.98 -711.18 R-squared 0.198 0.203 0.199 n 1180 1180 1180 1180 1180 1180 * p<.10 ** p<.05 A For probits, reported numbers are derivatives (dF/dX), represented estimated change in probability for one-unit change in independent variable, assuming other variables are at mean values. TABLE 9: Disability Characteristics, Participation, and Voting Difficulties Voted in 1998 Difficulties in voting Other forms of participation Diff. from Absentee Registered Avg. Diff. from At all non-disability ballot to vote Actual^ Expected^ number non-disability (1) (2) (3) (4) (5) (6) (7) (8) Any disability 53.4% -6.1% 7.4% 78.5% 8.2% 27.5% 1.16 -0.31 Severe disability 48.8% -10.7% 7.8% 75.3% 13.2% 43.6% 0.98 -0.49 Limited in work, housework, or other activities 50.7% -8.8% 6.2% 78.1% 10.6% 34.3% 1.17 -0.30 Severe vision impairment 47.3% -12.2% 6.0% 77.2% 14.9% 52.0% 1.02 -0.45 Blind 52.5% -7.0% 4.5% 79.9% 17.6% 43.5% 0.98 -0.49 Severe hearing impairment 58.8% -0.7% 10.6% 84.9% 7.8% 41.7% 1.14 -0.33 Difficulty with walking, climbing stairs, reaching, lifting, or carrying 53.2% -6.3% 8.3% 79.6% 9.1% 33.5% 1.20 -0.27 Difficulty walking 1/4 mile 49.9% -9.6% 8.3% 79.5% 11.1% 39.6% 1.10 -0.37 Unable to walk 1/4 mile 46.3% -13.2% 8.9% 79.4% 8.9% 50.3% 0.87 -0.60 Use wheelchair 38.2% -21.3% 13.5% 72.1% 15.3% 45.1% 1.12 -0.35 Use cane, crutches, walker 48.0% -11.5% 12.4% 78.4% 13.5% 50.9% 1.18 -0.29 Difficulty learning 33.4% -26.1% 2.8% 70.2% 17.1% 57.4% 0.85 -0.62 Difficulty remembering or concentrating 41.4% -18.1% 7.8% 72.5% 10.9% 40.7% 0.88 -0.59 Difficulty dressing, bathing, or getting around inside home 39.3% -20.2% 7.5% 67.7% 14.5% 52.3% 1.01 -0.46 Difficulty going outside home alone to shop or visit doctor's office 35.8% -23.7% 6.7% 72.8% 23.9% 52.7% 0.86 -0.61 Consider self to have disability Yes 47.0% -12.5% 7.5% 75.3% 11.8% 37.3% 1.03 -0.44 No 62.2% 2.7% 7.5% 82.9% 3.7% 8.6% 1.34 -0.13 Need help with any daily activities 42.3% -17.2% 5.4% 70.8% 19.2% 49.6% 0.97 -0.50 A "Actual" difficulties refer to those who voted since 1988, while "expected" difficulties refer to those who have not voted since 1988. See Table 2 for prevalence of conditions. TABLE 10: Prediction of Participation with Disability Characteristics Other Forms of Participation Whether Voted in 1998 (probits^) Any (probits^) Number (OLS) Indep. variables (1) (2) (3) (4) (5) (6) Functional limitations: Blind -0.060 (0.54) 0.007 (0.07) -0.116 (1.35) -0.091 (1.06) -0.175 (0.85) -0.112 (0.54) Severe vision impairment, but not blind -0.182 (2.12) ** -0.167 (1.92) * -0.028 (0.40) -0.027 (0.38) -0.024 (0.11) -0.001 (0.00) Severe hearing impairment -0.045 (0.68) -0.052 (0.77) -0.054 (0.89) -0.058 (0.95) -0.196 (1.16) -0.199 (1.17) Use wheelchair -0.345 (3.63) ** -0.154 (1.37) 0.055 (0.68) 0.126 (1.38) 0.072 (0.35) 0.247 (1.01) No wheelchair, but cane/crutches/walker -0.172 (2.56) ** -0.059 (0.77) -0.021 (0.35) 0.021 (0.32) -0.041 (0.24) 0.043 (0.23) No mobility aid, cannot walk 1/4 mile -0.198 (2.22) ** -0.176 (1.80) * -0.215 (2.64) ** -0.206 (2.42) ** -0.456 (2.49) ** -0.429 (2.16) ** No mobility aid, can walk 1/4 mile with difficulty -0.104 (1.50) -0.108 (1.40) 0.041 (0.68) 0.045 (0.70) 0.008 (0.05) 0.012 (0.07) Otherwise restricted in lifting, etc. 0.047 (0.56) 0.045 (0.51) 0.096 (1.47) 0.100 (1.48) 0.434 (2.01) ** 0.435 (1.96) ** Difficulty.learning -0.054 (0.64) -0.005 (0.06) 0.020 (0.26) 0.046 (0.61) 0.078 (0.42) 0.138 (0.74) Difficulty remembering or concentrating -0.159 (2.67) ** -0.115 (1.98) ** -0.116 (2.28) ** -0.101 (1.97) ** -0.296 (2.15) ** -0.247 (1.75) * Activity limitations: Limited in work, housework, other activity 0.031 (0.57) 0.001 (0.02) 0.036 (0.24) Difficulty inside home -0.055 (0.68) -0.036 (0.51) 0.093 (0.53) Difficulty going outside home alone -0.302 (4.07) ** -0.107 (1.66) * -0.329 (2.19) ** Need help with daily activities 0.002 (0.03) 0.014 (0.26) -0.056 (0.39) Log-likelihood -662.94 -658.84 -709.51 -708.3 R-squared 0.200 0.201 n 1171 1170 1180 1179 1180 1179 * p<.10 ** p<.05 All regressions include the demographic variables from the Table 7 regressions. TABLE 11: Mobilization in 1998 Elections Proportion contacted Voting rate With Without With Without Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (6) Contacted by either major political party Yes 0.306 0.370 -0.064 ** 65.5% 72.4% -6.9% No 47.0% 52.1% -5.1% Contacted by someone else about supporting specific candidates Yes 0.111 0.092 0.019 53.6% 70.4% -16.8% * No 52.8% 58.3% -5.5% Talked to by anyone about registering or getting out to vote Yes 0.238 0.235 0.003 59.1% 70.7% -11.6% No 51.1% 55.9% -4.8% Contacted about supporting candidates by: Any political party 0.325 0.377 -0.052 * Friends 0.013 0.011 0.002 Family members 0.002 0.003 -0.001 Co-workers 0.000 0.000 0.000 Organizations 0.025 0.029 -0.004 Someone else 0.003 0.009 -0.006 Talked to by anyone about registering or getting out to vote by: Friends 0.039 0.049 -0.010 Family members 0.010 0.029 -0.019 ** Co-workers 0.015 0.029 -0.014 * Political parties 0.092 0.092 0.000 Other organizations 0.050 0.032 0.018 Someone else 0.019 0.029 -0.010 * Statistically significant difference at p<.10 ** p<.05 TABLE 12: Political Efficacy, Competencies, and Participation Distribution within Participation if in group at left samples (proportions) Voter turnout Number of other forms With Without With Without With Without Disability DisabilityDifference Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (6) (7) (8) (9) Overall 1.000 1.000 53.4% 59.5% -6.1% ** 1.15 1.47 -0.32 ** 1. Follow gov't and public affairs: Most of time 0.575 0.581 -0.006 64.8% 71.6% -6.8% * 1.55 1.97 -0.42 ** Some of time 0.225 0.280 -0.055 ** 45.2% 51.7% -6.5% 0.84 0.96 -0.12 Now and then or hardly at all 0.190 0.139 0.051 ** 27.7% 24.4% 3.3% 0.39 0.42 -0.04 2. Public officials don't care much about Agree 0.565 0.474 0.091 ** 50.2% 53.4% -3.2% 1.08 1.47 -0.39 ** people like me Disagree 0.286 0.440 -0.154 ** 57.7% 65.3% -7.6% 1.56 1.48 0.08 3. People like me don't have say about gov't Agree 0.517 0.412 0.105 ** 48.1% 53.9% -5.8% 0.93 1.28 -0.35 ** Disagree 0.394 0.542 -0.148 ** 61.6% 64.1% -2.5% 1.58 1.69 -0.11 4. Politics and gov't sometimes too Agree 0.561 0.442 0.119 ** 49.6% 57.3% -7.7% * 0.89 1.16 -0.27 ** complicated to understand Disagree 0.380 0.519 -0.139 ** 58.2% 61.8% -3.6% 1.60 1.72 -0.12 5. Am well-qualified to participate in politics Agree 0.614 0.695 -0.081 ** 58.6% 61.6% -3.0% 1.46 1.69 -0.23 * Disagree 0.314 0.253 0.061 ** 42.5% 53.4% -10.9% * 0.66 0.91 -0.25 * 6. Am better-informed about gov't/politics Agree 0.470 0.504 -0.034 63.2% 68.0% -4.8% 1.66 2.03 -0.37 ** than most people Disagree 0.377 0.367 0.010 39.2% 47.0% -7.8% 0.65 0.83 -0.18 7. People w/disabilities have as much Agree 0.622 0.690 -0.068 ** 56.2% 60.1% -3.9% 1.24 1.51 -0.27 ** influence as other groups in gov't. Disagree 0.262 0.222 0.040 49.5% 59.2% -9.7% 1.09 1.45 -0.36 8. Gov't officials treat people w/disabilities Agree 0.493 0.592 -0.099 ** 56.5% 61.7% -5.2% 1.23 1.48 -0.25 * with as much respect as others Disagree 0.334 0.247 0.087 ** 48.6% 53.1% -4.5% 1.25 1.54 -0.29 Better than others at: 9. Ability to work with others 0.498 0.677 -0.179 ** 57.8% 63.4% -5.6% 1.44 1.65 -0.21 10. Ability to speak in public 0.276 0.342 -0.066 ** 59.0% 63.9% -4.9% 1.89 2.24 -0.35 * 11. Ability to lead a group 0.359 0.524 -0.165 ** 63.9% 68.4% -4.5% 1.90 2.13 -0.23 12. Ability to compose an effective 0.314 0.428 -0.114 ** 60.3% 61.8% -1.5% 1.82 1.93 -0.11 letter to elected official 13. Communicating ideas to others 0.451 0.601 -0.150 ** 59.0% 60.2% -1.2% 1.64 1.73 -0.09 TABLE 13: Employment and Participation Distribution within Political participation rates if in group at left samples (proportions) Voter turnout Number of other forms With Without With Without With Without Disability Disability Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (6) Age 18-64 Employed 0.467 ** 0.795 53.0% 54.8% -1.8% 1.52 1.50 0.02 Not employed 0.533 ** 0.205 42.6% 55.3% -12.7% * 1.02 1.27 -0.25 Age 65+ Employed 0.075 ** 0.175 69.1% 80.7% -11.6% 1.75 1.44 0.31 Not employed 0.925 ** 0.825 61.3% 93.8% -32.5% ** 0.94 1.83 -0.89 ** If employed Part time^ 0.275 ** 0.176 45.2% 51.7% -6.5% 1.72 1.34 0.38 Full-time^ 0.700 ** 0.818 58.8% 55.9% 2.9% 1.43 1.51 -0.08 Union member Yes 0.184 0.155 72.6% 58.6% 14.0% 1.87 2.16 -0.29 No 0.816 0.845 51.2% 