Survey: Voting Intentions During the 1988 Election: A Comparison of Disabled and Non-disabled Voters

Extent (Dublin Core)

41 Pages

File Name (Dublin Core)

c021_002_004_001

Title (Dublin Core)

Survey: Voting Intentions During the 1988 Election: A Comparison of Disabled and Non-disabled Voters

Description (Dublin Core)

This project was for the National Organization on Disability and was directed by Louis Genevie. There is a bound copy of the survey in this folder.

Date (Dublin Core)

1988

Date Created (Dublin Core)

1988

Congress (Dublin Core)

100th (1987-1989)

Topics (Dublin Core)

See all items with this value People with disabilities

See all items with this value Presidents--United States--Election--1988

See all items with this value Voting research

Policy Area (Curation)

Government Operations and Politics

Creator (Dublin Core)

Louis Harris and Associates, Inc.

Record Type (Dublin Core)

reports

Names (Dublin Core)

See all items with this value Bush, George, 1924-2018

See all items with this value Dukakis, Michael S. (Michael Stanley), 1933-

See all items with this value National Organization on Disability (U.S.)

Rights (Dublin Core)

http://rightsstatements.org/vocab/CNE/1.0/

Language (Dublin Core)

eng

Collection Finding Aid (Dublin Core)

https://dolearchivecollections.ku.edu/?p=collections/findingaid&id=54&q=

Physical Location (Dublin Core)

Collection 021, Box 2, Folder 4

Institution (Dublin Core)

Robert J. Dole Institute of Politics, University of Kansas, Lawrence, KS

Archival Collection (Dublin Core)

Alec Vachon Papers, 1969-2006

Full Text (Extract Text)

TABLE OF CONTENTS
Introduction... 1
A Note on Reading the Tables ...1
Public Release of Survey Findings ...2

EXECUTIVE SUMMARY...3
Why George Bush Won The Election...3
Changes in Voting Intentions During the Campaign...4

ANALYSIS OF THE RESULTS OF THE SURVEY...6
Section 1: Who is the Disabled Voter?...7

Section 2: Party Affiliation...8

Section 3: Voting Intentions During the Primaries...8

Section 4: Changes After the Democratic National Convention ...9

Section 5: Changes After the Republican National Convention ...9

Section 6: Who Voters Think Will Do The Best Job on Key Issues Facing the Nation ...10
Economic Policy...10
Domestic Policy...12
Foreign Policy and Defense...13
Inspiring Confidence ... 13
Leadership: The Bottom Line ...14

METHODOLOGY... 27
Introduction...28
Random Digit Dialing... 28
Sample Description...29
Stratification of Telephone Sample ...30
Sample Reliability... 31
Respondent Selection... 31
Callback Strategy...32
Weighting...33
Sampling Error... 33
Significance of Difference Between Proportions ... 34

LOUIS HARRIS AND ASSOCIATES

INDEX OF TABLES

1 DISTRIBUTION OF DISABILITY AMONG VOTERS ...15

2 A COMPARISON OF DISABLED AND NON-DISABLED VOTERS ON KEY DEMOGRAPHIC VARIABLES ... 16

3 PARTY AFFILIATION OF DISABLED AND NON-DISABLED VOTERS ... 17

4 VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS AT FOUR POINTS IN TIME DURING THE PRIMARY SEASON ...18

5 VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS BEFORE AND AFTER THE DEMOCRATIC AND REPUBLICAN CONVENTIONS. 19

6 VOTING INTENTIONS OF DISABLED VOTERS BEFORE AND AFTER THE REPUBLICAN CONVENTION ...20

7 WHO VOTERS THINK WILL DO THE BEST JOB ON KEY ECONOMIC POLICY ISSUES... 21

8 WHO VOTERS THINK WILL DO THE BEST JOB ON KEY DOMESTIC POLICY ISSUES ... 22

9 WHO VOTERS THINK WILL DO THE BEST JOB ON KEY FOREIGN POLICY AND DEFENSE ISSUES ... 23

10 WHO VOTERS THINK WILL DO THE BEST JOB ON INSPIRING CONFIDENCE FROM THE PEOPLE ...24

11 LEADERSHIP RATINGS OF BUSH ... 25

12 LEADERSHIP RATINGS OF DUKAKIS ...26
METHODOLOGY
A-1 RECOMMENDED ALLOWANCE FOR SAMPLING ERROR OF PROPORTIONS {PLUS OR MINUS) ...34
A-2 SAMPLING ERROR OF DIFFERENCE BETWEEN PROPORTIONS ... 35 A-3 SURVEY FIELD DATES...38

LOUIS HARRIS AND ASSOCIATES

-1-

Introduction

This report focuses on voting intentions of disabled Americans as they emerged during the 1988 presidential election. The findings are based on a series of polls (1) conducted during the campaign by Louis Harris and Associates for the National Organization on Disability. In all, fourteen nationally representative telephone surveys of voters' intentions were conducted between February and early November, 1988. (2)

A total of 1,538 disabled voters and 19,440 non-disabled voters were identified and interviewed during this period. The field dates for each of the fourteen surveys summarized in this document are identified in Appendix A, which also contains a complete description of the methods employed for Harris telephone surveys.

Figures for age, sex, race and education were weighted where necessary to bring them into line with their actual proportions in the population. In samples of this size, one can say with 95 percent certainty that the results have a statistical precision of plus or minus three percentage points of what they would be if the entire adult population had been polled.

A Note on Reading the Tables

An asterisk (*) on a table signifies a value of less than one-half percent (0.5%). A dash (-) represents a value of zero. Percentages may not always add to 100% because of computer rounding, multiple answers from

(1) copies of the questionnaires used during the course of the campaign are available from Louis Harris and Associates.

(2) The screening questions used for the surveys can be found in Appendix A.

LOUIS HARRIS AND ASSOCIATES

-2-

respondents, or the elimination of "no answers" from particular tables. Voters who were unsure of their intentions at the time of the survey, for example, were not included in the tables.

Public Release of Survey Findings

All Louis Harris and Associates surveys are designed to adhere to the code of standards of the Council of American Survey Research Organizations (CASRO) and the code of the National Council of Public Polls (NCPP). Because data from this survey will be released to the public, any release must stipulate that the complete report is also available, rather than simply an excerpt from the survey findings.

LOUIS HARRIS AND ASSOCIATES

-3-

EXECUTIVE SUMMARY

WHY GEORGE BUSH WON THE ELECTION
1. George Bush will be the 41st president of the United States because he convinced voters -- disabled and non-disabled alike -- that he will do a better job than Michael Dukakis on most of the key issues affecting the economy, foreign policy, and defense.

- Voters think that Bush will do a better job keeping the economy prosperous (54% vs. 38%), making America more competitive abroad (54% vs. 37%), negotiating arms agreements with the Russians (59% vs. 33%) and maintaining a strong defense for the country (63% vs. 30%).

- During the course of the campaign, Bush's strength on most of these issues increased more among disabled voters than among voters generally.

2. Most voters also think that Vice President Bush will do a better job controlling crime and handling the drug problem in the country.

- Bush's support on handling the drug problem increased from 30% immediately after the Democratic national convention to 47% in late October, an increase that was reflected in both the disabled and non-disabled vote.

3. As a result of the change in voters' opinions on these key issues, the Vice President's leadership ratings rose substantially during the course of the campaign.

- Forty-seven percent of voters rated the Vice President's leadership ability positively in July, and by November, 62% did so.

4. On which candidate had the greatest ability to inspire confidence
in the people, Bush beat Dukakis 55% to 40% among all voters.

- Among disabled voters, Bush's ability to inspire confidence was even greater, 54% to 36% for Dukakis.

LOUIS HARRIS AND ASSOCIATES

-4-

- The Vice President's rise in the polls on this issue during the course of the campaign was dramatic, moving from 30% in July to 55% in late October. During the same time, Dukakis's ability to inspire confidence fell by 23 points.

5. On issues affecting the more vulnerable segments of the society -- children, the poor, the elderly and the disabled -- voters, both disabled and non-disabled, think that Governor Dukakis would do a better job.

- Dukakis's lead on these issues was still strong going into the last week of the campaign, but declined slightly after the Democratic convention.

- Bush, while still trailing Dukakis, increased his support on these issues during the course of the campaign.

6. Voters also think that Dukakis would do a better job protecting the environment and improving the quality of education.

- Dukakis's lead on these issues declined slightly during the campaign, while Bush, although still trailing at the end, picked up support.

CHANGES IN VOTING INTENTIONS DURING THE CAMPAIGN

7. As leadership on the key issues changed during the course of the
campaign, so too did voters' intentions.

- The Harris poll conducted in March gave Governor Dukakis a two point lead over the Vice President among all voters. In the three months that followed, he increased that lead to five points.

8. Dukakis's lead over Bush during the primary season was much larger among disabled voters than among voters in general.
- Among disabled voters Dukakis started in March with an 11 point lead (55% vs. 44%) and by June had increased that lead to 24 points (58% vs. 34%).

- Thus, as the Democratic convention approached, Dukakis's overall lead stood at three points and was, to a significant extent, based on the very strong support he was receiving from disabled voters.

LOUIS HARRIS AND ASSOCIATES

-5-

9. After the Democratic convention, Michael Dukakis's lead increased among both disabled and non-disabled voters.

- Among all voters, Dukakis's lead increased ten points after the convention (54% for Dukakis; 42% for Bush).

- Among the disabled, the Governor's lead of 21 points going into the Democratic convention increased to 33 points afterward.

10. Our poll after the Republican national convention showed that Dukakis's substantial lead declined sharply among both disabled and non-disabled voters.

- Dukakis's 12 point lead among all voters after the Democratic convention turned into a five point deficit after the Republican convention when Vice President Bush led 50% to 45%.

- Among disabled people, one of Dukakis's strongholds, his lead deteriorated from 33 points after the Democratic convention to only ten points after the Republican convention.

11. By mid-October the Vice President had closed the gap and two of our last three polls showed Bush ahead of Dukakis among disabled people.

- Our final poll put Dukakis ahead of Bush among the disabled (44% vs. 49%) voters. Without question, however, the Vice President made substantial inroads into the disabled vote. In the end, the change in the disabled vote during the campaign accounted for between one and three percent of the Vice President's margin of victory.

LOUIS HARRIS AND ASSOCIATES

-6-

ANALYSIS OF THE RESULTS OF THE SURVEYS

LOUIS HARRIS AND ASSOCIATES

-7

Section 1: Who is the Disabled Voter?

As part of the Harris firm's regular election year polling of probable voters, respondents were asked whether or not they had a disability, handicap, or other physical or mental limitation that keeps them from participating fully in work, school, housework, or other activities. Overall, 10% were identified as having some form of disability (Table 1).(1)

In addition to their disability, the disabled voter also faces personal and social barriers to full participation in the political process. (2) Disabled voters tend to have less education and less income than other Americans. Forty-six percent have less than a high school education, compared to only 16% of non-disabled voters. Far fewer possess four year college degrees (9%) when compared to non-disabled voters (25%). Four times as many disabled voters have household incomes of $7,500 a year or less as compared to non-disabled voters (24% vs. 6%). Given this, it is not surprising to find that disabled people are more than two and a half times as likely to be out of work as the non-disabled. Even among those who do work, only one third as many work as professionals when compared to non-disabled people (Table 2).

Disabled voters are also older than their non-disabled counterparts. Forty-one percent are above the age of 65 compared to only 17% of non-disabled
voters (Table 2).

(1) This figure should not be confused with the number of disabled people in the country. Depending on how disability is measured, this figure stands at about 15 percent of the total adult population. See The ICD Survey of Disabled Americans: Bringing Disabled Americans into the Mainstream, Louis Harris and Associates, 1986.

(2) see Participation in Voting and Elections By Disabled Americans, Louis Harris and Associates, 1987.

LOUIS HARRIS AND ASSOCIATES

-8-

Section 2: Party Affiliation

Given the fact that the disabled voter tends to be older, less educated, and not as wealthy as the non-disabled voter traditional constituencies of the Democratic party --it is not surprising to find that more disabled people say they are Democrats than Republicans.

- Fifty-two percent of the disabled are registered as Democrats, compared to 40% of non-disabled voters. Twenty-four percent of the disabled say they are Republicans, compared to 33% of non-disabled voters. And 21% of the disabled voters say they belong to neither major party, compared to 25% of the non-disabled voters (Table 3).

