The 2018 Primaries Project: Appendix to voters data and acknowledgements

Elaine Kamarck and Alexander R. Podkul
alex podkul
Alexander R. Podkul Ph.D. Candidate in Political Science - Georgetown University

October 23, 2018

Editor's note:

The following is the appendix to “Political polarization and voters in the 2018 congressional primaries,” a report from The Primaries Project at Brookings.

Exit polls methodological statement from Edison Research

Survey methodology

Edison Research conducted exit polls for the 2018 state primaries in 20 U.S. congressional districts. Exit polls were conducted on Election Day, on the day of the U.S. House primary for that district.

Interviews were conducted at a randomly-selected sample of eight polling places in each of the twenty selected Congressional Districts for a total of 160 polling locations nationwide. A total of 13,372 interviews on primary voters were conducted: 7,198 among Democratic primary voters, 5,335 among Republican primary voters and 839 among voters in California who voted in a combined “Top-Two” primary election.

Interviewers were stationed at each sample polling location during all polling hours on Election Day. The interviewers attempted to invite every voter to participate in the survey as they left the polling location, after they had voted. Voters who were willing to participate were asked whether they had voted in the Democratic or Republican primary and were handed a short self-administered questionnaire. There was a different version of the questionnaire depending on whether the voter voted in the Democratic or Republican primary. Voters filled out the questionnaire themselves and placed the completed questionnaire in a ballot box outside the polling place. Voters were assured that their survey responses were private, confidential and anonymous.

Interviewers also noted the gender, approximate age and race of all voters who were approached but declined to complete the questionnaire. This allowed us to calculate the overall response rate and the response rate by each demographic group to determine any effect caused by differential non-response. The overall response rate was 33%.

The margin of error on the Republican dataset (excluding FL CD 10) is +/- 2. The margin of error on the Democratic dataset is +/- 1. The margin of error for the two California Congressional Districts (which had a “Top-Two” primary) is +/- 4.

The data for each individual congressional district primary was tabulated separately and weighted to match the final results of the U.S. House primary election in each district. Each congressional district survey was also weighted to account for the differential non-response by age, race and gender. The final combined dataset was weighted to reflect the number of voters for each congressional district primary.

How congressional district samples were selected

Edison Research compiled a list of all of the U.S. House congressional district primaries in 2018. All congressional districts where there were at least two candidates in either the Democratic or Republican primary were eligible for sample selection. Congressional districts with at least two candidates in states with “Top-Two” primary systems like California were also eligible for sample selection. All interviews were conducted with election day voters so congressional districts in states with all by-mail voting (Colorado, Oregon and Washington) were excluded. The congressional districts were selected randomly proportional to the size of the vote in the most recent contested primary in each congressional district. Within each congressional district sample locations were selected randomly, stratified by past presidential vote and proportional to the size of the vote in the previous presidential election.

Among the twenty congressional districts, thirteen had contested U.S. House primaries for both the Democratic and Republican Parties, five had contested Democratic primaries only and two had contested Republican primaries only.

Census comparisons

District averages come from the 2017 American Community Survey (ACS) 1-Year Estimates provided by the U.S. Census Bureau and were downloaded via the American Fact Finder, available on the Census website. Congressional district-level estimates were gathered for the sampled districts in exit polls and averaged together, weighted by the voting age population within each sampled district. A list of tables accessed used for extraction is available upon request.


