This Washington Post-Schar School poll was conducted Oct. 15-21, 2018, among a sample
of 1,545 registered voters in 69 Congressional battleground districts, including 1,269
likely voters. A random sample of voters from state voter registration files was
contacted by mail to complete a survey online or through an automated phone interview
in September and October. This survey is the second wave of interviews with these
respondents. All results are among likely voters and have a margin of sampling error
of plus or minus 3 percentage points. The sample of registered voters is weighted to
estimates of the population of registered voters in battleground districts. Sampling,
data collection and tabulation by SSRS of Glen Mills, Pa.
*= less than 0.5 percent
(Full methodological details appended at the end)
1. Please rate the chances that you will vote in the election for Congress on November
6th in your district: Are you absolutely certain to vote, will you probably vote, are the
chances 50-50, or less than that?

10/21/18
10/5/18

Certain
to vote
88
88

Probably
vote
4
9

Chances
50/50
2
2

Less than
50/50
1
1

Will not
vote
*
*

Already
voted
5
NA

No
opinion
*
*

2. If the election for Congress in your district were being held today, for whom would
you vote? Would you lean toward ([INSERT DEMOCRAT CANDIDATE IN RESPONDENT’S DISTRICT],
the Democrat) or ([INSERT REPUBLICAN CANDIDATE IN RESPONDENT’s DISTRICT], the
Republican)? (IF ALREADY VOTED) Confidentially and for statistical purposes only, for
whom did you vote...?
Summary table among likely voters

10/21/18
10/5/18

Democratic
candidate
50
50

Republican
candidate
47
46

Another
candidate
1
*

No
opinion
3
4

3. (ASKED IF SUPPORTING THE DEMOCRATIC CANDIDATE) Would you say your vote is more (FOR
the Democratic candidate) or more (AGAINST the Republican candidate)?

10/21/18

For the
Democrat
45

Against the
Republican
29

Both
equally
23

Neither
3

4. (ASKED IF SUPPORTING THE REPUBLICAN CANDIDATE) Would you say your vote is more (FOR
the Republican candidate) or more (AGAINST the Democratic candidate)?

10/21/18

For the
Republican
51

Against the
Democrat
27

Both
equally
18

Neither
4

 5. Which best describes your impression of the Democratic Party?

10/21/18

--------Favorable-------NET
Strongly
Somewhat
48
20
28

-------Unfavorable------NET
Somewhat
Strongly
52
15
37

6. Which best describes your impression of the Republican Party?

10/21/18

--------Favorable-------NET
Strongly
Somewhat
47
19
28

-------Unfavorable------NET
Somewhat
Strongly
53
15
38

On another subject…
7. How concerned are you that men you are close to might be unfairly accused of sexual
assault, if at all?

10/21/18

------Concerned-----NET
Very
Somewhat
57
30
26

------Less concerned-----NET
Not too
Not at all
43
21
22

8. How concerned are you that women in the U.S. are not believed when they report they
were sexually assaulted, if at all?

10/21/18

------Concerned-----NET
Very
Somewhat
78
45
32

------Less concerned-----NET
Not too
Not at all
22
16
7

9. Which do you think is the bigger problem in this country?

10/21/18

Men being unfairly
accused of sexual assault
41

Women not being believed
when reporting sexual assault
59

10. How often do you think men who commit sexual assault in the U.S. face serious
consequences?

10/21/18

Nearly all
of the time
3

Most of
the time
17

About half
of the time
25

Less than
half of the time
35

Almost
never
20

11. Lastly, did the debate over Brett Kavanaugh’s nomination to the Supreme Court make
you feel (more motivated to vote) this fall, (less motivated to vote), or did it not make
much of a difference?

10/21/18

More
Motivated
59

Less
motivated
2

Did not make
much difference
39

*** END ***
METHODOLOGICAL DETAILS
This poll was jointly sponsored and funded by The Washington Post and the Schar School
for Policy and Government at George Mason University. The poll is a random sample of

2

 adults in 69 congressional districts considered “tossup,” “lean Democrat” or “lean
Republican” by the Cook Political Report on Aug. 24, as well as additional districts
identified as competitive by Post staff.
This questionnaire was administered with the exact questions in the exact order as
they appear in this document. Demographic questions are not shown; a breakdown of
likely voter demographics is shown in the table below. If a question was asked of a
reduced base of the sample, a parenthetical preceding the question identifies the
group asked. Phrases surrounded by parentheticals within questions indicate clauses
that were randomly rotated for respondents.
Registration-based sampling was conducted within each of the 69 congressional districts
using voter file provided by L2, stratified by past turnout and whether voters recently
registered. Selected participants received letters in the mail with directions on how to
log onto an online portal to take the survey online or; voters without internet access
could take an automated survey interview by phone. The final sample included 1,425
interviews completed online (including cell phone/tablet) and 120 via automated phone.
This survey uses statistical weighting procedures to account for deviations in the survey
sample from known population characteristics in the congressional districts, which helps
correct for differential survey participation and random variation in samples. Base
weights adjusted for the proportion of voters who turned out in 2016 and 2014 general
elections, only 2016 or 2014, voters who recently registered and other voters. The
sample of registered voters was weighted to the share of selected districts won by
Clinton and Trump, as well as the sex and age makeup of voters in battleground
districts according to L2’s database of registered voters. The sample was weighted to
match the race/ethnicity and educational makeup of registered voters according to Post
analysis of the Census Bureau American Community Survey and Current Population Survey.
The sample was also weighted to match the share of voters who are registered as
Democrats and Republicans in states that allow registration by party. In states
without party registration, voter-file indicators of turnout in a recent Democratic
primary or Republican primary were used as according a weighting benchmark. Voters
whose record indicated they voted in 2016 were weighted to match the percentage of
voters who actually supported Clinton and Trump to final vote tallies in battleground
districts.
All error margins have been adjusted to account for the survey’s design effect, which
is 1.37 for the likely voter sample. The design effect is a factor representing the
survey’s deviation from a simple random sample and takes into account decreases in
precision due to sample design and weighting procedures. Surveys that do not
incorporate a design effect overstate their precision.
Table of likely voter demographics:
Male
Female
Age
Age
Age
Age
Age

18-29
30-39
40-49
50-64
65+

Likely voters
48%
51%
11%
10%
18%
31%
29%

White
Nonwhite
African American
Hispanic
Other

80%
19%
6%
8%
5%

No four-year college degree
High school or less

59%
26%

3

 Some college
College graduate
Bachelor’s degree
Post-grad degree

33%
41%
24%
16%

The Washington Post is a charter member of AAPOR’s Transparency Initiative, which
recognizes organizations that disclose key methodological details on the research they
produce.
Contact polls@washpost.com for further information about how The Washington Post
conducts polls.

4