Section 3: Voting Intentions During the Primaries

During the primary season, prior to the nominating conventions of both parties, the voting intentions of both disabled and non-disabled people were relatively stable (Table 4). During that time, Governor Dukakis started out with a two point lead over the Vice President, and in the three months that followed increased his lead to five points (49% vs. 44%).

Dukakis's lead in the primaries was much slimmer among non-disabled voters than among those who identified themselves as disabled. Beginning in March, Dukakis held a one point lead among non-disabled voters and by June his lead had increased to three points. Among disabled voters, however, Dukakis started out in March with an eleven point lead (55% vs. 44%) and by June had increased that lead to 24 points (58% vs. 34%).

As the Democratic convention approached, Dukakis's overall lead stood at three points (Table 5). His lead, however, was, to a large extent, based on the very strong support he was receiving from disabled voters, 58% of whom supported him compared to 37% who supported the Vice President. Among non-disabled voters Dukakis held a slim one point lead prior to the Democratic convention.

LOUIS HARRIS AND ASSOCIATES

-9-

Section 4: Changes After the Democratic National Convention

Dukakis received a major boost in support as a result of the Democratic nominating convention. Our two polls taken after the convention showed that the slim lead that Dukakis held prior to the convention increased from three percent to twelve percent. During this time, Dukakis's lead increased among both disabled and non-disabled voters, but was more substantial among the disabled.

- The Governor's lead of twenty-one points among the disabled prior to the Democratic convention increased to thirty-three points afterward. Among non-disabled voters the Governor's slim one point lead increased to ten points after the convention (Table 5).

Section 5: Changes After the Republican National Convention

Between the beginning of August and the beginning of September, 1988, the substantial lead that Dukakis held after the national Democratic convention deteriorated rapidly among both disabled and non-disabled voters. Overall, Dukakis's twelve point lead after the Democratic convention turned into a five point deficit after the Republican convention, when Vice President Bush held the lead over the Governor 50% to 45%. Even among disabled people, one of Dukakis's strongholds, his lead deteriorated from thirty-three points after the Democratic convention to only ten points after the Republican convention.

Exactly when disabled voters' intentions changed during this period can be seen in the three Harris polls taken in August and September (Table 6). In early August, prior to the Republican convention, Dukakis held a 30 point lead over the Vice President among the disabled. In the next poll, taken in late August, just after the Republican national convention, a dramatic shift in the voting preference of disabled people was found. At that time, 49%, an

LOUIS HARRIS AND ASSOCIATES

-10-

increase of 16% compared to just a few weeks prior, said that they were for Bush, which placed the Vice President two points ahead of Governor Dukakis (49% for Bush, 47% for Dukakis) among the disabled.

In two of the three polls that followed, George Bush led Michael Dukakis among the disabled and non-disabled alike. Our final poll, however, placed Bush ahead of Dukakis 52% to 45% among all voters, but trailing Dukakis among the disabled, 44% to 49% (Table 6).

Observation:
There is no doubt that the Vice President made inroads into this traditionally Democratic constituency. Dukakis's substantial lead among the disabled dissipated during the course of the election and disabled people, despite their traditionally Democratic leanings, voted for George Bush in numbers almost equal to that of the nation as a whole.

Section 6: Who Voters Think Will Do The Best Job on Rey Issues Facing the Nation

In addition to polling voters' intentions during the course of the campaign, the Harris firm also asked voters which candidate they thought would do the best job on a variety of problems facing the nation, including key domestic and foreign policy issues.

Economic Policy

On the issue of which candidate would be more likely to keep the country prosperous, the latest Harris poll* on this issue placed George Bush ahead of Michael Dukakis 54% to 38% (Table 7). Disabled voters thought that George Bush could do a better job in this regard by a similar margin. The

*October 13-17, 1988.

LOUIS HARRIS AND ASSOCIATES

-11-

disabled vote, however, changed more dramatically than the total national vote. Immediately after the Democratic national convention (late July), most voters felt that Michael Dukakis would do better keeping the economy prosperous (53% vs. 46% for Vice President Bush). Among the disabled, however, Dukakis led Bush, 67% to 24%. Between late July and late October the number of disabled voters who thought that Bush would do a better job with the economy increased 34 points from 24% to 53%. During that same time, the proportion of all voters who thought that Bush would do a better job with the economy increased by eight points, more than one and one-half of which were attributable to the change in disabled people's opinion.

On the issue of making America more competitive abroad, the majority of voters thought that George Bush would do a better job (53% of the disabled voters; 54% of all voters). These latest figures represented strong increases for Bush during the course of the campaign. Since late July, the proportion of all voters who thought that the Vice President would make America more competitive abroad increased 15 points; and the proportion of disabled voters who thought so increased by 23 points (Table 7).

Voters also thought that George Bush would do better than Michael Dukakis at controlling inflation. On this issue, Bush held a 51% to 39% lead overall, but the disabled vote went for Dukakis, 48% to 44%.

On their ability to cut the rate of unemployment in the country, Bush and Dukakis were in a virtual dead heat, 46% for Bush, 45% for Dukakis. Even though disabled people were going for Dukakis on this issue 49% to 38%, Vice President Bush picked up six points among disabled voters on this issue between May and October, while the disabled vote for Dukakis remained at 49%. Among all voters Vice President Bush picked up eight points between May and October, while Governor Dukakis lost three points.

LOUIS HARRIS AND ASSOCIATES

-12-

Voters were concerned about the Vice President's ability to cut the deficit and thought that Dukakis would do a better job cutting Federal spending (46% vs. 43% for Bush). That margin is wider among disabled voters, 51% for Dukakis, 40% for Bush.

As far as cutting the federal deficit is concerned, the electorate also felt that Dukakis would do a better job. However, his margin was very slim (45% to 44%). Disabled voters preferred George Bush, 43% to 36%, on this issue (Table 7).

Domestic Policy

By substantial margins, voters thought that Governor Dukakis would do a better job protecting the environment and improving the quality of education and health care in the country. Both disabled and non-disabled voters also thought that Dukakis would do a better job protecting the more vulnerable segments of society, including disabled people, children, the elderly and the poor. For example, voters -- disabled and non-disabled, thought that Dukakis would do a better job helping the disabled, (57% vs. 34% for Bush among all voters, and 55% vs. 33% among disabled voters) (Table 8).

While Dukakis continued to hold a substantial lead on these issues of conscience, in most instances, Vice President Bush made substantial gains during the course of the campaign among disabled and non-disabled voters alike. The proportion supporting Dukakis, on the other hand, either remained stable or declined slightly. The candidate voters thought would do the best job improving the quality of public education is a good example of this trend. When we took our last poll before the election, Dukakis led Bush on this issue 51% to 38% nationally, figures that are similar for the disabled vote. Since May, however, Bush increased his support among voters on this issue from 29% to 38%, while

LOUIS HARRIS AND ASSOCIATES

-13-

Dukakis's support declined slightly from 54% to 51%. This trend was even stronger among the disabled than among the voters generally. Michael Dukakis was still thought by most voters to be able to do the best job dealing with these problems, but George Bush made substantial gains in convincing voters that he could do the job better (Table 8).

Foreign Policy and Defense

The majority of voters thought that George Bush would do a better job maintaining a strong defense for the country (63% vs. 30%). However, by a similar margin (59% vs. 34%), voters thought that Michael Dukakis would do a better job controlling defense spending. The figures were similar for disabled voters on these two issues.

As far as working for world peace and negotiating nuclear arms control agreements with the Russians were concerned, voters -- both disabled and non-disabled favored George Bush by wide margins.

On bringing peace to Central America, however, Bush's margin was not anywhere near as great -- 46% -- compared to 42% for Dukakis. Disabled voters favored Dukakis on this issue 47% to 42% (Table 9).

Inspiring Confidence

Bush led Dukakis 55% to 35% on the key issue of inspiring confidence in the people. Among disabled voters Bush's lead was even greater, 59% vs. 30%. The Vice President's increase on this key issue was dramatic. Since late July, when only 30% of those polled said they thought that he would do the best job inspiring confidence, the Vice President's support rose by fully 25 points, to 55%, while Governor Dukakis's fell by 28 points to only 35%. (Table 10).

LOUIS HARRIS AND ASSOCIATES

-14-

The shift was even greater among disabled voters than among voters in general. Bush increased the proportion of disabled voters who thought he would do the best job inspiring confidence by fully 31 points while Dukakis's support declined by 37 points during the course of the campaign (Table 10).

Leadership: The Bottom Line

Perhaps the most telling evaluation by voters was that made of the leadership potential of the candidates. In many ways, George Bush's ratings during the course of the campaign tell the story of the campaign in a nutshell. Prior to the Republican national convention, the Vice President's positive rating on leadership was below 50%. After his strong performance at the convention, Bush's rating soared and by the end of the campaign stood at 62% (Table 11). Meanwhile, Dukakis's leadership rating was moving a similar magnitude in the opposite direction (Table 12).

Observation:

Like most voters, disabled people think that George Bush will do the best job on most of the key issues facing the nation. This, despite the fact that disabled and non-disabled voters alike think that Dukakis would have done a better job improving the quality of education, protecting the environment, and helping the more vulnerable segments of the society -- including the disabled.

The fact that George Bush mentioned his support for the disabled in his acceptance speech at the Republican national convention should not be overlooked. During that speech Bush said, "I'm going to do whatever it takes to make sure the disabled are included in the mainstream. For too long they've been left out, but they're not going to be left out anymore." This was the first time a candidate for national office had addressed the nation's disabled citizens directly; on at least two other occasions before major national audiences, the Vice President repeated his pledge. Although Governor Dukakis also had a very strong position in support of the disabled, he made no clear statement of support during his acceptance speech or in any national forum afterward.

LOUIS HARRIS AND ASSOCIATES

-15-

Table 1

DISTRIBUTION OF DISABILITY AMONG VOTERS

Q.: Does a disability, handicap, or other physical or mental limitation currently keep you from participating fully in work, school, housework, or other activities, or not?

Total 13613

%
Yes, keeps me from participating 10 (%)
No, does not 89 (%)
Not sure 1 (%)

LOUIS HARRIS AND ASSOCIATES

-16-

Table 2
A COMPARISON OF DISABLED AND NON-DISABLED VOTERS ON KEY DEMOGRAPHIC VARIABLES

Total
Disabled Voters
Non-Disabled Voters
Base
13613
1124
12322

%
%
%

Sex

Male
49
47
49
Female
51
53
51

Occupation

Professional
14
5
16
Manager, official
12
4
12
Proprietor
5
3
5
Clerical worker
5
4
6
Sales worker
5
2
5
Craftsman, foreman
13
6
14
Operative, unskilled labor
10
8
10
All others
13
16
12
Not working
23
51
20

Age

18-24
10
3
11
25-34
20
5
21
35-44
20
10
21
45-64
31
39
30
65-74
12
24
11
75+
7
17
6

Education

Less than high school
19
46
16
High school graduate
36
31
36
Some college
21
13
22
4-year college graduate or more
23
9
25

Household Income

$7,500 or less
8
24
6
$7,501 to $15,000
12
23
11
$15,001 to $25,000
19
20
19
$25,001 to $35,000
19
11
20
$35,001 to $50,000
19
9
20
$50,001 or more
16
4
18

LOUIS HARRIS AND ASSOCIATES

-17-

Table 3

PARTY AFFILIATION OF DISABLED AND NON-DISABLED VOTERS

Total
Disabled Voters
Non-Disabled Voters
Base
13613
1124
12322

%
%
%
Party Affiliation

Republican
32
24
33
Democrat
41
52
40
Independent
25
21
25
Other/ not sure
2
2
2

LOUIS HARRIS AND ASSOCIATES

-18-

Table 4

VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS AT FOUR POINTS IN TIME
DURING THE PRIMARY SEASON*

Q.a: Of course, this November for president, it will be Vice President George Bush for the Republicans and Governor Michael Dukakis for the Democrats. If you had to choose right now, would you be for Bush or Dukakis?

Q.b: Well, if you had to say, would you lean toward Bush the Republican or toward Dukakis the Democrat?