55.1% -3.9% 1.48 1.37 0.11 Spoken to by co-workers about voting 0.026 0.039 * Significant difference between disability and non-disability samples at 90% level ** 95% level ^ Part-time is usual weekly hours of 1-34, while full-time is usual weekly hours of 35 or more. Part-time and full-time figures do not sum to one due to missing hours data. TABLE 14: Transportation and Political Participation Distribution within Participation rates if in group at left samples (proportions) Voter turnout Number of other forms With Without With Without With Without Disability Disability Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (6) (7) (8) Overall 1.000 1.000 53.4% 59.5% -6.1% ** 1.15 1.47 -0.32 ** Days of travel in avg. week Mean 3.34 ** 5.18 0-2 0.441 ** 0.130 49.7% 62.6% -12.9% * 0.89 1.08 -0.20 3-4 0.150 * 0.111 58.1% 70.0% -11.9% 1.46 1.61 -0.15 5-7 0.409 ** 0.759 55.9% 57.8% -1.9% 1.37 1.52 -0.15 If employed 0-4 days 0.149 0.139 56.5% 56.3% 0.2% 2.48 1.36 1.12 ** 5-7 days 0.851 0.861 55.0% 55.6% -0.6% 1.37 1.57 -0.20 If not employed, age 18-64 0-4 days 0.786 ** 0.480 43.8% 60.7% -16.9% ** 1.01 0.84 0.17 5-7 days 0.214 ** 0.520 44.5% 50.1% -5.6% 1.10 1.57 -0.47 If not employed, age 65+ 0-4 days 0.823 ** 0.509 58.5% 92.1% -33.6% ** 0.80 2.03 -1.23 ** 5-7 days 0.177 ** 0.491 71.1% 95.4% -24.3% ** 1.69 1.62 0.07 Usual transportation mode Own or family vehicle 0.812 ** 0.916 56.2% 60.1% -3.9% 1.28 1.49 -0.21 ** Someone else's vehicle 0.063 ** 0.019 36.4% 63.9% -27.5% 0.59 1.42 -0.83 ** Public transportation 0.070 ** 0.040 49.0% 59.9% -10.9% 0.70 1.14 -0.44 Other 0.056 ** 0.024 30.7% 26.6% 4.1% 0.82 1.15 -0.33 Have vehicle one can drive Yes 0.681 ** 0.930 59.6% 61.0% -1.4% 1.40 1.51 -0.11 If employed 0.876 ** 0.944 57.7% 57.1% 0.6% 1.64 1.50 0.14 If not employed, age 18-6 0.624 ** 0.899 47.7% 56.7% -9.0% 1.31 1.27 0.04 If not employed, age 65+ 0.555 ** 0.893 75.4% 95.0% -19.6% ** 1.15 1.97 -0.82 ** No 0.319 ** 0.070 39.9% 39.5% 0.4% 0.64 1.03 -0.39 Difficulty in travel Never or rarely 0.660 0.633 56.5% 59.5% -3.0% 1.12 1.51 -0.39 ** Occasionally 0.170 * 0.213 47.8% 62.4% -14.6% ** 1.28 1.33 -0.05 Often/very often 0.096 0.119 56.2% 58.3% -2.1% 1.43 1.43 0.00 Always 0.048 0.033 31.5% 45,5% -14.0% 1.04 1.69 -0.65 * Statistically significant difference at p<.10 ** p<.05 TABLE 15: Group Involvement, Religious Attendance, and Participation Distribution within Participation rates if in group at left samples (proportions) Voter turnout Other forms With Without With Without With Without Disability Disability Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (6) Overall 1.000 1.000 53.4% 59.5% -6.1% ** 1.15 1.47 -0.32 ** Regularly meet with any groups or organizations Yes 0.413 ** 0.546 63.9% 65.3% -1.4% 1.86 1.97 -0.11 1/month 0.154 0.179 74.1% 66.0% 