Phil Swatton
philswatton.bsky.social
Phil Swatton
@philswatton.bsky.social
Work as a data scientist at the Alan Turing Institute, background in political science. Views my own and not necessarily shared by my employer.

https://philswatton.github.io/
Reposted by Phil Swatton
Again some excellent work by @philswatton.bsky.social!! Check it out:
I have a new preprint: 'Predictive Modelling Shows Demographics Do Not Predict Vote Choice at the Individual Level'

DOI: doi.org/10.31235/osf...

Thread below
November 24, 2025 at 10:27 PM
This whole thread is a wonderful defence of a humane and tolerant attitude, but how prescient was Stuart Hall here: " The capacity to live with difference is ... the coming question of the twenty-first century."
In my article, I quoted something from Stuart Hall's "Culture, Community, Nation," an essay first published in a 1993 issue of Cultural Studies. He writes:

"It should not be necessary to look, walk, feel, think, speak exactly like a paid-up member of the buttoned-up, stiff-upper-lipped, ...
November 24, 2025 at 11:45 AM
If, like me, you are an enjoyer of Ben's patented bubble plots, you'll have likely noticed that voters cluster by ideology and not demographics.

My new preprint is on exactly that: osf.io/preprints/so...
November 21, 2025 at 4:47 PM
This is every bit as good as people have said. Well worth a read
November 21, 2025 at 3:46 PM
Headline empirical findings:

Models using demographics are not much better at prediction than just assuming all voters vote the same way.

Models using ideology however are better, and are close to models using past vote choice w/ constituency vars

This result generalises from BES to EES
November 21, 2025 at 7:17 AM
Reposted by Phil Swatton
I have a new preprint: 'Predictive Modelling Shows Demographics Do Not Predict Vote Choice at the Individual Level'

DOI: doi.org/10.31235/osf...

Thread below
November 20, 2025 at 8:09 AM
I have a new preprint: 'