Independent survey research
An accurate account of what the public thinks.
Wit Research designs and conducts surveys of public opinion for institutions that need to understand it precisely. We measure carefully, report plainly, and publish the methods behind every number.
Coverage
What we measure
Politics and governance
Vote intention, approval, issue priorities and attitudes toward government, measured on a consistent schedule so change can be read with confidence.
Economy and households
Consumer sentiment, financial expectations and cost-of-living pressure, broken out by the groups that experience them differently.
Society and institutions
Trust, civic life, community and the values that shape them. Long-running series, asked the same way each time.
Reputation and policy
How the public regards organizations, sectors and proposals, including the arguments that move opinion and the ones that do not.
Sample findings
How we report results
The studies below are illustrative. Their figures are sample data, prepared to show the form our reporting takes: a clear finding, the question as asked, and the sample, dates and margin of error beside every chart. They are not results from actual surveys.
01/Institutions
Illustrative sample data
Confidence in national institutions remains uneven
Local institutions continue to command more confidence than national ones. Small-business owners and the military lead; Congress and national news organizations trail by a wide margin.
| Institution | 2026 (%) | 2025 (%) |
|---|---|---|
| Small businesses | 68 | 70 |
| The military | 61 | 63 |
| Local government | 47 | 44 |
| Public schools | 39 | 38 |
| The Supreme Court | 33 | 36 |
| Large corporations | 22 | 21 |
| National news organizations | 18 | 20 |
| Congress | 11 | 12 |
- Population
- U.S. adults
- Sample
- n = 2,412
- Fieldwork
- Sept. 8–21, 2026
- Margin of error
- ±2.3 points
- Mode
- Probability-based online panel
02/Economy
Illustrative sample data
Younger adults are the least optimistic about the year ahead
Expectations for personal finances divide sharply by age. Adults under 30 are nearly twice as likely as those 65 and older to expect their situation to worsen.
| Group | Better (%) | About the same (%) | Worse (%) |
|---|---|---|---|
| All adults | 31 | 42 | 27 |
| Ages 18–29 | 29 | 34 | 37 |
| Ages 30–49 | 35 | 38 | 27 |
| Ages 50–64 | 30 | 46 | 24 |
| Ages 65+ | 24 | 57 | 19 |
- Population
- U.S. adults
- Sample
- n = 3,086
- Fieldwork
- Oct. 1–12, 2026
- Margin of error
- ±2.0 points (total sample)
- Mode
- Mixed mode: online panel and live telephone
03/Tracking
Illustrative sample data
Sentiment about the country’s direction, eight quarters
The share saying the country is headed in the right direction has moved within a narrow band, rising modestly through mid-2026 before easing this autumn.
| Quarter | Right direction (%) |
|---|---|
| Q4 ’24 | 27 |
| Q1 ’25 | 29 |
| Q2 ’25 | 28 |
| Q3 ’25 | 31 |
| Q4 ’25 | 33 |
| Q1 ’26 | 34 |
| Q2 ’26 | 36 |
| Q3 ’26 | 32 |
- Population
- U.S. adults
- Sample
- n = ≈2,000 per wave
- Fieldwork
- Quarterly, Q4 2024 – Q3 2026
- Margin of error
- ±2.2 to ±2.6 points per wave
- Mode
- Probability-based online panel
Methodology
Standards we hold to
A survey is only as useful as the confidence one can place in it. Our methods are documented, consistent across waves, and open to examination by anyone who reads our work.
- Typical national margin of error
- ±2–3pts
- Adults in a standard national study
- 2,000+
- Of questionnaires published with results
- 100%
01
Sampling
Probability-based samples drawn from address-based and random-digit frames, supplemented by vetted online panels where the design calls for it.
02
Weighting
Results are weighted to current Census benchmarks for age, sex, education, race and ethnicity, region and, where relevant, past vote.
03
Uncertainty
Every margin of error we report accounts for the design effect of weighting. Subgroup estimates carry their own, wider, intervals.
04
Disclosure
Each release states its sponsor, population, field dates, mode, sample size, weighting variables, margin of error and full question wording.
Clients
Who we work with
We conduct custom studies and long-term tracking programs for organizations whose decisions depend on an honest reading of public sentiment. Engagements are confidential unless a client elects to publish.
- Public agencies
- Foundations and nonprofits
- News organizations
- Universities and think tanks
- Corporate leadership
- Associations and coalitions