Federal level · Current term
Hawaii
Hawaii’s congressional delegation
- statements analyzed
- 4,131
- officials covered
- 4
- last updated
- Oct 28, 2026
What members of Congress say in public
Floor speeches, press releases, newsletters and posts on X
We collect these statements and label each one as a personal attack, a criticism of policy, credit claiming, an appeal to bipartisanship or plain policy talk. The rules are the same for every member, so the numbers compare across members, parties and states.
What each party owns
This comparison needs both parties.
Each bar is the share of Democrats’ policy statements that name an area minus the share of Republicans’ that do, in percentage points, among 4 members of Hawaii’s congressional delegation in the current term. Bars to the right are areas Democrats name more often; bars to the left, areas Republicans name more often. The AI model tags each policy statement with any of the Congressional Research Service’s 25 policy areas. The figures beside each bar are both parties’ shares.
This comparison needs both parties. It is on the United States dashboard.
Who stands out
The five highest rates on four kinds of statement among 4 members of Hawaii’s congressional delegation. An official needs at least 50 statements in the current term to be ranked.
An AI model labels every statement. An official’s rate is the share of their statements that carry a label.
Most personal attacks
Overtly questions a named person’s character, integrity or loyalty, or calls them names. Criticising a record or a policy does not count.
- 1Mazie HironoDU.S. Senate · Hawaii18%
- 2Brian SchatzDU.S. Senate · Hawaii9.6%
- 3Jill TokudaDU.S. House · HI-022.6%
- 4Ed CaseDU.S. House · HI-010.7%
Most credit claiming
Claims credit for passing legislation, securing funding or grants, or other accomplishments in office.
- 1Ed CaseDU.S. House · HI-0148%
- 2Brian SchatzDU.S. Senate · Hawaii30%
- 3Jill TokudaDU.S. House · HI-0221%
- 4Mazie HironoDU.S. Senate · Hawaii16%
Most bipartisan language
Emphasises working across the aisle, compromise or common ground between the parties.
- 1Brian SchatzDU.S. Senate · Hawaii24%
- 2Ed CaseDU.S. House · HI-0120%
- 3Jill TokudaDU.S. House · HI-0210%
- 4Mazie HironoDU.S. Senate · Hawaii7.8%
Most policy-focused
Discusses a specific public policy or policy area (health care, taxes, defense, …).
- 1Mazie HironoDU.S. Senate · Hawaii81%
- 2Brian SchatzDU.S. Senate · Hawaii77%
- 3Ed CaseDU.S. House · HI-0173%
- 4Jill TokudaDU.S. House · HI-0273%
Each official’s rates are on their profile and in the bulk file.
Party comparison
Democrats talk about policy in 78% of statements; 10% attack opponents personally and 2.1% use extreme labels. The most-discussed policy area is Health (29% of policy statements).
Each bar is the share of a party’s statements that carry a label, pooled across 4 members of Hawaii’s congressional delegation in the current term. Officials who post more count more. A statement can carry several labels, so a party’s shares can sum to more than 100%.
Statements by label and party
Share of each party’s statements, the current term
- Personal attacks
- Policy criticism
- Policy discussion
- Bipartisanship
- Credit claiming
- Extreme labels
Show the data
| Measure | Democrats |
|---|---|
| Personal attacks | 10% |
| Policy criticism | 37% |
| Policy discussion | 78% |
| Bipartisanship | 14% |
| Credit claiming | 25% |
| Extreme labels | 2.1% |
The measures
Personal attacks. Overtly questions a named person’s character, integrity or loyalty, or calls them names. Criticising a record or a policy does not count.
Policy criticism. Objects to a specific policy, law or court ruling with fact-based arguments, without personal attacks or emotional appeals.
Policy discussion. Discusses a specific public policy or policy area (health care, taxes, defense, …).
Bipartisanship. Emphasises working across the aisle, compromise or common ground between the parties.
Credit claiming. Claims credit for passing legislation, securing funding or grants, or other accomplishments in office.
Extreme labels. Calls a politician, a party or a policy “far left”, “far right”, “MAGA”, “extreme”, “radical”, “woke”, “fascist” or “communist”.
Personal attacks by venue
Share of statements by 4 members of Hawaii’s congressional delegation, the current term
Democrats’ personal-attack rate runs from 4.1% of their newsletters to 13.9% of their floor speeches.
Show the data
| Venue | Democrats: personal attacks | Democrats: statements |
|---|---|---|
| Posts on X | 10.3% | 2766 |
| Press releases | 10.2% | 908 |
| Newsletters | 4.1% | 98 |
| Floor speeches | 13.9% | 359 |
Notes
Each point is the share of a party’s statements in one venue that the AI model labelled a personal attack, pooled across members. A line joins each party’s four venues. A statement counts once and carries the label if any of its passages does.
