Kevin Collins
@kwcollins
Co-founder, Survey 160. Loves survey and voter participation. Researcher. Democrat. YIMBY. Husband and Dad. Opinion haver and measurer. Outdoors and cooking enthusiast. He/Him
The polling on Iran has gotten steadily more negative since then. For example, here's Quinnipiac's trend poll.qu.edu/images/polli...
First, consider this longitudinal data on job approval (from high quality GOP firm Echelon Insights), generally and foreign policy specifically. Trump has pretty consistently fallen below Biden's at the same point in the term echeloninsights.substack.com/p/friday-fin...
But this add for the YouGov Parallax product (side note: please stop naming products after dystopian movies) raises something that hadn't previously crystalized for me: YouGov is selling a virtual version of me (as I am, like any self-respecting survey researchers, on as many panels as possible)
These numbers are pretty old (May 28- June 1), and more importantly Fox News (while a high quality pollster) is an outlier here. Which would have been clearer if the anonymous account retweeted here had included a link to the poll www.nytimes.com/interactive/...
There is a new Fox News poll for a Senate seat in the South showing the Democrat with a sizable lead ... but it's North Carolina not Georgia. Don't repost old polls like they're news, folks www.foxnews.com/politics/fox...
Note this is from last month and is a bit of an outlier, though most polls have Ossoff up comfortably www.nytimes.com/interactive/...
Now Trump is f-ing with the Cherry Blossom 10 miler course (from an email from the race organizers)
Republicans largely erasing their party id advantage among Americans without a college degree seems like a finding from the Pew NPORS study that should be getting more attention www.pewresearch.org/politics/fac...
Post a perfect album from the 90's that isn't Nirvana, Soundgarden, Pearl Jam, or Alice In Chains
Impressive collapse in Newsom's poll numbers in NH since February scholars.unh.edu/survey_cente...
Mode shifts over the period largely came from entrants and exits rather than methodology changes within firms (though there was more mode shifting between 2020 and 2024 than between 2016 and 2020). The new entrants poses a risk, as newer firms were on average less accurate
Nice little experiment new in Field Methods on the inclusion of QR codes in survey recruitment letters. They found ~3 pp increase response rates, but surprisingly that was concentrated among older respondents. journals.sagepub.com/doi/full/10....
This article on attention looks super interesting, supporting my view that attentiveness is not just a function of individual traits but also of survey traits. Field shorter surveys. And if someone with access wants to send me a copy I'd be very grateful. journals.sagepub.com/doi/abs/10.1....
Interesting new study looking at correlates of panel retention using the ANES. Finds political engagement predicts retention, but partisan strength and affective polarization do not, with effect of partisan strength on retention declining over time www.surveypractice.org/article/1614...
Who decided it was a good idea for the Smithsonian to start using the shittiest, most deceptive fundraising tactics developed from campaign donation mining?
Interesting study by Census on their Household Pulse Survey, now out in @surveypractice.bsky.social, suggesting that text messages outperform email as recruitment tool for Spanish-speaking respondents (needs an experiment!) www.surveypractice.org/article/1603...
The only difference between this and when the Klan marched on Washington 100 years ago is that the Klan actually showed their faces
For those interested, here's the methods from the Iowa poll (similar method for GA)
Most Important Issue questions should list "political corruption" as an option, generally but *especially* in Georgia where Ossoff has made it the centerpiece of his campaign (even before the GOP nominated Collins, which ... was a choice)
I would love to know why Iowa and Ohio are disproportionately affected by synthetic past vote weighting (6 and 4 points to the right, respectively) in the latest Times/Siena polls www.nytimes.com/2026/07/01/u...
As always, the question is “compared to what?” And compared to issue-free promises of being an independent voice that break from party leadership, “moderate” doesn’t perform any better, (suggesting but not proving) that “moderation” is just a signal of independence
Other polls from this race in June have El-Sayed leading Stevens, and McMorrow in single digits, but with many more undecideds www.nytimes.com/interactive/...