  1. Are you:
    1. Male
    2. Female
  2. To which age group do you belong?
    1. 18-24
    2. 25-29
    3. 30-39
    4. 40-44
    5. 45-49
    6. 50-59
    7. 60-64
    8. 65-74
    9. 74 or over
  3. Are you:
    1. White
    2. Black
    3. Hispanic/Latino
    4. Asian
    5. American Indian
    6. Other
  4. In today’s election for U.S. House of Representatives, did you just vote for:?
  5. Which of these five candidate qualities mattered the most in deciding how you voted for U.S. House of Representatives?
    1. Shares my values
    2. Can bring about needed change
    3. Has the right experience
    4. Cares about people like me
    5. Can defeat the other party’s candidate in November
  6. How closely did you follow the election for U.S. House of Representatives in your district?
    1. Extremely closely
    2. Very closely
    3. Somewhat closely
    4. Not too closely
  7. What election were you most concerned about voting in today:
  8. Was one reason for your vote today:
    1. To express support for Donald Trump
    2. To express opposition to Donald Trump
    3. Donald Trump was not a factor
  9. What was the last grade of school you completed?
    1. Did not complete high school
    2. High school graduate
    3. Some college or associate degree
    4. College graduate
    5. Postgraduate study
  10. No matter how you voted today, do you think of yourself as a:
    1. Strong Democrat
    2. Not Strong Democrat
    3. Independent
    4. Not Strong Republican
    5. Strong Republican
    6. Something else
  11. On most political matters, do you consider yourself:
    1. Very liberal
    2. Somewhat liberal
    3. Moderate
    4. Somewhat conservative
    5. Very conservative
  12. Do you think you are more liberal or more conservative than most of the general election voters in your Congressional district?
    1. More liberal
    2. More conservative
  13. Do you or does someone in your household belong to a labor union? (ONLY ON DEMOCRATIC QUESTIONNAIRE)
    1. Yes, I do
    2. Yes, someone else does
    3. Yes, I do and someone else does
    4. No one does
  14. Do you think the condition of the nation’s economy is:
    1. Excellent
    2. Good
    3. Not so good
    4. Poor
  15. Who do you think deserves more credit for the country’s economic situation?
    1. President Obama
    2. President Trump
    3. Neither
  16. Overall, would you say trade with other countries:
    1. Creates more U.S. jobs
    2. Takes away U.S. jobs
    3. Has no effect on U.S. jobs
  17. Should most illegal immigrants working the United States be:
    1. Offered a chance to apply for legal status
    2. Deported to the country they came from
  18. Do you think the use of marijuana should be:
    1. Legal and regulated
    2. Illegal
  19. In conducting foreign policy, should the U.S.:
    1. Take into account the interests of its allies
    2. Only follow its own national interests
  20. In the 2016 _______ presidential primary, did you vote for:
  21. If the Republican primary for president in 20202 were held today, whom would you support? (ONLY ON REPUBLICAN AND CALIFORNIA QUESTIONNAIRE)
    1. Donald Trump
    2. Another Republican candidate
  22. In the 2016 election for president, did you vote for:
    1. Hillary Clinton
    2. Donald Trump
    3. Someone else
    4. Did not vote
  23. Is your opinion of Nancy Pelosi/Paul Ryan? (VARIES BY PARTY)
    1. Strongly favorable
    2. Somewhat favorable
    3. Somewhat unfavorable
    4. Strongly unfavorable
  24. Is your opinion of Chuck Schumer/Mitch McConnell? (VARIES BY PARTY)
    1. Strongly favorable
    2. Somewhat favorable
    3. Somewhat unfavorable
    4. Strongly unfavorable
  25. How do you feel about the way Donald Trump is handling his job as president?
    1. Strongly approve
    2. Somewhat approve
    3. Somewhat disapprove
    4. Strongly disapprove
  26. Did anyone call or talk to you in person on behalf of a campaign about coming out to vote?
    1. Yes
    2. No
  27. How often do you attend religious services?
    1. More than once a week
    2. Once a week
    3. A few times a month
    4. A few times a year
    5. Never
  28. Would you describe yourself as a born-again or evangelical Christian?
    1. Yes
    2. No
  29. 2017 total family income:
    1. Under $15,000
    2. $15,000 – $29,999
    3. $30,000 – $49,999
    4. $50,000 – $74,999
    5. $75,000 – $99,999
    6. $100,000 – $149,999
    7. $150,000 – $199,999
    8. $200,000 or more
  30. If today’s primary was not held under a top-two primary system, would you have chosen to vote in the: (ONLY ON CALIFORNIA QUESTIONNAIRE)
    1. Democratic primary
    2. Republican primary
  31. Top-two primaries eliminate the need for registering for a political party in order to vote. Which statement most accurately reflects your opinion? (ONLY ON CALIFORNIA QUESTIONNAIRE)
    1. I prefer not having to register for a political party, and voting from one list of candidates
    2. I prefer voting in my own party’s primary
    3. It doesn’t matter to me how primary elections are held


Thanks to Christine Stenglein and Nick Zeppos for their research assistance.

The Brookings Institution is a nonprofit organization devoted to independent research and policy solutions. Its mission is to conduct high-quality, independent research and, based on that research, to provide innovative, practical recommendations for policymakers and the public. The conclusions and recommendations of any Brookings publication are solely those of its author(s), and do not reflect the views of the Institution, its management, or its other scholars.

Brookings gratefully acknowledges the program support provided to Governance Studies by Democracy Fund and the William and Flora Hewlett Foundation.

Brookings recognizes that the value it provides is in its absolute commitment to quality, independence, and impact. Activities supported by its donors reflect this commitment.


  • Footnotes
    1. Unfortunately, due to an issue with the exit polls questionnaire, the second wave (of four) accidentally included six responses. In the analyses above, the sixth response—which was “Agrees with you on major issues”—was collapsed and included in “Shares my values.” Tables with these responses broken out are available on request.