Total Voters
Disabled Voters
Non-Disabled Voters

Base
Bush
Dukakis
Base
Bush
Dukakis
Base
Bush
Dukakis
March, 1988

%
%

%
%

%
%
(3/9-3/13)
1260
47
49
119
44
55
1135
48
49
April, 1988

(4/1-4/5)
1248
47
49
88
35
58
791
47
49
May, 1988

(5/5-5/10)
1256
43
50
107
37
57
1143
44
50
June, 1988

(6/1-6/6)
1246
44
49
108
34
58
1135
45
48
*Totals do not add to 100% due to voters who were not sure how they would vote.

LOUIS HARRIS AND ASSOCIATES

-19-

Table 5

VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS BEFORE AND AFTER THE DEMOCRATIC AND REPUBLICAN CONVENTIONS*

Q.a: Of course, this November for president, it will be Vice President George Bush for the Republicans and Governor Michael Dukakis for the Democrats. If you had to choose right now, would you be for Bush or Dukakis?

Q.b: Well, if you had to say, would you lean toward Bush the Republican or toward Dukakis the Democrat?

Total Voters Disabled Voters Non-Disabled Voters
Base Bush Dukakis Base Bush Dukakis Base Bush Dukakis
Before the
Democratic
Convention
(6/1-6/6 and 7/7-7/12) 2491 % 46 49 208 % 37 58 2243 % 47 48
After the
Democratic
Convention
(7/22-7/25and 8/4-8/9) 2662 % 42 54 222 % 31 64 2415 % 43 53
After the
RepublicanConvention
(8/19-8/22and 9/1-9/6) 2586 % 50 45 186 % 43 53 2381 % 50 45
Mid-Oct. (10/14-10/17) 1350 % 53 43 90 % 54 36 1256 % 53 43
Late Oct. (10/28-10/30) 1249 % 52 45 80 % so 48 1161 % 52 44
Early Nov. (11/2-11/7) 3216 % 52 45 244 % 44 49 2638 % 53 44

*Totals do not add to 100% due to voters who were not sure how they would vote.

LOUIS HARRIS AND ASSOCIATES

Table 6
VOTING INTENTIONS OF DISABLED VOTERS BEFORE AND AFTER THE REPUBLICAN CONVENTION*
Q.a: Of course, this November for president, it will be Vice President George
Bush for the Republicans and Governor Michael Dukakis for the Democrats. If you had to choose right now, would you be. for Bush or Dukakis?
Q.b: Well, if you had to say, would you lean toward Bush the Republican or toward Dukakis the Democrat?
Disabled Voters
Base** Bush Dukakis
Early August (8/4-8/9) 125 % 33 63
Late August (8/19-8/22) 88 % 49 47
Early Sept. (9/1-9/6) 99 % 37 59
Early Oct. (10/6-10/10) 103 % 47 46
Mid-Oct.
(10/14-10/17) 90 % 54 36
Late Oct.
(10/28-10/30) 80 % 50 48
Early Nov. (11/2-11/-7) 244 % 44 49

*Totals do not add to 100% due to voters who were not sure how they would vote.
**Even though these bases are small, the changes of 16% for Bush from early August to late August is still significant at the 95% confidence level.
LOUIS HARRIS AND ASSOCIATES

Table 7
WHO VOTERS THINK WILL DO THE BEST JOB ON KEY ECONOMIC POLICY ISSUES Total Voters Disabled Voters
Base* Bush Du-Neither/ kakis Not Sure- Base Du-Neither/ Bush kakis Not Sure
KeeEing the economy ErOSEerous MayEarly July Late July Late August Early October Mid-October 1257 % 44 1245 % 46 1266 % 46 1295 % 55 1354 % 52 1353 % 54 44 45 53 40 42 38 12 8 7 6 6 9 107 % 30 100 %. 36 98 24 88 % 44 103 % 45 90 % 53 52 52 67 43 48 33 18 12 10 13 8 14
Making America more competitive abroad
May Early JulyLate July Late August Early October Mid-October , 1257 % 39 1246 % 47 1266 % 39 1295 % 54 1355 % 48 1356 % 54 46 42 53 40 41 37 15 11 8 6 10 9 107 % 28 100 % 40 98 % 30 87 % 46 103 % 39 90 % 53 53 so 59 44 45 32 18 10 12 11 16 15
Cutting federal SEending MayEarly July Early October 1257 % 41 1234 % 41 1355 % 43 44 48 46 15 11 11 107 % 28 100 % 36 103 % 40 51 52 51 21 12 9
Handling the problem of the federal deficit
MayEarly JulyEarly October Mid-October 1257 % 38 1240 % 42 1354 % 42 1356 % 44 45 46 47 45 16 11 11 12 107 % 24 99 % 32 103 % 38 90 % 43 53 58 54 36 22 10 8 20
Keeping inflation under control
May Early October 1257 % 48 1354 % 51 38 39 14 10 107 % 31 103 % 44 48 48 20 8
Cutting the rate of unemElovment
MayEarly October 1256 % 38 1354 % 46 48 45 14 9 107 % 32 103 % 38 49 49 19 13

*Bases vary according to how many times the question was asked during the course of the campaign.
LOUIS HARRIS AND ASSOCIATES

-22Table 8
WHO VOTERS THINK WILL DO THE BEST JOB ON
KEY DOMESTIC POLICY ISSUES
Total Voters Disabled Voters Du-Neither/ Du-Neither/ Base* Bush kakis Not Sure Base Bush kakis Not Sure
Handling the drug problem
in the country
May 1257 % 33 41 26 107 % 27 47 26
Early July 1245 % 35 44 21 100 % 28 48 23
Late July 1266 % 30 53 17 98 % 28 55 18
Late August 1296 \ 48 41 12 88 \ 49 42 9
Early October 1355 \ 43 45 11 103 \ 43 46 11
Mid-October 1356 \ 47 40 13 90 % 43 32 25

Controllin9 crime Mid-October 1356 % 63 27 10 90 \ 53 34 13
HelEing the disabled Late August 1296 % 33 58 9 88 % 36 52 12 Early October 1355 % 34 57 9 103 \ 33 55 12
Protectin9 the environment Late August 1294 % 38 so 12 88 \ 35 47 18 Early October 1355 \ 41 49 9 103 \ 39 47 14 Mid-October 1356 \ 40 48 11 90 \ 43 41 17
Helping children get a better break Early July 1245 \ 27 56 16 100 \ 24 62 15 Early October 1355 % 34 56 9 103 \ 34 53 13 Mid-October 1356 \ 35 54 10 90 \ 39 51 10
Improving the quality of Eublic education May 1256 % 29 54 17 107 \ 24 58 17 Early July 1245 \ 32 54 13 100 \ 27 59 14 Early October 1355 \ 37 53 10 103 % 38 49 13 Mid-October 1355 \ 38 51 11 90 % 37 48 16
Helping the elderly and EOor get a better break May 1257 % 21 66 11 107 \ 24 60 15 Early October 1355 \ 31 62 7 103 \ 35 60 6
Making sure all workers are covered with health insurance provided by their e!!!Eloyer
May 1257 \ 22 64 14 107 \ 22 64 14 Early October 1355 % 24 66 10 103 \ 29 60 11 Mid-October 1355 \ 28 62 11 90 \ 35 55 10
*Bases vary according to how many times the question was asked during the course of the campaign.
LOUIS HARRIS AND ASSOCIATES
Table 9
WHO VOTERS THINK WILL DO THE BEST JOB ON
KEY FOREIGN POLICY AND DEFENSE ISSUES

Total VoteI"s Disabled VoteI"s Du-NeitheI"I Du-NeitheI"/ Base* Bush kakis Not Sure Base Bush kakis Not SUI"e
Working for peace in the WOI"ld Late August 1296 % 53 39 8 88 % 52 41 7 EaI"ly octobeI" 1355 % 49 42 9 103 % 45 51 4 Mid-OctobeI" 1356 % 53 37 11 90 % 57 28 15
Negotiating nuclear arms contI"ol agreements with the Russians May 1257 % 54 32 14 107 % 41 40 18 EaI"ly October 1355 % 59 33 8 103 % 52 35 12
BI"inging peace to Central Amedca Euly OctobeI" 1355 % 46 42 12 103 % 42 47 11
Maintaining a strong defense foI" the countI"y May 1257 % 62 28 10 107 % 41 41 18 Eady July 1245 % 60 31 8 100 % 56 33 11 Eady OctobeI" 1355 % 63 30 6 103 % 55 34 11
ContI"olling defense spendingMay 1256 % 31 57 12 107 % 29 55 16 EaI"ly OctobeI" 1354 % 34 59 7 103 % 38 53 8
*Bases vaI"y accoI"ding to how many times the question was asked duI"ing the couI"se of the campaign.
-24
Table 10
WHO VOTERS THINK WILL DO THE BEST JOB
ON INSPIRING CONFIDENCE FROM THE PEOPLE

Total Voters Disabled VotersDu-Neither/ Du-Neither/Base* Bush kakis Not Sure ~ Bush kakis Not Sure
Inspiring confidence fromthe 12eo12leMay 12SS % 38 49 13 107 % 31 SS 14 Early July 1246 % 36 S2 12 100 % 27 S8 14 Late July 1266 % 30 63 6 98 % 28 67 6 Late August 1296 % 46 49 6 88 % 4S so s Early October 13S3 % 48 44 7 103 % 4S 4S 10 Mid-October 13S3 % SS 3S 9 90 % S9 30 11 Late October 12SO % SS 40 5 80 % S4 36 10
*Bases vary according to how many times the question was asked during the course of the campaign.
LOUIS HARRIS AND ASSOCIATES
Table 11
LEADERSHIP RATINGS OF BUSH
Q.: How would you rate (READ EACH NAME) as a leader, on a scale of 1 to 10, with 1 being the worst and 10 the best?
Total Voters Disabled Voters Pas-Neg-Pos-Neg-
Base itive ative Not Sure Base itive ative Not Sure
Early July
(7/7-7/12) 1245 % 47 50 3 100 % 44 48 8

Late July
(7/22-7/25) 1266 % 47 so 3 98 % 39 56 5

Early Auqust
(8/4-8/9) 1401 % 49 49 2 125 % 38 56 6

Late Auqust
(8/19-8/22) 1295 % 58 40 2 88 % 59 38 3

Early Sept.
(9/1-9/6) 1299 % 60 37 3 99 % 45 44 11

Early Oct.
(10/6-10/10) 1354 % 60 38 2 103 % 49 42 9

Mid-Oct.
(10/14-10/17) 1356 % 56 42 2 90 % 58 39 3

Late Oct.
(10/28-10/30) 1250 % 60 38 2 80 % 64 33 3

Early Nov.
(11/2-11/7) 3221 % 62 36 2 244 % 56 40 4

LOUIS HARRIS AND ASSOCIATES

Table 12
LEADERSHIP RATINGS OF DUKAKIS
Q. : How would you rate (READ EACH NAME) as a leader, on a scale of 1 to 10, with 1 being the worst and 10 the best?
Total Voters Disabled Voters Pos-Neg-Pos-Neg-
Base itive ative Not Sure Base itive ative Not Sure
Early July
(7/7-7/12) 1245 % 55 38 7 100 % 55 36 9