Posts on X are 67% of the 4,131 statements here, so the pooled rates elsewhere on this page sit closest to the X points.
A party needs 50 statements in a venue before its point is drawn.
Polarization
The median Democrat scores −1.12 on the CSIP scale.
CSIP, the lab’s Common-Space Ideal Point, places every official on one left–right scale estimated from roll-call votes, campaign contributions and the text of their statements. Negative scores are to the left and positive scores to the right. The gap is the distance between the two parties’ medians.
CSIP scores by party
← More liberal · 0 · More conservative →
Show the data
| Group | Officials | Median |
|---|---|---|
| Democrats | 4 | −1.1 |
Gap between the party medians
By election cycle
Not enough Congresses with scored members of both parties to draw a trend.
Notes
Each dot is one official, blue left of 0 and red right of 0, darker further from 0. Overlapping dots form darker patches where officials cluster. The rule in each row marks the party median.
Each official’s score is on their profile, and the full table is in Downloads.
Trends in what politicians say
Bipartisan language by Democrats moved most, from 17% of statements in Jan–Mar ’25 to 9.5% in Jul–Sep ’26.
Each strip is one label, with a row per party and a cell per month. Cells are shaded against the row’s own peak, and the peak and latest values are printed at the right. The sentence above each strip compares the first three complete months of the window with the last three. Party-months with fewer than 10 statements are hatched and excluded.
Monthly rates by party
Members of Congress, the current term · each row shaded against its own peak
Personal attacks
From Jan–Mar ’25 to Jul–Sep ’26, Democrats fell from 16% to 8.4%.
Show the data
| Column | Democrats |
|---|---|
| Jan ’25 | 17.3% |
| Feb ’25 | 17.9% |
| Mar ’25 | 12.2% |
| Apr ’25 | 9.7% |
| May ’25 | 10.8% |
| Jun ’25 | 6.5% |
| Jul ’25 | 4.6% |
| Aug ’25 | 6.3% |
| Sep ’25 | 12.4% |
| Oct ’25 | 11.4% |
| Nov ’25 | 14.2% |
| Dec ’25 | 10.0% |
| Jan ’26 | 12.3% |
| Feb ’26 | 9.8% |
| Mar ’26 | 8.8% |
| Apr ’26 | 8.9% |
| May ’26 | 11.4% |
| Jun ’26 | 9.6% |
| Jul ’26 | 9.8% |
| Aug ’26 | 10.0% |
| Sep ’26 | 5.4% |
| Oct ’26 | 7.2% |
Extreme labels
From Jan–Mar ’25 to Jul–Sep ’26, Democrats fell from 2.0% to 1.2%.
Show the data
| Column | Democrats |
|---|---|
| Jan ’25 | 1.0% |
| Feb ’25 | 4.3% |
| Mar ’25 | 0.8% |
| Apr ’25 | 2.3% |
| May ’25 | 0.9% |
| Jun ’25 | 1.9% |
| Jul ’25 | 1.7% |
| Aug ’25 | 3.6% |
| Sep ’25 | 5.1% |
| Oct ’25 | 2.7% |
| Nov ’25 | 1.1% |
| Dec ’25 | 1.1% |
| Jan ’26 | 2.5% |
| Feb ’26 | 2.3% |
| Mar ’26 | 1.8% |
| Apr ’26 | 0.6% |
| May ’26 | 1.4% |
| Jun ’26 | 4.8% |
| Jul ’26 | 2.8% |
| Aug ’26 | 0.0% |
| Sep ’26 | 0.8% |
| Oct ’26 | 0.0% |
Bipartisanship
From Jan–Mar ’25 to Jul–Sep ’26, Democrats fell from 17% to 9.5%.
Show the data
| Column | Democrats |
|---|---|
| Jan ’25 | 19.2% |
| Feb ’25 | 16.2% |
| Mar ’25 | 16.1% |
| Apr ’25 | 12.4% |
| May ’25 | 19.4% |
| Jun ’25 | 10.2% |
| Jul ’25 | 12.9% |
| Aug ’25 | 8.3% |
| Sep ’25 | 16.1% |
| Oct ’25 | 16.4% |
| Nov ’25 | 14.2% |
| Dec ’25 | 12.6% |
| Jan ’26 | 16.6% |
| Feb ’26 | 10.9% |
| Mar ’26 | 9.7% |
| Apr ’26 | 12.2% |
| May ’26 | 15.0% |
| Jun ’26 | 14.5% |
| Jul ’26 | 9.8% |
| Aug ’26 | 4.0% |
| Sep ’26 | 14.6% |
| Oct ’26 | 9.6% |
Policy discussion
From Jan–Mar ’25 to Jul–Sep ’26, Democrats held steady (75% to 80%).
