Signature index · Quarterly · First wave Q1 2027
The Trust Tracker
Two numbers, published four times a year, built the same way every time: how much Americans trust the institutions that govern them, and whether they feel represented by anyone at all.
Get an email when each wave publishes
Four emails a year, each with the two indices, every component, and the data file.
Trust and representation, 0–100
Demonstration series. No wave has been fielded; these values exist only to design the chart. First wave fielding Q1 2027.
Each point is a quarterly wave of about 2,000 U.S. adults on a probability-based panel. Standard error of each index about 0.9 points. See the demonstration pilot release for the full template.
What it measures
Trust, and the feeling of being represented
Most trust-in-institutions series ask about one institution at a time and are published when a client pays for them. Ours is a fixed battery, self-funded, on a fixed calendar.
Index 1
Institutional Trust Index
The mean of nine 0–100 trust scores. The nine institutions:
- Congress
- The Supreme Court
- The presidency
- Federal agencies
- State government
- Local government
- Local election officials
- The news media
- Public opinion polls
Yes, polls are on the list. We publish our own industry’s number every quarter.
Index 2
Representation Index
The share of adults who say the Democratic or Republican Party “represents people like me,” weighted by how strongly they feel it. People who choose “neither” score zero.
It is the mirror image of the “neither” share, which in most horse-race polling is visible only as undecided or other. Here it is the headline.
Every wave also reports the raw three-way split by party, age, region and education in the crosstab file.
How it is constructed
Five rules we will not bend
- 01
One battery, nine institutions, every quarter
Respondents rate how much they trust each institution to act in the public interest on a five-point scale, from “none at all” to “a great deal.” The nine are rotated. The wording never changes between waves, because a tracker that edits its question is two trackers.
- 02
Each institution is scored 0–100
None at all = 0, very little = 25, some = 50, quite a lot = 75, a great deal = 100. The institution score is the weighted mean. We publish all nine so you can see which institutions moved the composite.
- 03
The Institutional Trust Index is the plain average
No institution is weighted more than another. A simple mean is easy to rebuild from the crosstab file and hard to accuse of tuning. If we ever change the component list, the old index continues alongside the new one for four waves.
- 04
The Representation Index measures the “neither” problem directly
We ask which party, if either, represents people like you, then how strongly you feel it. Feeling represented very strongly scores 100, somewhat 67, not that strongly 33; “neither” and “not sure” score 0. The index is the mean.
- 05
We report the uncertainty every wave
Each index carries a standard error (about 0.9 points at n ≈ 2,000). A quarter-to-quarter change smaller than twice that is reported as “no significant change,” in those words, even when the headline would be better with a trend.
Wave schedule
Fixed calendar
Fielding windows are set a year ahead and published here. We do not move a wave to catch or avoid news.
Email alerts
Wave alerts
An email the morning each wave publishes.
| Wave | Quarter | Fielding window | Status |
|---|---|---|---|
| Pilot | Q4 2026 | Instrument testing, n ≈ 600 | In design |
| Wave 1 | Q1 2027 | Late January – early February | Scheduled |
| Wave 2 | Q2 2027 | Late April – early May | Scheduled |
| Wave 3 | Q3 2027 | Late July – early August | Scheduled |
| Wave 4 | Q4 2027 | Late October – early November | Scheduled |
Each wave publishes within ten business days of the close of fielding, with the full release template: toplines, crosstabs, verbatim questions, methodology, CSV.