Late July
(7/22-7/25) 1266 % 64 33 3 98 % 71 25 4

Early August
(8/4-8/9) 1401 % 57 38 5 125 % 67 28 5

Late August
(8/19-8/22) 1295 % 55 42 3 88 % 62 35 3

Early Sept.
(9/1-9/6) 1299 % 55 41 4 99 % 54 32 14

Early Oct.
(10/6-10/10) 1354 % 55 42 3 103 % 53 40 7

Mid-Oct.
(10/14-10/17) 1356 % 49 47 4 90 % 42 52 6

Late Oct.
(10/28-10/30) 1250 % 53 44 3 80 % 48 48 4

Early Nov.
(11/2-11/7) 3221 % 51 46 3 .244 % 56 40 4

LOUIS HARRIS AND ASSOCIATES

-27
APPENDIX A: METHODOLOGY

LOUIS HARRIS AND ASSOCIATES

METHODOLOGY
I. Introduction
Louis Harris and Associates' telephone surveys are based on a national sample of the civilian population 18 years of age and older of the United States. Those living in Alaska and Hawaii are not represented, nor are those in prisons, hospitals, or religious and educational institutions, unless such individuals have their own outside phone line. Samples are based on the Census Bureau's adult population figures for each state in the country. These figures are updated by intercensal estimates produced annually by the Census Bureau, and sample locations are selected biannually to reflect changes in the country's demographic and geographic profile.
National samples are stratified on two dimensions --geographic region and metropolitan vs. nonmetropolitan residence. Stratification ensures that the sample will reflect within 1% those living in different regions of the country and those living in SMSAs and non-SMSAs. Within each stratum, the selection of the primary sampling unit (PSU) is achieved through multistage unclustered sampling. Each PSU yields one interview. First states, ·then counties, and then minor civil divisons are selected, with probability of selection proportionate to Census Bureau estimates of their respective populations.
II. Random Digit Dialing
Louis Harris and Associates employs a random digit dialing procedure to select households within primary sample units. Random digit dialing is a significant improvement over previously used techniques, extending the potential coverage rate to almost 95% of the U.S. population.
Unless some method of random digit dialing is used, telephone samples must be drawn directly from published lists. However, since the population of
unlisted phone number subscribers is large and demographically dissimilar to
subscribers with listed phone numbers, reliance on telephone directories alone
yields seriously biased samples. For this reason, using published phone listing as the universe is inadequate for telephone surveys and inferior to using random digit dailing.
The use of a random digit dialing method offers other important
advantages:
The sample is highly representative.
Unlisted telephone numbers have the same probability of inclusion in the sample as listed numbers. This is particularly important in reaching both high income and minority populations.
Respondents are geographically dispersed, rather than clustered.
Households with multiple phone listing have the same probability of inclusion in the sample as households with a single phone listing.
III. Sample Description The Louis Harris and Associates national telephone sample is selected via a multistage, stratified selection process.
The first stage of sampling involves the selection of 1,000 county sampling points (or primary selections). These sampling points are selected within 144 strata defined on the basis of the cross classification of individual
states and metropolitan/non-metropolitan areas. The selection of these points is based upon probabilities proportional to population estimates, using updates
from the 1980 Census. The second stage of sampling involves the selection of a specific telephone directory for each of the 1,000 primary selections.
LOUIS HARRIS AND ASSOCIATES

-30
This is accomplished in several steps. First, all telephone
directories for counties containing primary selections are obtained by the sampling office. From census definitions, a unique list of all separate
geographic listings is obtained. In this process duplicate geographic listings which appear in multiple directories are eliminated. Next, these geographic
listings are stratified by size. Finally, using the initial random numbers used to select the primary selection, a relative within primary unit number is used to obtain a specific geographic listing for each primary selection. In the third step of sampling, each primary selection (which is now located within a telephone list) is followed down to a specific telephone number
consisting of an area code, a 3 place prefix and a 4 place suffix. Deleting the last two digits of this suffix defines a bank of 100 telephone numbers.
For each selected bank, a series of random two place numbers are generated. Dialing is then sequentially initiated. If the number is not a household, or if it does not result in a successful screen/interview, a
replacement is selected by going to the next complete phone number. This process is repeated until a successful screen/interview is completed, or the
field period is closed.
IV. Stratification of Telephone Sample The core telephone exchanges are stratified by two variables -geographic region and place of residence. The United States is divided into
four regions, as follows:
1. East: Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, Connecticut, New York, New Jersey, Pennsylvania, Delaware, Maryland, District of Columbia, and West Virginia.
2. South: Virginia, North Carolina, South Carolina,
Florida, Georgia, Kentucky, Tennessee, Alabama, Mississippi,Arkansas, Louisiana, Texas, and Oklahoma.
LOUIS HARRIS AND ASSOCIATES
3.
Midwest: Ohio, Michigan, Indiana, Illinois, Wisconsin, Minnesota, Iowa, Missouri, Kansas, Nebraska, South Dakota, and North Dakota.

4.
West: Montana, Wyoming, Colorado, New Mexico, Arizona, Utah, Idaho, Nevada, California, Oregon, and Washington.

We also use three categories for place of residence:
1.
Central City: Every place defined by the Census Bureau as a central city of a Standard metropolitan Statistical Area. (SMSA).

2.
Rest of Metropolitan Area: Every place that is not a central city but is within SMSA boundaries.

3.
Outside Metropolitan Area: Every place that is not included in any of the other two categories.

We then define each place or residence category within each region as a stratum.
V. Sample Reliability
In order to maintain reliability and integrity in the sample, the
telephone field staff follows these procedures when a respondent contact is
attempted: A nonanswering telephone is dialed three times over a three-day period. At the end of this time, if no contact is made, a new telephone number is generated for that PSU.
If a business telephone is reached or if contact is made with a household in which a potential respondent presents a language barrier, or a telephone is not in service, a new telephone number is generated for that PSU.
Once a residential contact is established, the interviewer uses a respondent selection procedure to designate someone in the household for the interview.
VI. Respondent Selection
Within each household contacted for the survey, the interviewer uses a
respondent selection procedure to determine which adult member of the household
will be interviewed. Under the standardized procedure, the interviewer asks for
one of the following household members, in order of priority: (1) youngest
LOUIS HARRIS AND ASSOCIATES

adult male at home; (2) next youngest adult male at home; (3) youngest adult female-at home; or (4) next youngest adult female at home. This procedure results in an acceptable distribution of respondents' sex and age.
If a respondent selection grid is required to select a respondent in a household, the household member with the nearest birthday will be selected as the eligible respondent. This procedure eliminates the complexities involved with the Troldahl-Carter grid while increasing respondents' willingness to participate in the survey. Use of this method results in an acceptable distribution of respondents' sex and age.
VII. Callback Strategy In order to attain the highest possible response rates within reasonable cost constraints, callbacks are made according to the following guidelines: No Answer/Not-at-Homes: An initial attempt and then three callbacks to reach an adult member of the household. Callbacks are made on different days and at different times of the day. After the third call, the household is replaced by another number in that PSU.
Refusals: One callback to try to convert any designated respondent who has refused or terminated an interview. If after the conversion attempt the designated respondent still declines the interview, another household is
selected in that PSU.
Unavailable Respondents: An initial attempt and two callbacks to reach the designated member of the household. If after the third call the respondent is still not available for the interview, another household is
selected in that PSU. Busy Signals: An initial call, a follow-up fifteen minutes later, and two callbacks to reach a member of the household. Callbacks are made on LOUIS HARRIS AND ASSOCIATES
different days and at different times of the day. If the telephone is still
busy after the fourth call, a new number is selected in that PSU and the
household is replaced.

VIII. Weighting
All national public cross sections are weighted to the Census Bureau's latest population parameters on education, sex, race, and age. This adjusts these key variables, where necessary, to their actual proportions in the
population. Only moderate weighting is necessary in Harris samples. When the nearest birthday respondent selection procedure is used the data is usually · weighted to size of household.
IX. Sampling Error The results achieved from national public cross sections are subject to sampling error. Sampling error is defined as the difference between the
results obtained from the sample and those that would have been obtained had the entire population been surveyed. The size of sampling error varies both with the size of the sample and with the percentage giving a particular answer. The
following table sets forth the range of ~rror in samples of different sizes at different percentages of response:
LOUIS HARRIS AND ASSOCIATES

Table A-1
RECOMMENDED ALLOWANCE FOR SAMPLING ERROR OF PROPORTIONS (PLUS OR MINUS)
Sampling Tolerances (at 95% Confidence Level) To Use in Evaluating Any Individual Percentage Result
ApproximateSample Size Approximate Magnitude of Results
Of Any Group Survey Survey Survey Survey Asked Question Percentage Percentage Percentage Percentage Survey on Which Survey Result at Result at Result at Result at Percentage Result Is Based 10% or 90% 20% or 80% 30% or 70% 40% or 60% Result at 50%
1,250 2 2 3 3 3
1,000 2 2 3 3 3 500 3 4 4 4 4 300 3 5 5 6 6 200 4 6 6 7 7 100 6 8 9 10 10
50 8 11 13 14 14
For example, if the response for a sample size of 1,000 is 30%, in 95
cases out of 100 the response in the total population would be between 27% and
33%.
X. Significance of Difference Between Proportions
The difference between the percentage responses given by two
independent samples to the same question may or may not be significant. To
determine whether or not such a difference is indeed significant, the size of
the samples involved and the percentage giving each response must be taken into
account. The following table shows the margin of error that must be allowed for
different sample sizes at different percentages of response:
l
Table A-2
SAMPLING ERROR OF DIFFERENCE BETWEEN PROPORTIONS
Sampling Tolerances (at 95% Confidence Level)
To Use in Evaluating Differences Between
Two Percentage Results

AQQroximate Magnitude of Results Approximate Sample Size Survey Survey Survey Survey Survey of Two Groups Asked Percentage Percentage Percentage Percentage Percentage Question on Which Results at Results at Results at Results at Results Survey Results Are Based 10% or 90% 20% or 80% 30% or 70% 40% or 60% at 50%
1,250 vs. 1,250 2 3 4 4 4 1,000 2 3 4 4 4 500 3 4 5 5 5 300 4 5 6 6 6 200 4 6 7 7 7 100 6 8 9 10 10 50 8 11 13 14 14
1,000 vs. 1,000 3 4 4 4 4 500 3 4 5 5 5 300 4 5 6 6 6 200 5 6 7 7 8 100 6 8 9 10 10 50 9 11 13 14 14
500 vs. 500 4 4 6 6 6 300 4 6 7 7 7 200 6 7 8 8 8 100 7 9 10 11 11 50 9 12 13 14 15
300 vs. 300 5 6 7 8 8 200 5 7 8 9 9 100 7 9 10 11 11 50 9 12 14 15 15
l 200 vs. 200 6 8 9 10 10
100 7 10 11 12 12 50 9 12 14 15 15
l 100 vs. 100 8 11 13 14 14 50 10 14 16 17 17
50 vs. 50 12 16 18 19 20
l
Study No. 874027

ISSUED BY
NATIONAL ORGANIZATION ON DISABILITY

VOTING INTENTIONS DURING THE 1988 ELECTION:
A Comparison of Disabled and Non-disabled Voters
For:
The National Organization on Disability
Project Director:
Louis Genevie, Ph.D. Vice President
LOUIS HARRIS AND ASSOCIATES, INC.
630 FIFTH AVENUE
NEW YORK, NY 10111
(212) 698-9600

For example, suppose one group of size 1,000 has a response of 34% "yes11 for a question, and an independent group of size 300 has a response of 27% "yes" for the same question, for a difference of 7 percentage points. According to the table, this difference is subject to a potential sampling error of plus or minus 6 percentage points. Since the difference is greater than the
potential sampling error, the difference is significant. Moreover, if the entire population were interviewed, the difference would be from 1 to 13 percentage points (the 7 point difference plus or minus the 6 point potential sampling error) in 95 cases out of 100. The recommended allowances for significance of difference were calculated based on a a simple random sample.
LOUIS HARRIS AND ASSOCIATES

In this series of surveys, voters were selected from the pool of
eligibles using the following questions1
Q.: Did you vote in the 1984 presidential election when just over half the voters came out to vote, or didn't you get around to voting then?
Voted...( (__-1

Did not vote ...__-2

Not sure ...__-3

Q.: Are you 21 years of age and over or are you under 21 years of age?
21 and over ...( (__-1 18-20 ...__-2 Not sure/refused...__-3
Q.: How certain are you that you will vote in the 1988 presidential election or
absolutely certain, quite certain, probably will vote, probably will not, certainly will not?
Absolutely certain to vote ...( (__-1
Quite certain to vote ...__-2
Probably will vot~...__-3
Probably will not vote ...__-4
Certainly will not vote ...__-5
Not sure ...__-6

1In the surveys conducted from February-September, 1988, voters were identified as follows:
1.
Voted in the 1984 presidential election; or

2.
18 to 21 years of age and say they are absolutely or quite certain to vote in the 1988 presidential election.

voters
In the surveys conducted October 6-10, 1988, and October 13-17, 1988,
were identified as follows:
1.
Voted 1984 presidential election and absolutely or quite certain to vote in 1988; or

2.
Registered to vote and absolutely or quite certain to vote in 1988; or

3.
18-21 years of age and registered and absolutely or quite certain to
vote in 1988.

For the October 28-30, 1988, and November 2-7, 1988, surveys, voters were
identified as follows:
1.
Voted in 1984 presidential election and absolutely or quite certain to vote in 1988; or

2.
18-21 years of age and registered and absolutely or quite certain to
vote in 1988.