Show the data
| Column | Democrats |
|---|---|
| Jan ’25 | 75.0% |
| Feb ’25 | 76.1% |
| Mar ’25 | 74.8% |
| Apr ’25 | 80.9% |
| May ’25 | 85.1% |
| Jun ’25 | 78.6% |
| Jul ’25 | 80.8% |
| Aug ’25 | 74.0% |
| Sep ’25 | 71.0% |
| Oct ’25 | 76.7% |
| Nov ’25 | 75.6% |
| Dec ’25 | 81.2% |
| Jan ’26 | 87.1% |
| Feb ’26 | 73.6% |
| Mar ’26 | 69.6% |
| Apr ’26 | 80.0% |
| May ’26 | 70.0% |
| Jun ’26 | 75.3% |
| Jul ’26 | 72.0% |
| Aug ’26 | 84.0% |
| Sep ’26 | 83.1% |
| Oct ’26 | 92.8% |
Notes
Oct ’26 is incomplete (data through Oct 28, 2026). It is drawn but excluded from every comparison.
No party-month here has fewer than ten statements.
Personal attacks by month
Members of Congress, Jan 2025 to Oct 2026 · Oct ’26 incomplete
Democrats’ personal attacks peaked in Feb 2025, at 17.9% of statements, the month after the inauguration. In Sep 2026, the last complete month, the rates were 5.4% for Democrats.
Show the data
| Date | Democrats |
|---|---|
| Oct 2026 | 7.2% |
| Sep 2026 | 5.4% |
| Aug 2026 | 10.0% |
| Jul 2026 | 9.8% |
| Jun 2026 | 9.6% |
| May 2026 | 11.4% |
| Apr 2026 | 8.9% |
| Mar 2026 | 8.8% |
| Feb 2026 | 9.8% |
| Jan 2026 | 12.3% |
| Dec 2025 | 10.0% |
| Nov 2025 | 14.2% |
| Oct 2025 | 11.4% |
| Sep 2025 | 12.4% |
| Aug 2025 | 6.3% |
| Jul 2025 | 4.6% |
| Jun 2025 | 6.5% |
| May 2025 | 10.8% |
| Apr 2025 | 9.7% |
| Mar 2025 | 12.2% |
| Feb 2025 | 17.9% |
| Jan 2025 | 17.3% |
- 1Jan 20, 2025 Inauguration · calendar
- 2Apr 13, 2025 Arson at the Pennsylvania governor’s residence · lab violence record
- 3Jun 14, 2025 Two Minnesota legislators and their spouses shot at their homes · lab violence record
- 4Sep 10, 2025 Charlie Kirk shot in Orem, Utah · lab violence record
Notes
Each line is the monthly share of a party’s statements that the AI model labelled a personal attack, pooled across officials. The numbered rules mark acts of political violence from the lab’s record and fixed dates on the political calendar. They date events and make no claim about cause.
Party-months with fewer than 10 statements are omitted and break the line. Oct ’26 is incomplete; it is drawn but never counted as a peak.
Policy topics
Emergency Management led in Sep ’26, named in 27% of policy statements. Science, Technology, Communications rose most from Aug ’26 (+12 pts); Economics and Public Finance fell most (−9.0 pts).
The AI model tags each policy statement with any of 25 policy areas, so shares can sum to more than 100%. A share is the percentage of a month’s policy statements by members of Congress that name the area. The lists compare the last two complete months, with changes in percentage points. A month needs at least ten policy statements to be compared.
Most discussed, Sep ’26
108 policy statements; change since Aug ’26 in grey
- Emergency Management27% −0.5 pts
- Government Operations and Politics22% +3.2 pts
- Armed Forces and National Security21% −3.7 pts
- Education19% +8.7 pts
- Science, Technology, Communications18% +12 pts
Rising
Up most since Aug ’26
- Science, Technology, Communications+12 pts
- Education+8.7 pts
- Labor and Employment+5.0 pts
Falling
Down most since Aug ’26
- Economics and Public Finance−9.0 pts
- Health−6.8 pts
- Agriculture and Food−6.6 pts
States
The state maps are on the United States dashboard.
The left map is the share of statements by each state’s congressional delegation that the AI model labelled a personal attack, the current term (members of Congress, by the state they represent). The right map is the lab’s survey estimate of affective polarization in each state: residents’ rating of their own party minus their rating of the other, on a 0–100 thermometer. Each state is a square of the same size, shaded by quintile; darker is higher.
The two maps are on the United States dashboard.
Challengers and incumbents
No race in this scope has an incumbent and a challenger with 10 or more posts on X.
Every 2026 House and Senate race with an incumbent running and at least one challenger. A candidate’s rate is the share of their posts on X that the AI model labelled a personal attack. Both sides are measured over the same months, since Oct 1, 2025, and incumbents’ press releases, newsletters and floor speeches are excluded. Candidates with fewer than 10 posts are left out.
No race in this scope has both an incumbent and a challenger with 10 or more posts on X.