Integrity Polling
Releases Horse race

Demonstration release: a statewide gubernatorial ballot test

A worked example of how Integrity Polling reports a horse-race poll, with every ballot-qualified candidate named.

By Integrity Polling research staff

Sample data. This release is a demonstration of our reporting template. No survey was fielded. Every number on this page is illustrative.

Sample size
n = 1,012
Likely voters
Field dates
Sep 14–18, 2026
Margin of error
±3.4 pts
95% confidence, design effect included
Mode
Mixed mode
Full mode mix in methodology

Topline results

Demo data

If the election for governor were held today, for whom would you vote?

Likely voters, n = 1,012. Candidate order rotated. All four ballot-qualified candidates read by name.

Whiskers show ±3.4 percentage points, the margin of sampling error.

Integrity Polling, Sep 14–18, 2026. n = 1,012 likely voters. Question Q3.

Demo data

Candidate favorability

Likely voters. Rows sum to 100 across favorable, unfavorable, and never heard / no opinion.

Integrity Polling, Sep 14–18, 2026. n = 1,012 likely voters. Question Q5–Q8.

Demo data

Most important issue facing the state

Likely voters. One response accepted; list rotated.

Integrity Polling, Sep 14–18, 2026. n = 1,012 likely voters. Question Q9.

Key findings

The race is close, and we say so. Delgado’s four-point edge among likely voters is only slightly larger than the poll’s ±3.4-point margin of sampling error, which applies to each candidate’s share. The margin on the difference between two candidates is roughly twice that. The honest summary is “Delgado narrowly ahead, within the range where a different sample could show a tie,” not “Delgado leads.”

Nine percent of likely voters chose a candidate other than the two major-party nominees. Alicia Park’s 7 percent is more than the gap between the leaders. A poll that offered only Delgado and Reed and bucketed everyone else as “other” would have hidden the single most consequential fact about this race: the outcome likely depends on where Park’s voters land, or whether they stay with her.

Independents are the swing, and they are not undecided so much as unaligned. Among independent and other registrants, 15 percent back Park and 10 percent are undecided. Both candidates have more room to grow with this group than with their own partisans, who are already north of 85 percent consolidated.

The geography is familiar. Delgado leads by 26 in metro counties, the two are close in the suburbs, and Reed leads by 28 in rural counties. Nothing in the regional pattern is surprising, which is one reason we flag it: when the geography of a poll looks like the geography of the last election, the sample is at least plausible.

What to notice about this release

This page exists to show the template, not to inform a real campaign. Everything we promise in the Integrity Standard is on this page:

  • The methodology box lists field dates, sample size, mode mix, sampling frame, weighting targets, response rate, and margin of error, with the design effect included.
  • Every question appears below exactly as it was read, in the order it was asked, including the screener that defines “likely voter.”
  • Every ballot-qualified candidate is named in the question and in the chart. Nobody is “other” unless they are not on the ballot.
  • The crosstabs are downloadable as a CSV with unweighted sample sizes for every column.
  • After the election this poll would be graded on the Scorecard, with the signed error on the Delgado–Reed margin and on each minor-party share.

The headline question broken out by group. The CSV includes every question and the unweighted n for every column.

Gubernatorial ballot test by party registration. Column percentages.
ResponseDemocratRepublicanIndependent / other
Maria Delgado (D)89%5%38%
Thomas Reed (R)4%86%31%
Alicia Park (L)2%5%15%
Devon Okafor (G)3%0%4%
Another candidate0%1%2%
Undecided2%3%10%
Unweighted n368351293

Bold marks the largest share in each row. Columns may not sum to 100 because of rounding. The overall margin of error is ±3.4 points; subgroup margins are larger and grow as the unweighted n shrinks.

Question wording, in the order asked

Exactly as respondents heard or read it. Bracketed text is interviewer instruction. Rotations are noted where they occurred.

  1. S1

    Are you currently registered to vote at your current address?

    • Yes
    • No
    • Not sure

    Respondents answering No or Not sure were thanked and the interview ended.

  2. S2

    How likely are you to vote in the November general election for governor? Would you say you will definitely vote, probably vote, that the chances are about 50-50, that you probably will not vote, or that you definitely will not vote?

    • Definitely vote
    • Probably vote
    • 50-50
    • Probably will not vote
    • Definitely will not vote

    Likely voters are respondents who said they will definitely or probably vote, plus 50-50 respondents with a record of voting in two of the last three general elections.