LOUIS HARRIS AND ASSOCIATES

Disabled people were identified using the following question:
Q.: Does a disability , handicap, or other physical or mental limitation currently keep you from participating fully in work, school, housework, or other activities, or not?
Yes, keeps me from participating.. ( (__-1 No, does not ...__-2 Not sure ...__-3
The exact fieldwork dates for each survey summarized in this document
can be found in Table A-3, below.
Table A-3
SURVEY FIELD DATES
Month
February
March
April
May
June
July
July
August
August
September
October
October
October
November
Dates 2/18/88-2/24/88 3/9/88-3/13/88 4/1/88-4/5/88 5/5/88-5/10/88 6/1/88-6/6/88 7/7/88-7/12/88 7/22/88-7/25/88 8/4/88-8/9/88 8/19/88-8/22/88 9/1/88-9/6/88 10/6/88-10/10/88 10/14/88-10/17/88 10/28/88-10/30/88 11/2/88-11/7/88
LOUIS HARRIS AND ASSOCIATES
TABLE OF CONTENTS
Introduction... 1
A Note on Reading the Tables ...1
Public Release of Survey Findings ...2

EXECUTIVE SUMMARY...3
Why George Bush Won The Election...3
Changes in Voting Intentions During the Campaign...4

ANALYSIS OF THE RESULTS OF THE SURVEY...6
Section 1: Who is the Disabled Voter?...7

Section 2: Party Affiliation...8

Section 3: Voting Intentions During the Primaries...8

Section 4: Changes After the Democratic National Convention ...9

Section 5: Changes After the Republican National Convention ...9

Section 6: Who Voters Think Will Do The Best Job on Key Issues Facing the Nation ...10
Economic Policy...10
Domestic Policy...12
Foreign Policy and Defense...13
Inspiring Confidence ... 13
Leadership: The Bottom Line ...14

METHODOLOGY... 27
Introduction...28
Random Digit Dialing... 28
Sample Description...29
Stratification of Telephone Sample ...30
Sample Reliability... 31
Respondent Selection... 31
Callback Strategy...32
Weighting...33
Sampling Error... 33
Significance of Difference Between Proportions ... 34

LOUIS HARRIS AND ASSOCIATES

INDEX OF TABLES

1 DISTRIBUTION OF DISABILITY AMONG VOTERS ...15

2 A COMPARISON OF DISABLED AND NON-DISABLED VOTERS ON KEY DEMOGRAPHIC VARIABLES ... 16

3 PARTY AFFILIATION OF DISABLED AND NON-DISABLED VOTERS ... 17

4 VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS AT FOUR POINTS IN TIME DURING THE PRIMARY SEASON ...18

5 VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS BEFORE AND AFTER THE DEMOCRATIC AND REPUBLICAN CONVENTIONS. 19

6 VOTING INTENTIONS OF DISABLED VOTERS BEFORE AND AFTER THE REPUBLICAN CONVENTION ...20

7 WHO VOTERS THINK WILL DO THE BEST JOB ON KEY ECONOMIC POLICY ISSUES... 21

8 WHO VOTERS THINK WILL DO THE BEST JOB ON KEY DOMESTIC POLICY ISSUES ... 22

9 WHO VOTERS THINK WILL DO THE BEST JOB ON KEY FOREIGN POLICY AND DEFENSE ISSUES ... 23

10 WHO VOTERS THINK WILL DO THE BEST JOB ON INSPIRING CONFIDENCE FROM THE PEOPLE ...24

11 LEADERSHIP RATINGS OF BUSH ... 25

12 LEADERSHIP RATINGS OF DUKAKIS ...26
METHODOLOGY
A-1 RECOMMENDED ALLOWANCE FOR SAMPLING ERROR OF PROPORTIONS {PLUS OR MINUS) ...34
A-2 SAMPLING ERROR OF DIFFERENCE BETWEEN PROPORTIONS ... 35 A-3 SURVEY FIELD DATES...38

LOUIS HARRIS AND ASSOCIATES

-1-

Introduction

This report focuses on voting intentions of disabled Americans as they emerged during the 1988 presidential election. The findings are based on a series of polls (1) conducted during the campaign by Louis Harris and Associates for the National Organization on Disability. In all, fourteen nationally representative telephone surveys of voters' intentions were conducted between February and early November, 1988. (2)

A total of 1,538 disabled voters and 19,440 non-disabled voters were identified and interviewed during this period. The field dates for each of the fourteen surveys summarized in this document are identified in Appendix A, which also contains a complete description of the methods employed for Harris telephone surveys.

Figures for age, sex, race and education were weighted where necessary to bring them into line with their actual proportions in the population. In samples of this size, one can say with 95 percent certainty that the results have a statistical precision of plus or minus three percentage points of what they would be if the entire adult population had been polled.

A Note on Reading the Tables

An asterisk (*) on a table signifies a value of less than one-half percent (0.5%). A dash (-) represents a value of zero. Percentages may not always add to 100% because of computer rounding, multiple answers from

(1) copies of the questionnaires used during the course of the campaign are available from Louis Harris and Associates.

(2) The screening questions used for the surveys can be found in Appendix A.

LOUIS HARRIS AND ASSOCIATES

-2-

respondents, or the elimination of "no answers" from particular tables. Voters who were unsure of their intentions at the time of the survey, for example, were not included in the tables.

Public Release of Survey Findings

All Louis Harris and Associates surveys are designed to adhere to the code of standards of the Council of American Survey Research Organizations (CASRO) and the code of the National Council of Public Polls (NCPP). Because data from this survey will be released to the public, any release must stipulate that the complete report is also available, rather than simply an excerpt from the survey findings.

LOUIS HARRIS AND ASSOCIATES

-3-

EXECUTIVE SUMMARY

WHY GEORGE BUSH WON THE ELECTION
1. George Bush will be the 41st president of the United States because he convinced voters -- disabled and non-disabled alike -- that he will do a better job than Michael Dukakis on most of the key issues affecting the economy, foreign policy, and defense.

- Voters think that Bush will do a better job keeping the economy prosperous (54% vs. 38%), making America more competitive abroad (54% vs. 37%), negotiating arms agreements with the Russians (59% vs. 33%) and maintaining a strong defense for the country (63% vs. 30%).

- During the course of the campaign, Bush's strength on most of these issues increased more among disabled voters than among voters generally.

2. Most voters also think that Vice President Bush will do a better job controlling crime and handling the drug problem in the country.

- Bush's support on handling the drug problem increased from 30% immediately after the Democratic national convention to 47% in late October, an increase that was reflected in both the disabled and non-disabled vote.

3. As a result of the change in voters' opinions on these key issues, the Vice President's leadership ratings rose substantially during the course of the campaign.

- Forty-seven percent of voters rated the Vice President's leadership ability positively in July, and by November, 62% did so.

4. On which candidate had the greatest ability to inspire confidence
in the people, Bush beat Dukakis 55% to 40% among all voters.

- Among disabled voters, Bush's ability to inspire confidence was even greater, 54% to 36% for Dukakis.

LOUIS HARRIS AND ASSOCIATES

-4-

- The Vice President's rise in the polls on this issue during the course of the campaign was dramatic, moving from 30% in July to 55% in late October. During the same time, Dukakis's ability to inspire confidence fell by 23 points.

5. On issues affecting the more vulnerable segments of the society -- children, the poor, the elderly and the disabled -- voters, both disabled and non-disabled, think that Governor Dukakis would do a better job.

- Dukakis's lead on these issues was still strong going into the last week of the campaign, but declined slightly after the Democratic convention.

- Bush, while still trailing Dukakis, increased his support on these issues during the course of the campaign.

6. Voters also think that Dukakis would do a better job protecting the environment and improving the quality of education.

- Dukakis's lead on these issues declined slightly during the campaign, while Bush, although still trailing at the end, picked up support.

CHANGES IN VOTING INTENTIONS DURING THE CAMPAIGN

7. As leadership on the key issues changed during the course of the
campaign, so too did voters' intentions.

- The Harris poll conducted in March gave Governor Dukakis a two point lead over the Vice President among all voters. In the three months that followed, he increased that lead to five points.

8. Dukakis's lead over Bush during the primary season was much larger among disabled voters than among voters in general.
- Among disabled voters Dukakis started in March with an 11 point lead (55% vs. 44%) and by June had increased that lead to 24 points (58% vs. 34%).

- Thus, as the Democratic convention approached, Dukakis's overall lead stood at three points and was, to a significant extent, based on the very strong support he was receiving from disabled voters.

LOUIS HARRIS AND ASSOCIATES

-5-

9. After the Democratic convention, Michael Dukakis's lead increased among both disabled and non-disabled voters.

- Among all voters, Dukakis's lead increased ten points after the convention (54% for Dukakis; 42% for Bush).

- Among the disabled, the Governor's lead of 21 points going into the Democratic convention increased to 33 points afterward.

10. Our poll after the Republican national convention showed that Dukakis's substantial lead declined sharply among both disabled and non-disabled voters.

- Dukakis's 12 point lead among all voters after the Democratic convention turned into a five point deficit after the Republican convention when Vice President Bush led 50% to 45%.

- Among disabled people, one of Dukakis's strongholds, his lead deteriorated from 33 points after the Democratic convention to only ten points after the Republican convention.

11. By mid-October the Vice President had closed the gap and two of our last three polls showed Bush ahead of Dukakis among disabled people.

- Our final poll put Dukakis ahead of Bush among the disabled (44% vs. 49%) voters. Without question, however, the Vice President made substantial inroads into the disabled vote. In the end, the change in the disabled vote during the campaign accounted for between one and three percent of the Vice President's margin of victory.

LOUIS HARRIS AND ASSOCIATES

-6-

ANALYSIS OF THE RESULTS OF THE SURVEYS

LOUIS HARRIS AND ASSOCIATES

-7

Section 1: Who is the Disabled Voter?

As part of the Harris firm's regular election year polling of probable voters, respondents were asked whether or not they had a disability, handicap, or other physical or mental limitation that keeps them from participating fully in work, school, housework, or other activities. Overall, 10% were identified as having some form of disability (Table 1).(1)

In addition to their disability, the disabled voter also faces personal and social barriers to full participation in the political process. (2) Disabled voters tend to have less education and less income than other Americans. Forty-six percent have less than a high school education, compared to only 16% of non-disabled voters. Far fewer possess four year college degrees (9%) when compared to non-disabled voters (25%). Four times as many disabled voters have household incomes of $7,500 a year or less as compared to non-disabled voters (24% vs. 6%). Given this, it is not surprising to find that disabled people are more than two and a half times as likely to be out of work as the non-disabled. Even among those who do work, only one third as many work as professionals when compared to non-disabled people (Table 2).

Disabled voters are also older than their non-disabled counterparts. Forty-one percent are above the age of 65 compared to only 17% of non-disabled
voters (Table 2).

(1) This figure should not be confused with the number of disabled people in the country. Depending on how disability is measured, this figure stands at about 15 percent of the total adult population. See The ICD Survey of Disabled Americans: Bringing Disabled Americans into the Mainstream, Louis Harris and Associates, 1986.

(2) see Participation in Voting and Elections By Disabled Americans, Louis Harris and Associates, 1987.

LOUIS HARRIS AND ASSOCIATES

-8-

Section 2: Party Affiliation

Given the fact that the disabled voter tends to be older, less educated, and not as wealthy as the non-disabled voter traditional constituencies of the Democratic party --it is not surprising to find that more disabled people say they are Democrats than Republicans.

- Fifty-two percent of the disabled are registered as Democrats, compared to 40% of non-disabled voters. Twenty-four percent of the disabled say they are Republicans, compared to 33% of non-disabled voters. And 21% of the disabled voters say they belong to neither major party, compared to 25% of the non-disabled voters (Table 3).

Section 3: Voting Intentions During the Primaries

During the primary season, prior to the nominating conventions of both parties, the voting intentions of both disabled and non-disabled people were relatively stable (Table 4). During that time, Governor Dukakis started out with a two point lead over the Vice President, and in the three months that followed increased his lead to five points (49% vs. 44%).

Dukakis's lead in the primaries was much slimmer among non-disabled voters than among those who identified themselves as disabled. Beginning in March, Dukakis held a one point lead among non-disabled voters and by June his lead had increased to three points. Among disabled voters, however, Dukakis started out in March with an eleven point lead (55% vs. 44%) and by June had increased that lead to 24 points (58% vs. 34%).

As the Democratic convention approached, Dukakis's overall lead stood at three points (Table 5). His lead, however, was, to a large extent, based on the very strong support he was receiving from disabled voters, 58% of whom supported him compared to 37% who supported the Vice President. Among non-disabled voters Dukakis held a slim one point lead prior to the Democratic convention.

