iiinova — where your digital self keeps evolving

Reposted by Ariel Edwards-Levy

Edward Swaine @edswaine.bsky.social · 29/09/2026

i no longer understand what's on the ballot

Brad Pitt and daughter are on the ballot

3213

Ariel Edwards-Levy @aedwardslevy.bsky.social · 5h

the driver is a better person than me because the temptation to place this on the windshield must be astounding

5422

Ariel Edwards-Levy @aedwardslevy.bsky.social · 5h

mostly just want @ethanbencohen.bsky.social to be mildly disappointed by this joke

3100

Ariel Edwards-Levy @aedwardslevy.bsky.social · 5h

flipping ratings so quickly, they must be a short-order Cook

6814

Ariel Edwards-Levy @aedwardslevy.bsky.social · 6h

IPUMS is extremely useful, and also I have a mental block where I can only process "IPUMS" as a pet name akin to sweetums or snookums aw, ipums

290

Ariel Edwards-Levy @aedwardslevy.bsky.social · 6h

also, would it kill The Media to call home once in a while

040

Reposted by Ariel Edwards-Levy

Kathleen Weldon @kathleenweldon.bsky.social · 10h

Daylight savings = perennial hot topic. But what I find interesting here is how invested the survey org (Minn Poll) was in respondents enjoying the survey. Respondents were expected to find taking a poll a bit of communal fun, not an unpleasant civic obligation or something done for a bit of cash

Our opener this month is our all-time favorite opener, accor-
ding to veteran interviewers. No question seems to get a
respondent more in the mood to be interviewed than questions about daylight saving time. We hope it's as popular this year as it has been in the past.

1153

Ariel Edwards-Levy @aedwardslevy.bsky.social · 10h

I really like this question. KFF asked voters who said they were more motivated than usual for this election to explain why they felt that way -- top answers to bring change, or anti-Trump sentiment www.kff.org/public-opini...

2182

Ariel Edwards-Levy @aedwardslevy.bsky.social · 23h

would have been such a good bit for Handel's Water Music just to be a John Cage situation where you spend ~45 minutes really listening to the rain

4342

Ariel Edwards-Levy @aedwardslevy.bsky.social · 06/10/2026

how is there not a Finding Nemo/Oliver crossover where Nemo falls in with a bunch of impish urchins

6181

Reposted by Ariel Edwards-Levy

Steven Rich @datab.ae · 05/10/2026

fun fact: the tuba section of the university of missouri marching band is called the mizzousaphones

3575

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

for reasons I do not profess to understand, some of my acquaintances are not active on bsky dot app

160

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

no force in existence could prevent me

060

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

oddly enough, also his response to Calvin

110

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

(obviously I am still paying close attention and there is no shortage of things to find interesting and important, but there are delightful people who can happily rattle off facts about downballot state races from 60 years ago and I am not necessarily one of them)

1120

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

my greatest secret is that because I am a public-opinion nerd, everyone thinks I am also an elections nerd, and by nature I am not

6462

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

i am not allowed to write any more emails today

7240

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

the rubber soul outtakes are so good, she said youthfully

130

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

area public opinion researcher not entirely sure if she's kidding

0250

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

solving survey nonresponse by convincing the lonely of how much fun they'll have talking to interviewers

1422

Ariel Edwards-Levy @aedwardslevy.bsky.social · 05/10/2026

aspects of gongoozling en.wikipedia.org/wiki/Gongooz...

4221

Ariel Edwards-Levy @aedwardslevy.bsky.social · 03/10/2026

you can't expect us all to just travel out there and incur Roman charges

0260

Ariel Edwards-Levy @aedwardslevy.bsky.social · 03/10/2026

the correct scansion for this would be "some say evites," and it bothers me deeply

3110

Ariel Edwards-Levy @aedwardslevy.bsky.social · 03/10/2026

some say the world will end in promos/some say in evites

5312

Ariel Edwards-Levy @aedwardslevy.bsky.social · 02/10/2026

congratulations on being among today's winners of random digit dialing!