  3. Q3

    If the election for governor were held today and the candidates were [ROTATE: Maria Delgado, the Democrat; Thomas Reed, the Republican; Alicia Park, the Libertarian; Devon Okafor, the Green Party candidate], for whom would you vote?

    • Maria Delgado
    • Thomas Reed
    • Alicia Park
    • Devon Okafor
    • Another candidate (accepted, not read)
    • Undecided (accepted, not read)

    Every candidate who had qualified for the general-election ballot as of the first day of fielding was read by name, in rotated order. Undecided respondents were not asked a leaner question; we report the undecided share as is.

  4. Q4

    How certain are you of your choice? Would you say you are certain, or that you might change your mind before the election?

    • Certain
    • Might change my mind
    • Not sure

    Asked of respondents who named a candidate in Q3.

  5. Q5

    Do you have a favorable or unfavorable opinion of Maria Delgado, or have you never heard of her?

    • Favorable
    • Unfavorable
    • Never heard of
    • Heard of, no opinion
  6. Q6

    Do you have a favorable or unfavorable opinion of Thomas Reed, or have you never heard of him?

    • Favorable
    • Unfavorable
    • Never heard of
    • Heard of, no opinion
  7. Q7

    Do you have a favorable or unfavorable opinion of Alicia Park, or have you never heard of her?

    • Favorable
    • Unfavorable
    • Never heard of
    • Heard of, no opinion
  8. Q8

    Do you have a favorable or unfavorable opinion of Devon Okafor, or have you never heard of him?

    • Favorable
    • Unfavorable
    • Never heard of
    • Heard of, no opinion

    Q5–Q8 asked in the same rotated order as the ballot test.

  9. Q9

    Which one of the following is the most important issue facing the state today? [ROTATE LIST]

    • Cost of living
    • Housing
    • Public safety
    • Education
    • Health care
    • Immigration
    • Something else (specify)
  10. D1–D7

    Demographic items: age, gender, race and ethnicity, highest level of education, county of residence, party registration confirmation, and 2024 presidential vote. Exact wording is in the questionnaire download.

Methodology

How we poll, in full
Field dates
Sep 14–18, 2026
Sample size
n = 1,012
Population
Likely voters (screened from 1,418 registered voters), statewide
Mode
Mixed mode: text-to-web (58%), live telephone to cell and landline (31%), probability-based online panel (11%)
Sampling frame
Statewide registered-voter file, stratified by region, age, party registration and vote history
Weighting targets
Age · Sex · Race/ethnicity · Education · Region · Party registration · 2024 presidential vote recall
Margin of sampling error
±3.4 percentage points at 95% confidence. Includes a design effect of 1.21. Subgroup margins are larger and are noted beneath each crosstab.
Response rate
AAPOR RR3 3.9% (telephone); 12.4% completion rate (text-to-web)
Languages
English and Spanish
Sponsor / funding
Integrity Polling. No outside funding, no candidate or party involvement.

Percentages may not sum to 100 because of rounding. The margin of error for subgroups is larger than the overall margin. In addition to sampling error, question wording, order and practical difficulties in fielding surveys can introduce error or bias; we publish every question in the order asked so you can judge for yourself.

Share X Bluesky LinkedIn Email
Embed the headline chart

Paste this into your page. The frame stays current if we issue a correction, and always links back to the full release.

<iframe src="https://integritypolling.com/embed/sample-statewide-governor-ballot-test/" title="Demonstration release: a statewide gubernatorial ballot test — Integrity Polling" width="100%" height="440" style="border:1px solid #E4E0EA;border-radius:8px" loading="lazy"></iframe>

More releases

All releases
  • Tracker wave

    Demonstration release: Trust Tracker pilot wave

    In this illustrative pilot, the Institutional Trust Index reads 41 on a 0–100 scale and the Representation Index reads 31. Local election officials are the most trusted institution tested; Congress the least. Thirty-eight percent of adults say neither major party represents them.

    41 Institutional Trust Index, 0–100 (pilot, illustrative)
    • Sample data
    • Trust Tracker
    • Institutions
    • Representation
  • Issue poll

    Demonstration release: Americans on ranked-choice voting and ballot access

    After hearing a neutral description, 58 percent of U.S. adults support letting voters rank candidates in order of preference. Nearly two-thirds say independent and third-party candidates should face the same signature requirements as major-party nominees.

    58% support ranked-choice voting after hearing it described
    • Sample data
    • Elections
    • Ranked-choice voting
    • Ballot access