LOUIS HARRIS AND ASSOCIATES

-9-

Section 4: Changes After the Democratic National Convention

Dukakis received a major boost in support as a result of the Democratic nominating convention. Our two polls taken after the convention showed that the slim lead that Dukakis held prior to the convention increased from three percent to twelve percent. During this time, Dukakis's lead increased among both disabled and non-disabled voters, but was more substantial among the disabled.

- The Governor's lead of twenty-one points among the disabled prior to the Democratic convention increased to thirty-three points afterward. Among non-disabled voters the Governor's slim one point lead increased to ten points after the convention (Table 5).

Section 5: Changes After the Republican National Convention

Between the beginning of August and the beginning of September, 1988, the substantial lead that Dukakis held after the national Democratic convention deteriorated rapidly among both disabled and non-disabled voters. Overall, Dukakis's twelve point lead after the Democratic convention turned into a five point deficit after the Republican convention, when Vice President Bush held the lead over the Governor 50% to 45%. Even among disabled people, one of Dukakis's strongholds, his lead deteriorated from thirty-three points after the Democratic convention to only ten points after the Republican convention.

Exactly when disabled voters' intentions changed during this period can be seen in the three Harris polls taken in August and September (Table 6). In early August, prior to the Republican convention, Dukakis held a 30 point lead over the Vice President among the disabled. In the next poll, taken in late August, just after the Republican national convention, a dramatic shift in the voting preference of disabled people was found. At that time, 49%, an

LOUIS HARRIS AND ASSOCIATES

-10-

increase of 16% compared to just a few weeks prior, said that they were for Bush, which placed the Vice President two points ahead of Governor Dukakis (49% for Bush, 47% for Dukakis) among the disabled.

In two of the three polls that followed, George Bush led Michael Dukakis among the disabled and non-disabled alike. Our final poll, however, placed Bush ahead of Dukakis 52% to 45% among all voters, but trailing Dukakis among the disabled, 44% to 49% (Table 6).

Observation:
There is no doubt that the Vice President made inroads into this traditionally Democratic constituency. Dukakis's substantial lead among the disabled dissipated during the course of the election and disabled people, despite their traditionally Democratic leanings, voted for George Bush in numbers almost equal to that of the nation as a whole.

Section 6: Who Voters Think Will Do The Best Job on Rey Issues Facing the Nation

In addition to polling voters' intentions during the course of the campaign, the Harris firm also asked voters which candidate they thought would do the best job on a variety of problems facing the nation, including key domestic and foreign policy issues.

Economic Policy

On the issue of which candidate would be more likely to keep the country prosperous, the latest Harris poll* on this issue placed George Bush ahead of Michael Dukakis 54% to 38% (Table 7). Disabled voters thought that George Bush could do a better job in this regard by a similar margin. The

*October 13-17, 1988.

LOUIS HARRIS AND ASSOCIATES

-11-

disabled vote, however, changed more dramatically than the total national vote. Immediately after the Democratic national convention (late July), most voters felt that Michael Dukakis would do better keeping the economy prosperous (53% vs. 46% for Vice President Bush). Among the disabled, however, Dukakis led Bush, 67% to 24%. Between late July and late October the number of disabled voters who thought that Bush would do a better job with the economy increased 34 points from 24% to 53%. During that same time, the proportion of all voters who thought that Bush would do a better job with the economy increased by eight points, more than one and one-half of which were attributable to the change in disabled people's opinion.

On the issue of making America more competitive abroad, the majority of voters thought that George Bush would do a better job (53% of the disabled voters; 54% of all voters). These latest figures represented strong increases for Bush during the course of the campaign. Since late July, the proportion of all voters who thought that the Vice President would make America more competitive abroad increased 15 points; and the proportion of disabled voters who thought so increased by 23 points (Table 7).

Voters also thought that George Bush would do better than Michael Dukakis at controlling inflation. On this issue, Bush held a 51% to 39% lead overall, but the disabled vote went for Dukakis, 48% to 44%.

On their ability to cut the rate of unemployment in the country, Bush and Dukakis were in a virtual dead heat, 46% for Bush, 45% for Dukakis. Even though disabled people were going for Dukakis on this issue 49% to 38%, Vice President Bush picked up six points among disabled voters on this issue between May and October, while the disabled vote for Dukakis remained at 49%. Among all voters Vice President Bush picked up eight points between May and October, while Governor Dukakis lost three points.

LOUIS HARRIS AND ASSOCIATES

-12-

Voters were concerned about the Vice President's ability to cut the deficit and thought that Dukakis would do a better job cutting Federal spending (46% vs. 43% for Bush). That margin is wider among disabled voters, 51% for Dukakis, 40% for Bush.

As far as cutting the federal deficit is concerned, the electorate also felt that Dukakis would do a better job. However, his margin was very slim (45% to 44%). Disabled voters preferred George Bush, 43% to 36%, on this issue (Table 7).

Domestic Policy

By substantial margins, voters thought that Governor Dukakis would do a better job protecting the environment and improving the quality of education and health care in the country. Both disabled and non-disabled voters also thought that Dukakis would do a better job protecting the more vulnerable segments of society, including disabled people, children, the elderly and the poor. For example, voters -- disabled and non-disabled, thought that Dukakis would do a better job helping the disabled, (57% vs. 34% for Bush among all voters, and 55% vs. 33% among disabled voters) (Table 8).

While Dukakis continued to hold a substantial lead on these issues of conscience, in most instances, Vice President Bush made substantial gains during the course of the campaign among disabled and non-disabled voters alike. The proportion supporting Dukakis, on the other hand, either remained stable or declined slightly. The candidate voters thought would do the best job improving the quality of public education is a good example of this trend. When we took our last poll before the election, Dukakis led Bush on this issue 51% to 38% nationally, figures that are similar for the disabled vote. Since May, however, Bush increased his support among voters on this issue from 29% to 38%, while

LOUIS HARRIS AND ASSOCIATES

-13-

Dukakis's support declined slightly from 54% to 51%. This trend was even stronger among the disabled than among the voters generally. Michael Dukakis was still thought by most voters to be able to do the best job dealing with these problems, but George Bush made substantial gains in convincing voters that he could do the job better (Table 8).

Foreign Policy and Defense

The majority of voters thought that George Bush would do a better job maintaining a strong defense for the country (63% vs. 30%). However, by a similar margin (59% vs. 34%), voters thought that Michael Dukakis would do a better job controlling defense spending. The figures were similar for disabled voters on these two issues.

As far as working for world peace and negotiating nuclear arms control agreements with the Russians were concerned, voters -- both disabled and non-disabled favored George Bush by wide margins.

On bringing peace to Central America, however, Bush's margin was not anywhere near as great -- 46% -- compared to 42% for Dukakis. Disabled voters favored Dukakis on this issue 47% to 42% (Table 9).

Inspiring Confidence

Bush led Dukakis 55% to 35% on the key issue of inspiring confidence in the people. Among disabled voters Bush's lead was even greater, 59% vs. 30%. The Vice President's increase on this key issue was dramatic. Since late July, when only 30% of those polled said they thought that he would do the best job inspiring confidence, the Vice President's support rose by fully 25 points, to 55%, while Governor Dukakis's fell by 28 points to only 35%. (Table 10).

LOUIS HARRIS AND ASSOCIATES

-14-

The shift was even greater among disabled voters than among voters in general. Bush increased the proportion of disabled voters who thought he would do the best job inspiring confidence by fully 31 points while Dukakis's support declined by 37 points during the course of the campaign (Table 10).

Leadership: The Bottom Line

Perhaps the most telling evaluation by voters was that made of the leadership potential of the candidates. In many ways, George Bush's ratings during the course of the campaign tell the story of the campaign in a nutshell. Prior to the Republican national convention, the Vice President's positive rating on leadership was below 50%. After his strong performance at the convention, Bush's rating soared and by the end of the campaign stood at 62% (Table 11). Meanwhile, Dukakis's leadership rating was moving a similar magnitude in the opposite direction (Table 12).

Observation:

Like most voters, disabled people think that George Bush will do the best job on most of the key issues facing the nation. This, despite the fact that disabled and non-disabled voters alike think that Dukakis would have done a better job improving the quality of education, protecting the environment, and helping the more vulnerable segments of the society -- including the disabled.

The fact that George Bush mentioned his support for the disabled in his acceptance speech at the Republican national convention should not be overlooked. During that speech Bush said, "I'm going to do whatever it takes to make sure the disabled are included in the mainstream. For too long they've been left out, but they're not going to be left out anymore." This was the first time a candidate for national office had addressed the nation's disabled citizens directly; on at least two other occasions before major national audiences, the Vice President repeated his pledge. Although Governor Dukakis also had a very strong position in support of the disabled, he made no clear statement of support during his acceptance speech or in any national forum afterward.

LOUIS HARRIS AND ASSOCIATES

-15-

Table 1

DISTRIBUTION OF DISABILITY AMONG VOTERS

Q.: Does a disability, handicap, or other physical or mental limitation currently keep you from participating fully in work, school, housework, or other activities, or not?

Total 13613

%
Yes, keeps me from participating 10 (%)
No, does not 89 (%)
Not sure 1 (%)

LOUIS HARRIS AND ASSOCIATES

-16-

Table 2
A COMPARISON OF DISABLED AND NON-DISABLED VOTERS ON KEY DEMOGRAPHIC VARIABLES
Total Disabled Voters Non-Disabled Voters
Base 13613 1124 12322
% % %

Sex
Male 49 47 49
Female 51 53 51

Occupation
Professional 14 5 16
Manager, official 12 4 12
Proprietor 5 3 5
Clerical worker 5 4 6
Sales worker 5 2 5
Craftsman, foreman 13 6 14
Operative, unskilled labor 10 8 10
All others 13 16 12
Not working 23 51 20

Age
18-24 10 3 11
25-34 20 5 21
35-44 20 10 21
45-64 31 39 30
65-74 12 24 11
75+ 7 17 6

Education
Less than high school 19 46 16
High school graduate 36 31 36
Some college 21 13 22
4-year college graduate or more 23 9 25

Household Income
$7,500 or less 8 24 6
$7,501 to $15,000 12 23 11
$15,001 to $25,000 19 20 19
$25,001 to $35,000 19 11 20
$35,001 to $50,000 19 9 20
$50,001 or more 16 4 18

LOUIS HARRIS AND ASSOCIATES

-17-

Table 3

PARTY AFFILIATION OF DISABLED AND NON-DISABLED VOTERS

Total Disabled Voters Non-Disabled Voters
Base 13613 1124 12322
% % %
Party Affiliation
Republican 32 24 33
Democrat 41 52 40
Independent 25 21 25
Other/ not sure 2 2 2

LOUIS HARRIS AND ASSOCIATES

-18-

Table 4

VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS AT FOUR POINTS IN TIME
DURING THE PRIMARY SEASON*

Q.a: Of course, this November for president, it will be Vice President George Bush for the Republicans and Governor Michael Dukakis for the Democrats. If you had to choose right now, would you be for Bush or Dukakis?

Q.b: Well, if you had to say, would you lean toward Bush the Republican or toward Dukakis the Democrat?

Total Voters Disabled Voters Non-Disabled Voters
Base Bush Dukakis Base Bush Dukakis Base Bush Dukakis
March, 1988 % % % % % %
(3/9-3/13) 1260 47 49 119 44 55 1135 48 49
April, 1988
(4/1-4/5) 1248 47 49 88 35 58 791 47 49
May, 1988
(5/5-5/10) 1256 43 50 107 37 57 1143 44 50
June, 1988
(6/1-6/6) 1246 44 49 108 34 58 1135 45 48
*Totals do not add to 100% due to voters who were not sure how they would vote.

LOUIS HARRIS AND ASSOCIATES

-19-

Table 5

VOTING INTENTIONS OF DISABLED AND NON-DISABLED VOTERS BEFORE AND AFTER THE DEMOCRATIC AND REPUBLICAN CONVENTIONS*

Q.a: Of course, this November for president, it will be Vice President George Bush for the Republicans and Governor Michael Dukakis for the Democrats. If you had to choose right now, would you be for Bush or Dukakis?

Q.b: Well, if you had to say, would you lean toward Bush the Republican or toward Dukakis the Democrat?

