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
Continuing my obsession with the age crosstabs, in the Times latest Boomers (well, 65+) are more supportive of Biden than any other group in the reg voters universe, but narrowed to likely voters, Boomers and Zs (well, under 30) support at equal rates
Since survey results from inherently flawed question formats shouldn't dominate the discourse, let's talk about other findings from the NYT/Siena poll. For instance: more of the public is expecting Trump to perform well in the debate than are expecting the same from Biden.
Also, according to this poll about 3/4 of registered voters plan to watch the debate, including 61% of people who reported not voting in 2020. If we use CPS 2022 estimates of registered adult population, that means an audience of just shy of 120m adults.
Fair enough, I was more generally interested in the "which candidate is better on x" question format. The article does highlight some other questions asking about descriptions of forms of government though. The responses are not as strongly pro democratic as I'd personally like to see
Folks reposting this: note that this isn't a crosstab. 38% is of the "deciders" (a group of young, irregular, and not-definitely-decided voters). It's not of the 7 in 10 that think Trump will reject a defeat
The people writing Heritage's Project 2025 are so detail oriented that ... they got the acronym for the ACS wrong. It's American Community Survey (community singular), which you would know if you looked at the Census website
In the latest estimates of cell phone use from the National Health Interview Survey (the authoritative source on this question), as of December 2023, 75% of US adults now live in a cell-only household, and another 14% are "wireless-mostly" www.cdc.gov/nchs/data/nh...
For example, Gorsuch’s opinion in Kennedy vs Bremerton School District is premised on a claim that can only be accurately construed as a lie, given the clear contradictory photographic evidence provided in the dissent www.msnbc.com/rachel-maddo...
First, let's start with acknowledging the fact that 2020 had the worst polling errors on average in at least 2 decades, according to the survey research industry-group AAPOR task force's post mortem on 2020 pre-election polls (cite: aapor.org/wp-content/u...)
But let's put that in context. Being 4 percentage points off makes a difference in close elections, to be sure, but it's also relatively small errors in the overall distribution of attitudes. More importantly, it's much smaller errors than the early days of polling which were far more inaccurate.
Cohn doesn’t address this directly, probably because the Times’ surveys haven’t included a survey item that tries to measure voters’ dedication to democracy. (He does talk about issue prioritization of democracy and abortion among Dems, though that’s not quite the same thing)
So if you’re running a national voter-file sample or -matched survey, let me suggest that you ask people directly how much they value elections as a mechanism for selecting leaders. Something like below
Weirdly there is not a canonical cite I can find for why you should do this (recs welcome!), though the Caughey et al Cambridge element makes an analagous point about weighting for southern respondents devincaughey.github.io/files/caughe...
That they think you can field a high quality online survey with 50,000 respondents for 18,000 tells you everything you need to know about how much experience they have with survey research. (No, mechanical turk responses are not surveys)
I'm also laughing on the floor that they think you can get a high quality survey of 50,000 respondents for 18,000. Tell me you've never fielded a survey without telling me you've never fielded a survey