Independent · Non-partisan · Methods disclosed

Rigorous polling, clearly disclosed.

American Polling Company is a research firm for political and public-opinion work, built on disclosed methods, honest margins, and the discipline to flag what isn’t ready to report.

estimate ± margin of error95% confidence
100%of our methods, disclosed publicly
0undisclosed sponsors, ever
95%confidence, stated on every estimate
±MoEattached to every number we publish

How we work

Methodology you can inspect, not just trust.

Polling is only as good as the assumptions behind it. We make ours explicit, so a reader can judge the numbers on their merits.

  1. 01

    We show our work

    Every release ships with its full crosstabs, question wording, and weighting targets. If you want to check us, you can.

  2. 02

    We disclose the method

    Mode, frame, field dates, and response rate are stated up front, not on request, in line with professional survey-research transparency norms.

  3. 03

    We flag what isn't ready

    Low-n subgroups and low-confidence estimates are marked directional and not for publication. Honest uncertainty beats a false decimal.

  4. 04

    We report undecideds

    We don't allocate away the people who haven't decided. The margin includes them, because the electorate does.

  5. 05

    We respect the people we contact

    Respondent identity is never mixed into published results, and an opt-out is honored permanently and across every future study.

Services

Built for whoever needs the truth.

  • Political

    Campaign & party research

    Ballot tests, message and issue testing, and tracking, fielded to the same standard we would defend in public.

    • Benchmark & tracking polls
    • Message & ad testing
  • Public opinion

    Neutral public research

    Foundations, universities, and agencies who need defensible reads on where the public actually stands.

    • Attitudinal & policy studies
    • Longitudinal panels
  • Media

    Media & partner polls

    Co-branded surveys built for publication, with disclosure language and toplines newsrooms can run as-is.

    • Election-cycle polling
    • Data journalism support

The format we ship

What a release looks like.

⚠ Illustrative sample · not actual releases
Topline · horizontal bar with margin of errorwhisker = 95% confidence interval
Candidate A47% ±2.8
Candidate B44% ±2.8
Undecided9% ±1.7

Methodology

Sample
n = 1,214 likely voters
Margin of error
±2.8 pts at 95%
Mode
SMS + web, probability-based
Status
Illustrative sample

See the full sample report format →

Commission a poll

Have a question worth measuring?

Tell us the decision you’re trying to make. We’ll scope a design, disclose the method up front, and tell you honestly what a sample this size can and can’t say.

By submitting, you agree to our Privacy Policy. This is a business-inquiry form; it does not subscribe you to text messages.