Measure the race before you bet the budget on it.
American Polling and Research fields fast, modern surveys for campaigns, institutions, and newsrooms. Text-to-web sampling, likely-voter models built from real turnout history, weighted results in days. Every number ships with its margin and its method.
- Text-to-web sampling
- Weighted in days, not weeks
- Margin and universe per study
- Full crosstabs in every release
Services
Know what to say, where you stand, and who will vote.
Every study is built around a decision you are trying to make, and reported so you can act on it with your eyes open.
Message
Message & ad testing
Know which message actually moves your voters, and which falls flat, before you spend a dollar on media.
- Message & ad testing
- Issue & framing tests
- Ballot tests
Position
Benchmark & tracking
See where you stand at the outset, then watch the number move as the campaign goes on.
- Benchmark surveys
- Brushfire & tracking waves
Opposition
Incumbent-vulnerability reads
Find out whether an incumbent can actually be beaten, and where the opening is.
- Approval & re-elect reads
- Head-to-head ballot tests
Turnout
Likely-voter modeling
Weight to the electorate that will actually show up, modeled from real turnout history.
- Turnout-based LV models
- Universe design & screening
How an engagement works
From your question to a briefing you can act on.
- 01
Scope
We pin down the decision, the universe, and the questions, and put the method in writing before anyone fields.
- 02
Field
We field by text-to-web to a probability sample, watching response and quality as it comes in.
- 03
Weight
We weight to the likely electorate and mark any estimate too thin to trust.
- 04
Brief
You get a topline, the full crosstabs, and a plain read of what it means for your call.
Why us
Numbers you can hand a donor or a reporter.
We report undecided as its own line, because that is where the electorate actually is.
Modern platform
A multi-mode text-to-web platform reaches voters on the phone in their hand, including the ones a landline sample never will.
Fast enough to matter
Fielded and weighted in days, so the number lands while the decision is still open.
No black box
Every release states its margin of error and its likely-voter universe. You can see exactly how we got there.
Holds up to a reporter
Crosstabs, question wording, and field dates travel with the topline, so a number stands up when someone checks it.
The format we ship
What a release looks like.
Scope a poll
Have a question worth measuring?
Tell us the decision you’re trying to make. We’ll scope the design, put the method in writing, and tell you what a sample that size can and cannot say before you commit.