Total Voters Disabled Voters Non-Disabled Voters
Base Bush Dukakis Base Bush Dukakis Base Bush Dukakis
Before the
Democratic
Convention
(6/1-6/6 and 7/7-7/12) 2491 % 46 49 208 % 37 58 2243 % 47 48
After the
Democratic
Convention
(7/22-7/25and 8/4-8/9) 2662 % 42 54 222 % 31 64 2415 % 43 53
After the
RepublicanConvention
(8/19-8/22and 9/1-9/6) 2586 % 50 45 186 % 43 53 2381 % 50 45
Mid-Oct. (10/14-10/17) 1350 % 53 43 90 % 54 36 1256 % 53 43
Late Oct. (10/28-10/30) 1249 % 52 45 80 % so 48 1161 % 52 44
Early Nov. (11/2-11/7) 3216 % 52 45 244 % 44 49 2638 % 53 44

*Totals do not add to 100% due to voters who were not sure how they would vote.

LOUIS HARRIS AND ASSOCIATES

Table 6
VOTING INTENTIONS OF DISABLED VOTERS BEFORE AND AFTER THE REPUBLICAN CONVENTION*
Q.a: Of course, this November for president, it will be Vice President George
Bush for the Republicans and Governor Michael Dukakis for the Democrats. If you had to choose right now, would you be. for Bush or Dukakis?
Q.b: Well, if you had to say, would you lean toward Bush the Republican or toward Dukakis the Democrat?
Disabled Voters
Base** Bush Dukakis
Early August (8/4-8/9) 125 % 33 63
Late August (8/19-8/22) 88 % 49 47
Early Sept. (9/1-9/6) 99 % 37 59
Early Oct. (10/6-10/10) 103 % 47 46
Mid-Oct.
(10/14-10/17) 90 % 54 36
Late Oct.
(10/28-10/30) 80 % 50 48
Early Nov. (11/2-11/-7) 244 % 44 49

*Totals do not add to 100% due to voters who were not sure how they would vote.
**Even though these bases are small, the changes of 16% for Bush from early August to late August is still significant at the 95% confidence level.
LOUIS HARRIS AND ASSOCIATES

Table 7
WHO VOTERS THINK WILL DO THE BEST JOB ON KEY ECONOMIC POLICY ISSUES Total Voters Disabled Voters
Base* Bush Du-Neither/ kakis Not Sure- Base Du-Neither/ Bush kakis Not Sure
KeeEing the economy ErOSEerous MayEarly July Late July Late August Early October Mid-October 1257 % 44 1245 % 46 1266 % 46 1295 % 55 1354 % 52 1353 % 54 44 45 53 40 42 38 12 8 7 6 6 9 107 % 30 100 %. 36 98 24 88 % 44 103 % 45 90 % 53 52 52 67 43 48 33 18 12 10 13 8 14
Making America more competitive abroad
May Early JulyLate July Late August Early October Mid-October , 1257 % 39 1246 % 47 1266 % 39 1295 % 54 1355 % 48 1356 % 54 46 42 53 40 41 37 15 11 8 6 10 9 107 % 28 100 % 40 98 % 30 87 % 46 103 % 39 90 % 53 53 so 59 44 45 32 18 10 12 11 16 15
Cutting federal SEending MayEarly July Early October 1257 % 41 1234 % 41 1355 % 43 44 48 46 15 11 11 107 % 28 100 % 36 103 % 40 51 52 51 21 12 9
Handling the problem of the federal deficit
MayEarly JulyEarly October Mid-October 1257 % 38 1240 % 42 1354 % 42 1356 % 44 45 46 47 45 16 11 11 12 107 % 24 99 % 32 103 % 38 90 % 43 53 58 54 36 22 10 8 20
Keeping inflation under control
May Early October 1257 % 48 1354 % 51 38 39 14 10 107 % 31 103 % 44 48 48 20 8
Cutting the rate of unemElovment
MayEarly October 1256 % 38 1354 % 46 48 45 14 9 107 % 32 103 % 38 49 49 19 13

*Bases vary according to how many times the question was asked during the course of the campaign.
LOUIS HARRIS AND ASSOCIATES

-22Table 8
WHO VOTERS THINK WILL DO THE BEST JOB ON
KEY DOMESTIC POLICY ISSUES
Total Voters Disabled Voters Du-Neither/ Du-Neither/ Base* Bush kakis Not Sure Base Bush kakis Not Sure
Handling the drug problem
in the country
May 1257 % 33 41 26 107 % 27 47 26
Early July 1245 % 35 44 21 100 % 28 48 23
Late July 1266 % 30 53 17 98 % 28 55 18
Late August 1296 \ 48 41 12 88 \ 49 42 9
Early October 1355 \ 43 45 11 103 \ 43 46 11
Mid-October 1356 \ 47 40 13 90 % 43 32 25

Controllin9 crime Mid-October 1356 % 63 27 10 90 \ 53 34 13
HelEing the disabled Late August 1296 % 33 58 9 88 % 36 52 12 Early October 1355 % 34 57 9 103 \ 33 55 12
Protectin9 the environment Late August 1294 % 38 so 12 88 \ 35 47 18 Early October 1355 \ 41 49 9 103 \ 39 47 14 Mid-October 1356 \ 40 48 11 90 \ 43 41 17
Helping children get a better break Early July 1245 \ 27 56 16 100 \ 24 62 15 Early October 1355 % 34 56 9 103 \ 34 53 13 Mid-October 1356 \ 35 54 10 90 \ 39 51 10
Improving the quality of Eublic education May 1256 % 29 54 17 107 \ 24 58 17 Early July 1245 \ 32 54 13 100 \ 27 59 14 Early October 1355 \ 37 53 10 103 % 38 49 13 Mid-October 1355 \ 38 51 11 90 % 37 48 16
Helping the elderly and EOor get a better break May 1257 % 21 66 11 107 \ 24 60 15 Early October 1355 \ 31 62 7 103 \ 35 60 6
Making sure all workers are covered with health insurance provided by their e!!!Eloyer
May 1257 \ 22 64 14 107 \ 22 64 14 Early October 1355 % 24 66 10 103 \ 29 60 11 Mid-October 1355 \ 28 62 11 90 \ 35 55 10
*Bases vary according to how many times the question was asked during the course of the campaign.
LOUIS HARRIS AND ASSOCIATES
Table 9
WHO VOTERS THINK WILL DO THE BEST JOB ON
KEY FOREIGN POLICY AND DEFENSE ISSUES

Total VoteI"s Disabled VoteI"s Du-NeitheI"I Du-NeitheI"/ Base* Bush kakis Not Sure Base Bush kakis Not SUI"e
Working for peace in the WOI"ld Late August 1296 % 53 39 8 88 % 52 41 7 EaI"ly octobeI" 1355 % 49 42 9 103 % 45 51 4 Mid-OctobeI" 1356 % 53 37 11 90 % 57 28 15
Negotiating nuclear arms contI"ol agreements with the Russians May 1257 % 54 32 14 107 % 41 40 18 EaI"ly October 1355 % 59 33 8 103 % 52 35 12
BI"inging peace to Central Amedca Euly OctobeI" 1355 % 46 42 12 103 % 42 47 11
Maintaining a strong defense foI" the countI"y May 1257 % 62 28 10 107 % 41 41 18 Eady July 1245 % 60 31 8 100 % 56 33 11 Eady OctobeI" 1355 % 63 30 6 103 % 55 34 11
ContI"olling defense spendingMay 1256 % 31 57 12 107 % 29 55 16 EaI"ly OctobeI" 1354 % 34 59 7 103 % 38 53 8
*Bases vaI"y accoI"ding to how many times the question was asked duI"ing the couI"se of the campaign.
-24
Table 10
WHO VOTERS THINK WILL DO THE BEST JOB
ON INSPIRING CONFIDENCE FROM THE PEOPLE

Total Voters Disabled VotersDu-Neither/ Du-Neither/Base* Bush kakis Not Sure ~ Bush kakis Not Sure
Inspiring confidence fromthe 12eo12leMay 12SS % 38 49 13 107 % 31 SS 14 Early July 1246 % 36 S2 12 100 % 27 S8 14 Late July 1266 % 30 63 6 98 % 28 67 6 Late August 1296 % 46 49 6 88 % 4S so s Early October 13S3 % 48 44 7 103 % 4S 4S 10 Mid-October 13S3 % SS 3S 9 90 % S9 30 11 Late October 12SO % SS 40 5 80 % S4 36 10
*Bases vary according to how many times the question was asked during the course of the campaign.
LOUIS HARRIS AND ASSOCIATES
Table 11
LEADERSHIP RATINGS OF BUSH
Q.: How would you rate (READ EACH NAME) as a leader, on a scale of 1 to 10, with 1 being the worst and 10 the best?
Total Voters Disabled Voters Pas-Neg-Pos-Neg-
Base itive ative Not Sure Base itive ative Not Sure
Early July
(7/7-7/12) 1245 % 47 50 3 100 % 44 48 8

Late July
(7/22-7/25) 1266 % 47 so 3 98 % 39 56 5

Early Auqust
(8/4-8/9) 1401 % 49 49 2 125 % 38 56 6

Late Auqust
(8/19-8/22) 1295 % 58 40 2 88 % 59 38 3

Early Sept.
(9/1-9/6) 1299 % 60 37 3 99 % 45 44 11

Early Oct.
(10/6-10/10) 1354 % 60 38 2 103 % 49 42 9

Mid-Oct.
(10/14-10/17) 1356 % 56 42 2 90 % 58 39 3

Late Oct.
(10/28-10/30) 1250 % 60 38 2 80 % 64 33 3

Early Nov.
(11/2-11/7) 3221 % 62 36 2 244 % 56 40 4

LOUIS HARRIS AND ASSOCIATES

Table 12
LEADERSHIP RATINGS OF DUKAKIS
Q. : How would you rate (READ EACH NAME) as a leader, on a scale of 1 to 10, with 1 being the worst and 10 the best?
Total Voters Disabled Voters Pos-Neg-Pos-Neg-
Base itive ative Not Sure Base itive ative Not Sure
Early July
(7/7-7/12) 1245 % 55 38 7 100 % 55 36 9

Late July
(7/22-7/25) 1266 % 64 33 3 98 % 71 25 4

Early August
(8/4-8/9) 1401 % 57 38 5 125 % 67 28 5

Late August
(8/19-8/22) 1295 % 55 42 3 88 % 62 35 3

Early Sept.
(9/1-9/6) 1299 % 55 41 4 99 % 54 32 14

Early Oct.
(10/6-10/10) 1354 % 55 42 3 103 % 53 40 7

Mid-Oct.
(10/14-10/17) 1356 % 49 47 4 90 % 42 52 6

Late Oct.
(10/28-10/30) 1250 % 53 44 3 80 % 48 48 4

Early Nov.
(11/2-11/7) 3221 % 51 46 3 .244 % 56 40 4

LOUIS HARRIS AND ASSOCIATES

-27
APPENDIX A: METHODOLOGY

LOUIS HARRIS AND ASSOCIATES

METHODOLOGY
I. Introduction
Louis Harris and Associates' telephone surveys are based on a national sample of the civilian population 18 years of age and older of the United States. Those living in Alaska and Hawaii are not represented, nor are those in prisons, hospitals, or religious and educational institutions, unless such individuals have their own outside phone line. Samples are based on the Census Bureau's adult population figures for each state in the country. These figures are updated by intercensal estimates produced annually by the Census Bureau, and sample locations are selected biannually to reflect changes in the country's demographic and geographic profile.
National samples are stratified on two dimensions --geographic region and metropolitan vs. nonmetropolitan residence. Stratification ensures that the sample will reflect within 1% those living in different regions of the country and those living in SMSAs and non-SMSAs. Within each stratum, the selection of the primary sampling unit (PSU) is achieved through multistage unclustered sampling. Each PSU yields one interview. First states, ·then counties, and then minor civil divisons are selected, with probability of selection proportionate to Census Bureau estimates of their respective populations.
II. Random Digit Dialing
Louis Harris and Associates employs a random digit dialing procedure to select households within primary sample units. Random digit dialing is a significant improvement over previously used techniques, extending the potential coverage rate to almost 95% of the U.S. population.
Unless some method of random digit dialing is used, telephone samples must be drawn directly from published lists. However, since the population of
unlisted phone number subscribers is large and demographically dissimilar to
subscribers with listed phone numbers, reliance on telephone directories alone
yields seriously biased samples. For this reason, using published phone listing as the universe is inadequate for telephone surveys and inferior to using random digit dailing.
The use of a random digit dialing method offers other important
advantages:
The sample is highly representative.
Unlisted telephone numbers have the same probability of inclusion in the sample as listed numbers. This is particularly important in reaching both high income and minority populations.
Respondents are geographically dispersed, rather than clustered.
Households with multiple phone listing have the same probability of inclusion in the sample as households with a single phone listing.
III. Sample Description The Louis Harris and Associates national telephone sample is selected via a multistage, stratified selection process.
The first stage of sampling involves the selection of 1,000 county sampling points (or primary selections). These sampling points are selected within 144 strata defined on the basis of the cross classification of individual
states and metropolitan/non-metropolitan areas. The selection of these points is based upon probabilities proportional to population estimates, using updates
from the 1980 Census. The second stage of sampling involves the selection of a specific telephone directory for each of the 1,000 primary selections.
LOUIS HARRIS AND ASSOCIATES

