Do Polls Overlook Republican Voters?

 Many posts have discussed pollsters' challenges, including the plunge in response rates.

Erica Pandey, Russell Contreras andMarc Caputo at Axios:

Research suggests Republicans have become somewhat less likely than Democrats to participate in telephone surveys, and response rates overall have plummeted since the late 1990s, per Pew.

• "We definitely have more people answering surveys who voted for [former Vice President Kamala] Harris than who voted for President Trump," says Brent Buchanan, a Republican pollster and founder of Cygnal. "But if you're a good pollster, you still get the other people."
• Pollsters can compensate by giving greater weight to Trump voters or adjusting samples to better reflect the Republican electorate, which skews male, white and working class.
• Advanced statistical modeling tools can also take a poor sample and make it look like the true electorate, though results may be inaccurate for subgroups like young people or Hispanic voters, Pew's Keeter says.

But that has risks: "The adjustment assumes that the Republicans who respond to a poll are similar to the Republicans who don't respond to the poll," Rice University political science assistant professor Matthew Tyler tells Axios. Tyler said weighting GOP respondents more heavily can backfire if those willing to answer polls are also more willing to vote Democratic than the Republicans who never respond. That could inflate the apparent size of a Democratic wave.

Between the lines: Predicting who will actually vote can dramatically change the results. YouGov chief scientist Douglas Rivers found Democratic leads ranging from six to 16 points depending on how likely voters were identified in a September survey — and called the 16-point margin "almost certainly an overstatement."

The bottom line: Even with the questions about the accuracy of recent polls, most pollsters believe Democrats have the edge heading into Election Day. "I think it's telling that we're hanging our hopes on, 'The polls are wrong,'" Buchanan says. "That's not very strong."

October Surprises Have Changed

 AT WP, Dan Merica writes that the "October surprise" -- a last-minute story that changes the course of a campaign -- is not what it used to be.

“Unless you get a Cal Cunningham-level sex scandal,” Hunt said, “it is really hard to have anything cut through the clutter and the noise that is pumping into voters’ consciences this time of year.”

Instead, operatives said major stories that end campaigns are coming earlier in cycles, often ahead of primaries, so parties have time to replace candidates on the ballot.

There are three key reasons for this shift:

• Early voting: To expand access, many states are starting voting earlier. States like California, Maine and Montana began voting earlier this week, and Iowa, Kansas and North Carolina, among others, will begin next week.
• Changes in political communication: Now, campaigns can push their message earlier and without media involvement through social media and paid communication.
• Self-research: One operative working on House races this cycle said committees and campaigns “have gotten better at recognizing bad candidates early on before it is too late” and “know what to look out for before we fully embrace a candidate, before they become our nominee.”

The October Surprise is not dead. It has just changed.

House v. Senate

 In Federalist 51, Madison wrote:

In republican government, the legislative authority necessarily predominates. The remedy for this inconveniency is to divide the legislature into different branches; and to render them, by different modes of election and different principles of action, as little connected with each other as the nature of their common functions and their common dependence on the society will admit.

An old saying in the House:  “The other party is the opposition; the Senate is the enemy."

 Hans Nichols at Axios:

House Majority Whip Tom Emmer (R-Minn.) is saying out loud what House Republicans have been grumbling about for much of the year: They're fed up with the Senate.

Why it matters: You won't ever hear Speaker Mike Johnson or Majority Leader Steve Scalise talk about their Senate counterparts like this."There are no Democrats. There are no Republicans. They are just senators," Emmer told donors last week at a closed-door fundraiser for Rep. Mark Harris (R-N.C.). We were in attendance.

Emmer said Senate Majority Leader John Thune "is doing the job that he thinks he's supposed to do. There's a certain fraternity sorority that is the U.S. Senate."

 The intrigue: Emmer had a blunt message for senators who repeatedly say they can't pass House GOP priorities: "Don't just keep telling me 'Don't have the votes. Don't have the votes.' You suck."

Emmer said he once thought as a House member that he should stay out of the filibuster debate.

Now, he thinks his colleagues should "get rid of the darn filibuster."

Perceptions of Media Freedom

 Many posts have dealt with media problems.

Benedict Vigers at Gallup:

Even though most Americans continue to say the U.S. media have a lot of freedom, the percentage saying so has reached a new low this year. Slightly more than two in three Americans (69%) surveyed in May and June say there is a lot of media freedom in the U.S., down six percentage points from the previous low of 75% in 2025.

...

For much of the past 15 years, perceptions of media freedom in the U.S. have outpaced those in other advanced economies. Across 38 Organisation for Economic Co-operation and Development (OECD) countries, median perceptions of media freedom have typically hovered around 80%, slightly lower than in the United States. Last year was the first time on record that the two were tied, both at 75%, despite similarities in 2023 and 2015, when the U.S. advantage was smaller. While Gallup has not yet completed data collection in all OECD countries this year, based on available data, the OECD median will be between 70% and 76%. This means that 2026 is the first year on record in which the U.S. ranks below the bloc's median.