8.1% 1.87 2.07 -0.20 2-3/month 0.097 ** 0.138 61.4% 67.6% -6.2% 2.32 2.12 0.20 4+/month 0.143 ** 0.219 54.3% 63.8% -9.5% 1.59 1.81 -0.22 No 0.587 ** 0.454 45.7% 52.4% -6.7% 0.67 0.87 -0.20 * Participate in disability groups or organizations 0.053 44.5% 2.33 Ever attend religious services Yes 0.590 ** 0.700 58.9% 65.5% -6.6% * 1.32 1.56 -0.24 * Every week or almost every week 0.377 * 0.432 65.2% 68.2% -3.0% 1.36 1.67 -0.31 ** Once or twice per month 0.115 0.138 57.4% 55.4% 2.0% 1.42 1.65 -0.23 Few times a year 0.097 * 0.131 37.8% 67.6% -29.8% ** 1.04 1.07 -0.03 No 0.411 ** 0.300 45.1% 45.0% 0.1% 0.92 1.29 -0.37 ** Think of self as part of church/denom. 0.230 ** 0.135 39.5% 40.0% -0.5% 0.93 1.29 -0.36 Don't think of self as part of church/denom. 0.181 0.165 52.7% 49.1% 3.6% 0.91 1.29 -0.38 * Difference between disability and non-disability samples is significant at the 90% level ** 95% level TABLE 16: Household Income and Political Participation Distribution within Participation rates if in group at left samples (proportions) Voter turnout Number of other forms With Without With Without With Without Disability Disability Disability Disability Difference Disability Disability Difference (1) (2) (3) (4) (5) (6) (7) (8) All respondents Less than $25,000 0.437 ** 0.148 43.9% 49.9% -6.0% 0.66 0.95 -0.29 * $25-50,000 0.236 ** 0.289 57.4% 49.4% 8.0% 1.57 1.19 0.38 * $50,000 or more 0.164 ** 0.426 66.4% 66.0% 0.4% 1.97 1.84 0.13 Don't know or refused 0.163 0.136 60.4% 70.8% -10.4% 1.08 1.51 -0.43 * Less than $10,000 0.143 ** 0.027 36.7% 59.6% -22.9% 0.65 0.81 -0.16 $10,000-$19,999 0.197 ** 0.064 50.4% 49.7% 0.7% 0.75 0.70 0.05 $20,000-$29,999 0.123 0.105 48.2% 45.5% 2.7% 1.00 1.28 -0.28 $30,000-$39,999 0.088 ** 0.146 49.7% 44.4% 5.3% 1.19 1.31 -0.12 $40,000-$49,999 0.060 0.075 65.4% 55.1% 10.3% 1.81 0.92 0.89 ** $50,000-$59,999 0.039 ** 0.099 66.2% 63.2% 3.0% 1.88 1.41 0.47 $60,000-$74,999 0.038 ** 0.123 70.2% 63.1% 7.1% 2.40 1.56 0.84 $75,000 or more 0.076 ** 0.184 61.2% 68.1% -6.9% 1.80 2.23 -0.43 Employed Less than $25,000 0.249 ** 0.120 39.7% 37.0% 2.7% 0.54 1.01 -0.47 * $25-50,000 0.330 0.302 54.2% 44.0% 10.2% 1.49 1.22 0.27 $50,000 or more 0.269 ** 0.463 62.5% 64.1% -1.6% 2.55 1.80 0.75 ** Don't know or refused 0.151 0.115 67.8% 69.2% -1.4% 1.36 1.44 -0.08 Not employed, age 18-64 Less than $25,000 0.512 ** 0.198 35.2% 49.0% -13.8% 0.73 0.74 -0.01 $25-50,000 0.206 0.239 43.3% 44.6% -1.3% 1.54 0.79 0.75 ** $50,000 or more 0.158 ** 0.385 64.6% 61.6% 3.0% 1.26 1.84 -0.58 Don't know or refused 0.124 0.177 53.3% 62.0% -8.7% 1.12 0.97 0.15 Not employed, age 65+ Less than $25,000 0.543 ** 0.244 53.7% 92.7% -39.0% ** 0.64 1.03 -0.39 $25-50,000 0.180 0.296 78.5% 92.5% -14.0% 1.71 1.55 0.16 $50,000 or more 0.073 ** 0.250 83.3% 100.0% -16.7% * 1.53 2.34 -0.81 