-30
This is accomplished in several steps. First, all telephone
directories for counties containing primary selections are obtained by the sampling office. From census definitions, a unique list of all separate
geographic listings is obtained. In this process duplicate geographic listings which appear in multiple directories are eliminated. Next, these geographic
listings are stratified by size. Finally, using the initial random numbers used to select the primary selection, a relative within primary unit number is used to obtain a specific geographic listing for each primary selection. In the third step of sampling, each primary selection (which is now located within a telephone list) is followed down to a specific telephone number
consisting of an area code, a 3 place prefix and a 4 place suffix. Deleting the last two digits of this suffix defines a bank of 100 telephone numbers.
For each selected bank, a series of random two place numbers are generated. Dialing is then sequentially initiated. If the number is not a household, or if it does not result in a successful screen/interview, a
replacement is selected by going to the next complete phone number. This process is repeated until a successful screen/interview is completed, or the
field period is closed.
IV. Stratification of Telephone Sample The core telephone exchanges are stratified by two variables -geographic region and place of residence. The United States is divided into
four regions, as follows:
1. East: Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, Connecticut, New York, New Jersey, Pennsylvania, Delaware, Maryland, District of Columbia, and West Virginia.
2. South: Virginia, North Carolina, South Carolina,
Florida, Georgia, Kentucky, Tennessee, Alabama, Mississippi,Arkansas, Louisiana, Texas, and Oklahoma.
LOUIS HARRIS AND ASSOCIATES
3.
Midwest: Ohio, Michigan, Indiana, Illinois, Wisconsin, Minnesota, Iowa, Missouri, Kansas, Nebraska, South Dakota, and North Dakota.

4.
West: Montana, Wyoming, Colorado, New Mexico, Arizona, Utah, Idaho, Nevada, California, Oregon, and Washington.

We also use three categories for place of residence:
1.
Central City: Every place defined by the Census Bureau as a central city of a Standard metropolitan Statistical Area. (SMSA).

2.
Rest of Metropolitan Area: Every place that is not a central city but is within SMSA boundaries.

3.
Outside Metropolitan Area: Every place that is not included in any of the other two categories.

We then define each place or residence category within each region as a stratum.
V. Sample Reliability
In order to maintain reliability and integrity in the sample, the
telephone field staff follows these procedures when a respondent contact is
attempted: A nonanswering telephone is dialed three times over a three-day period. At the end of this time, if no contact is made, a new telephone number is generated for that PSU.
If a business telephone is reached or if contact is made with a household in which a potential respondent presents a language barrier, or a telephone is not in service, a new telephone number is generated for that PSU.
Once a residential contact is established, the interviewer uses a respondent selection procedure to designate someone in the household for the interview.
VI. Respondent Selection
Within each household contacted for the survey, the interviewer uses a
respondent selection procedure to determine which adult member of the household
will be interviewed. Under the standardized procedure, the interviewer asks for
one of the following household members, in order of priority: (1) youngest
LOUIS HARRIS AND ASSOCIATES

adult male at home; (2) next youngest adult male at home; (3) youngest adult female-at home; or (4) next youngest adult female at home. This procedure results in an acceptable distribution of respondents' sex and age.
If a respondent selection grid is required to select a respondent in a household, the household member with the nearest birthday will be selected as the eligible respondent. This procedure eliminates the complexities involved with the Troldahl-Carter grid while increasing respondents' willingness to participate in the survey. Use of this method results in an acceptable distribution of respondents' sex and age.
VII. Callback Strategy In order to attain the highest possible response rates within reasonable cost constraints, callbacks are made according to the following guidelines: No Answer/Not-at-Homes: An initial attempt and then three callbacks to reach an adult member of the household. Callbacks are made on different days and at different times of the day. After the third call, the household is replaced by another number in that PSU.
Refusals: One callback to try to convert any designated respondent who has refused or terminated an interview. If after the conversion attempt the designated respondent still declines the interview, another household is
selected in that PSU.
Unavailable Respondents: An initial attempt and two callbacks to reach the designated member of the household. If after the third call the respondent is still not available for the interview, another household is
selected in that PSU. Busy Signals: An initial call, a follow-up fifteen minutes later, and two callbacks to reach a member of the household. Callbacks are made on LOUIS HARRIS AND ASSOCIATES
different days and at different times of the day. If the telephone is still
busy after the fourth call, a new number is selected in that PSU and the
household is replaced.

VIII. Weighting
All national public cross sections are weighted to the Census Bureau's latest population parameters on education, sex, race, and age. This adjusts these key variables, where necessary, to their actual proportions in the
population. Only moderate weighting is necessary in Harris samples. When the nearest birthday respondent selection procedure is used the data is usually · weighted to size of household.
IX. Sampling Error The results achieved from national public cross sections are subject to sampling error. Sampling error is defined as the difference between the
results obtained from the sample and those that would have been obtained had the entire population been surveyed. The size of sampling error varies both with the size of the sample and with the percentage giving a particular answer. The
following table sets forth the range of ~rror in samples of different sizes at different percentages of response:
LOUIS HARRIS AND ASSOCIATES

Table A-1
RECOMMENDED ALLOWANCE FOR SAMPLING ERROR OF PROPORTIONS (PLUS OR MINUS)
Sampling Tolerances (at 95% Confidence Level) To Use in Evaluating Any Individual Percentage Result
ApproximateSample Size Approximate Magnitude of Results
Of Any Group Survey Survey Survey Survey Asked Question Percentage Percentage Percentage Percentage Survey on Which Survey Result at Result at Result at Result at Percentage Result Is Based 10% or 90% 20% or 80% 30% or 70% 40% or 60% Result at 50%
1,250 2 2 3 3 3
1,000 2 2 3 3 3 500 3 4 4 4 4 300 3 5 5 6 6 200 4 6 6 7 7 100 6 8 9 10 10
50 8 11 13 14 14
For example, if the response for a sample size of 1,000 is 30%, in 95
cases out of 100 the response in the total population would be between 27% and
33%.
X. Significance of Difference Between Proportions
The difference between the percentage responses given by two
independent samples to the same question may or may not be significant. To
determine whether or not such a difference is indeed significant, the size of
the samples involved and the percentage giving each response must be taken into
account. The following table shows the margin of error that must be allowed for
different sample sizes at different percentages of response:
l
Table A-2
SAMPLING ERROR OF DIFFERENCE BETWEEN PROPORTIONS
Sampling Tolerances (at 95% Confidence Level)
To Use in Evaluating Differences Between
Two Percentage Results

AQQroximate Magnitude of Results Approximate Sample Size Survey Survey Survey Survey Survey of Two Groups Asked Percentage Percentage Percentage Percentage Percentage Question on Which Results at Results at Results at Results at Results Survey Results Are Based 10% or 90% 20% or 80% 30% or 70% 40% or 60% at 50%
1,250 vs. 1,250 2 3 4 4 4 1,000 2 3 4 4 4 500 3 4 5 5 5 300 4 5 6 6 6 200 4 6 7 7 7 100 6 8 9 10 10 50 8 11 13 14 14
1,000 vs. 1,000 3 4 4 4 4 500 3 4 5 5 5 300 4 5 6 6 6 200 5 6 7 7 8 100 6 8 9 10 10 50 9 11 13 14 14
500 vs. 500 4 4 6 6 6 300 4 6 7 7 7 200 6 7 8 8 8 100 7 9 10 11 11 50 9 12 13 14 15
300 vs. 300 5 6 7 8 8 200 5 7 8 9 9 100 7 9 10 11 11 50 9 12 14 15 15
l 200 vs. 200 6 8 9 10 10
100 7 10 11 12 12 50 9 12 14 15 15
l 100 vs. 100 8 11 13 14 14 50 10 14 16 17 17
50 vs. 50 12 16 18 19 20
l
Study No. 874027

ISSUED BY
NATIONAL ORGANIZATION ON DISABILITY

VOTING INTENTIONS DURING THE 1988 ELECTION:
A Comparison of Disabled and Non-disabled Voters
For:
The National Organization on Disability
Project Director:
Louis Genevie, Ph.D. Vice President
LOUIS HARRIS AND ASSOCIATES, INC.
630 FIFTH AVENUE
NEW YORK, NY 10111
(212) 698-9600

For example, suppose one group of size 1,000 has a response of 34% "yes11 for a question, and an independent group of size 300 has a response of 27% "yes" for the same question, for a difference of 7 percentage points. According to the table, this difference is subject to a potential sampling error of plus or minus 6 percentage points. Since the difference is greater than the
potential sampling error, the difference is significant. Moreover, if the entire population were interviewed, the difference would be from 1 to 13 percentage points (the 7 point difference plus or minus the 6 point potential sampling error) in 95 cases out of 100. The recommended allowances for significance of difference were calculated based on a a simple random sample.
LOUIS HARRIS AND ASSOCIATES

In this series of surveys, voters were selected from the pool of
eligibles using the following questions1
Q.: Did you vote in the 1984 presidential election when just over half the voters came out to vote, or didn't you get around to voting then?
Voted...( (__-1

Did not vote ...__-2

Not sure ...__-3

Q.: Are you 21 years of age and over or are you under 21 years of age?
21 and over ...( (__-1 18-20 ...__-2 Not sure/refused...__-3
Q.: How certain are you that you will vote in the 1988 presidential election or
absolutely certain, quite certain, probably will vote, probably will not, certainly will not?
Absolutely certain to vote ...( (__-1
Quite certain to vote ...__-2
Probably will vot~...__-3
Probably will not vote ...__-4
Certainly will not vote ...__-5
Not sure ...__-6

1In the surveys conducted from February-September, 1988, voters were identified as follows:
1.
Voted in the 1984 presidential election; or

2.
18 to 21 years of age and say they are absolutely or quite certain to vote in the 1988 presidential election.

voters
In the surveys conducted October 6-10, 1988, and October 13-17, 1988,
were identified as follows:
1.
Voted 1984 presidential election and absolutely or quite certain to vote in 1988; or

2.
Registered to vote and absolutely or quite certain to vote in 1988; or

3.
18-21 years of age and registered and absolutely or quite certain to
vote in 1988.

For the October 28-30, 1988, and November 2-7, 1988, surveys, voters were
identified as follows:
1.
Voted in 1984 presidential election and absolutely or quite certain to vote in 1988; or

2.
18-21 years of age and registered and absolutely or quite certain to
vote in 1988.

LOUIS HARRIS AND ASSOCIATES

Disabled people were identified using the following question:
Q.: Does a disability , handicap, or other physical or mental limitation currently keep you from participating fully in work, school, housework, or other activities, or not?
Yes, keeps me from participating.. ( (__-1 No, does not ...__-2 Not sure ...__-3
The exact fieldwork dates for each survey summarized in this document
can be found in Table A-3, below.
Table A-3
SURVEY FIELD DATES
Month
February
March
April
May
June
July
July
August
August
September
October
October
October
November
Dates 2/18/88-2/24/88 3/9/88-3/13/88 4/1/88-4/5/88 5/5/88-5/10/88 6/1/88-6/6/88 7/7/88-7/12/88 7/22/88-7/25/88 8/4/88-8/9/88 8/19/88-8/22/88 9/1/88-9/6/88 10/6/88-10/10/88 10/14/88-10/17/88 10/28/88-10/30/88 11/2/88-11/7/88
LOUIS HARRIS AND ASSOCIATES

Survey: Voting Intentions During the 1988 Election: A Comparison of Disabled and Non-disabled Voters

Item