1100

Ariel Edwards-Levy @aedwardslevy.bsky.social · 02/10/2026

i find it deeply annoying to suspect it's not going to be as warm and lovely as it is today again until like April or May

1160

Reposted by Ariel Edwards-Levy

depths of wikipedia @depthsofwikipedia.bsky.social · 02/10/2026

via wiktionary

351567

Reposted by Ariel Edwards-Levy

Amanda Katz @katzish.bsky.social · 01/10/2026

NEW in @porchparty.bsky.social: The brilliant @zephoria.bsky.social embedded in the Census Bureau during the 2020 count. In this Q&A on her new book, she describes how civil servants stuck the landing despite Covid, and explains why the time to save 2030 is NOW www.porchpartynews.com/p/why-the-us...

porchpartynews.com

Why the US Census is on the brink

Researcher danah boyd watched heroic civil servants pull off the 2020 census amid a global pandemic. But the 2030 count is in peril—and she says the time to save it is now.

13617

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

̷r̷e̷a̷l̷ ̷d̷i̷c̷k̷ ̷m̷o̷v̷e̷

1171

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

enormously tempted to update my banner image to "Women find the Poll questions interesting"

2221

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

this feels like @dhmontgomery.com's, er, music

180

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

before she hung up, she said she was a skeletoooooooon

080

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

time to have spirit halloween stuck in my head for another month straight

030

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

personally I'll be getting the best of both worlds by relying solely on the longform writings produced by historical users of MTurk

010

Ariel Edwards-Levy @aedwardslevy.bsky.social · 01/10/2026

I think this qualifies you to be time's person of the year

24123324

Reposted by Ariel Edwards-Levy

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2024

the october surprise is october being tomorrow

381844318

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

I just used it to code a tool that largely automates some tedious document formatting tasks. That was pretty useful!

100

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

but it does always crack me up when people outside the field start in with "do pollsters ever worry about..." because the answer is almost invariably "yes" and quite often additionally "would you like to see the task force report"

2272

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

lol, this is not incorrect, but the real question, as always, is "trust it to do what" have realistic expectations for your data and you'll get more out of it!

4483

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

In fairness, I think there's a whole scale in terms of the level of modeling applied to a dataset, starting with basic weighting and working up through things like modeled/inferred voter file demos (see www.pewresearch.org/methods/2026...), but this strikes me as pretty far over the line

pewresearch.org

3 ways to correct a poll’s political balance: Party affiliation, past vote and voter file information

About this research This report explains political weighting – adjusting a survey to reflect the political composition of the United States as accurately as possible. Why did we do this?…

1160

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

I also find this fascinating: "More broadly, our AI respondents didn’t like to admit to uncertainty. Across all the questions on our three surveys where a 'not sure' option was offered, human panelists were around four times as likely as the AI model to choose it."

1758

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

From the report: "[T]the opinions and attitudes of the public can sometimes shift quickly and dramatically...This is part of what’s interesting and challenging about studying public opinion....[inference] can struggle to identify changes in sentiment when they happen quickly."

1464

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

I think AI has its uses, but I continue to believe that quantitative public opinion research is fundamentally a way of talking to people at scale, emphasis here on "people"

61087

Ariel Edwards-Levy @aedwardslevy.bsky.social · 30/09/2026

There is a lot of "oh dear" to go around in the results of this very thorough Pew study on the issues with synthetic samples, but this strikes me as particularly "oh dear" www.pewresearch.org/data-labs/20...

AI polling results can lean on racial or partisan stereotypes

In many cases, our AI poll took beliefs or attitudes that are reasonably common among a particular subgroup and portrayed them as nearly ubiquitous. For example, our AI poll would indicate that:

97% of Hispanic adults are at least somewhat likely to follow the World Cup (while the actual share is 43%).

7457160

Ariel Edwards-Levy @aedwardslevy.bsky.social · 29/09/2026

I always have very conflicting emotions when respondents say something wildly conspiratorial or bigoted in an open-ended question because on the one hand, yikes, and on the other hand at least they're still taking surveys?

172

Ariel Edwards-Levy @aedwardslevy.bsky.social · 29/09/2026

FWIW a) reaching enough/the right kinds of these folks is obviously a huge and perennial concern, but b) the answer here is "at least some of 'em will (which we know because they're telling pollsters they won't vaccinate their kids)" www.kff.org/public-opini...

191