Don't know or refused 0.204 0.211 58.4% 89.4% -31.0% ** 0.86 2.55 -1.69 ** * Statistically significant difference at p<.10 ** p<.05 TABLE 17: Prediction of Voting with All Characteristics Probits^ predicting probability of voting in 1998. Indep. variables (1) (2) (3) (4) (5) Any disability -0.194 (4.87) ** -0.146 (3.01) ** Political efficacy (high score=disagree) Public officials don't care 0.032 (2.02) ** 0.032 (2.01) ** 0.032 (2.02) ** People like me don't have say -0.001 (0.04) -0.001 (0.03) 0.000 (0.02) Politics too complicated 0.010 (0.67) 0.012 (0.82) 0.012 (0.80) Well-qualified for politics -0.018 (1.17) -0.018 (1.15) -0.018 (1.14) Better-informed than most -0.010 (0.63) -0.008 (0.55) -0.008 (0.54) Mobilization Contacted by pol. party in campaign 0.105 (1.99) ** 0.102 (1.92) * 0.104 (1.96) ** Contacted by someone else in campaign 0.031 (0.37) 0.036 (0.43) 0.036 (0.43) Encouraged to register or vote 0.073 (1.23) 0.071 (1.19) 0.070 (1.17) Transportation Days of travel per week -0.002 (0.13) 0.055 (0.90) 0.061 (0.97) Have vehicle one can drive 0.147 (2.02) ** 0.118 (1.56) 0.092 (1.10) Frequency of transportation problems 0.010 (0.56) 0.011 (0.63) 0.011 (0.58) Employment and social activities Employed -0.027 (0.42) -0.020 (0.31) -0.018 (0.27) Union member 0.009 (0.11) 0.013 (0.17) 0.013 (0.17) Meet regularly with groups/orgs. 0.051 (1.07) 0.049 (1.01) 0.040 (0.81) Ever attend religious services 0.084 (1.59) 0.078 (1.47) 0.075 (1.41) Household income Household income $25,000-$50,000 -0.065 (0.99) -0.067 (1.01) -0.072 (1.07) Household income $50,000+ 0.030 (0.42) 0.029 (0.40) 0.027 (0.38) Household income not reported 0.096 (1.18) 0.087 (1.06) 0.084 (1.01) Disability breakdowns Difficulty remembering/concentrating -0.107 (1.96) ** -0.104 (1.63) -0.112 (1.71) * Severe vision impairment -0.150 (1.70) * -0.105 (1.06) -0.101 (0.98) Cannot walk 1/4 mile, no mobility devices -0.158 (1.69) * -0.142 (1.40) -0.144 (1.37) Mobility devices or difficulty walking 1/4 mile -0.101 (1.89) * -0.089 (1.48) -0.092 (1.52) Difficulty going outside alone -0.322 (5.38) ** -0.286 (3.78) ** 0.087 (0.58) Other type of disability -0.052 (1.02) -0.041 (0.72) -0.048 (0.82) Difficulty going outside and: No regular meetings with groups/orgs. -0.215 (1.60) No vehicle one can drive -0.172 (1.17) Take trips less than 5 days/week -0.203 (1.29) Log-likelihood -669.58 -553.22 -662.48 -548.50 -547.04 n 1171 999 1171 999 999 * p<.10 ** p<.05 A Figures represent probability effects (dF/dX) of one-unit change at means of other variables. All regressions include demographic variables. FIGURE 1: Age Patterns in Voter Turnout, November 1998 90.0% 80.0% 70.0% Citizens.without 60.0% Disabilities Probability of voting 50.0% Citizens with Disabilities 40.0% 30.0% 20.0% 10.0% 0.0% 81 30 Age