Virtually every president has complained about the press. Trump says that reporters who irk him "should be arrested." Can't you see that his instincts are authoritarian? I know a lot of "anti-anti-Trumpers." Can't you see it, my friends? Doesn't it matter? https://t.co/qDyxSZxwfy

— Jay Nordlinger (@jaynordlinger) October 6, 2026

Rose Research, Fall 2026

Watch this clip from The Wire.

It is the best description of research, ever.

The big stuff:

Beware hallucinations and incomplete answers. Always ask for sources and links.

Finding the literature: https://consensus.app/

Great resources at Honnold Library -- which students usually overlook! 

• Nexis Uni: news sources and law journals
• Political science journals
• Dissertation abstracts (search for "California" and "redistricting" in abstracts, and you will see a couple of Rose Institute names)

The Internet Archive—if a link is broken or altered, this site might help you find it. 

California 

• CA.gov
• California ballot measures -- archive going back to 1911
• California bills and laws
• LAO -- Legislative Analyst's Office -- analysis of CA issues
• Public Policy Institute of California -- California surveys and policy analysis
• News sites: CalMatters and Rough & Tumble
• CA elections from the Secretary of State
• California Elections Data Archive -- local elections in CA
• California State Library Digital Collections
• State archives
• California open data

General State Policy and Politics

• NCSL -- National Conference of State Legislatures -- analysis & data on state issues  Partisan composition of legislatures
• Open States -- bill information for all state legislatures
• Urban Institute State Economic Monitor
• Ballotpedia
• Tax Foundation

General Public Policy and Finance

• CRS
• CBO
• GAO
• GovInfo
• Budgets
• Biden White House Archives
• FRED: Federal Reserve Economic Data
  
• Census
• Bureau of Labor Statistics

Crime

• Bureau of Justice Statistics
• Prisoner statistics
• Crime data explorer
• Gun Violence Archive

Health

• Local data
• HealthData.gov
• Kaiser Family Foundation

National Elections, Parties, Campaign Finance

• Precinct-level national maps
• 2016
• 2020
• 2024 
• US Presidential Election Atlas -- just what the title says, great historical data (but note the unusual color scheme: red is D, blue is R)
• US Elections Project -- turnout data
• Open Secrets -- Center for Responsive Politics site for federal campaign finance and lobbying e information
• FEC -- Federal Election Commission
• FCC Public Inspection Files (ad buys on broadcast stations)

Public Opinion

• Polling Report -- an aggregation of national poll data
• Gallup -- the best-known US pollster, delivering new numbers daily
• Pew Research Center
• 2018 exit poll
• 2020 exit poll
• 2024 exit poll

Killing Off Disability Data

"Every record has been destroyed or falsified, every book has been rewritten, every picture has been repainted, every statue and street and building has been renamed, every date has been altered. And that process is continuing day by day and minute by minute."   -- George Orwell, 1984

A number of posts have discussed presidential documents, official data, and government websites.  The Trump administration has killed off or limited access to many such resources, including the CIA World Factbook, State Department X posts, and DOJ press releases about the January 6 insurrection.

The administration is also stopping the collection of key statistics.

 Michelle Diament at Disability Scoop:

A proposed revamp of a key federal survey could lead to a drastic undercount of Americans with developmental disabilities, advocates warn.

The U.S. Centers for Disease Control and Prevention is calling for major changes to the 2028 version of the National Health Interview Survey. The annual questionnaire asks about 27,000 adults about their experiences with illness, chronic conditions, injuries, health access and more.

The current iteration includes 49 questions related to disabilities and functioning, according to Lindsay DuBois, a research associate at the nonprofit Human Services Research Institute, which published guidance for disability organizations and other stakeholders on the proposal. Under the new plan, she said that number would drop to 15.

Importantly, advocates say the revisions would leave out a question about whether a person’s difficulties with cognition, communication, mobility or other issues originated before the age of 22 as well as questions about challenges learning or completing everyday tasks, all of which were specifically designed to improve identification of people with developmental disabilities.

“The bottom line: The proposed revisions will make it harder to identify in the data people with disabilities — especially people with intellectual and developmental disabilities,” DuBois said.

Tariffs and Capital Investment

 Many posts have dealt with tariffs and trade. 

Kyle Pomerleau, Thomas Brosy, Robert McClelland, et al. "The Impact of Tariffs on the Cost of Capital in the US, AEI, September 30, 2026

Since 2025, the second Trump administration has imposed significant levies on most imports. Although tariffs are routinely discussed as a consumption tax, 36 percent of all imports are capital assets used to produce goods and services. As a result, tariffs place a meaningful burden on new investment in the United States.

We estimate that Trump’s tariffs (before the Supreme Court’s decision) raised the cost of capital by as much as 2.7 percent. This is equivalent to raising the METR [marginal effective tax rate] on new investment by 4.2 percentage points. The burden varies significantly by asset, industry, and legal form of organization, placing the largest burden on equipment. A 4.2 percentage point increase in the METR roughly offsets the OBBBA’s tax cut on investment. The burden that tariffs place on investment is an underappreciated effect of these taxes that has important implications for tax reform, the macroeconomy, and the distribution of the tax burden.