Working with a city

Municipal work has its own procurement, its own approval chain and its own question about who counts as the public. Commissioning a community survey covers the decisions a city actually has to make: the registered-voter frame versus the all-resident frame, sample size against district cuts, languages, and what arrives at the end.

The Pembroke Pines case study is the worked example: a two-wave program, 3,569 responses from 2,876 unique residents, and both dashboards published with the sample, the frame and the error band beside the numbers.

A method you can check

A number without its method is an opinion about an opinion. Every study we publish states the universe it was drawn from, the field dates, the sample size, how the data were weighted, and the error band that follows from all of it. We show figures weighted and unweighted, because a reader who disagrees with our weighting is entitled to see what the raw data said.

In the 2026 Pembroke Pines Community Survey we weighted to the city’s active registered voters by age, gender, race, party, and district, against a universe of 100,067 registered voters.

Representative means multilingual

In South Florida, an English-only instrument is not a representative one. We field multilingual surveys because leaving out the residents who do not answer in English does not just shrink a sample, it biases it in a direction you cannot correct for afterward with weights.

We field in English, Spanish, Haitian Creole, and French. Which of those a given study uses depends on the jurisdiction it covers.

Research that changes a decision, not just a deck

The most useful question a survey can answer is what happens when people learn more. In Wave 2 of the Pembroke Pines survey we asked about a proposed $80 million public-safety bond twice. Asked cold, support was 27% yes to 37% no. Asked again after a short factual briefing, it was 42% yes to 34% no, an 18-point swing in the margin.

Evan Ross presented both waves to the City Commission on August 5, 2026. Later in that same meeting, the motion to place the bond on the November ballot died for lack of a second. The full results are published as a live public dashboard in our research index of published studies.

Who we field for

Governments, nonprofits, advocacy organizations, and corporations. Community and resident surveys for cities, message and issue testing for advocacy campaigns, and membership or audience research for nonprofits and associations.

A study appears in our published research only after the client clears it for release. Most of what we field is never published, which is normal for research commissioned to inform a decision rather than to make an announcement.

Common questions

What is the difference between this page and the research index?

This page describes how we conduct public opinion research as a service. The research index at /research is the library of studies we have published, each with its full dashboard, sample, weighting, and error band.

Do you publish every survey you conduct?

No. A study appears publicly only after the client clears it for release. Most research we field is commissioned to inform a decision and is never published.

Do you survey in languages other than English?

Yes. We field in English, Spanish, Haitian Creole, and French. In South Florida an English-only instrument produces a biased sample rather than merely a smaller one, so the language set is chosen to match the jurisdiction.

How large are your samples?

It depends on the jurisdiction and the precision required. Wave 1 of the 2026 Pembroke Pines Community Survey was n=1,933 with a modeled band of plus or minus 2.9 points; Wave 2 was n=1,636 at plus or minus 3.4 points.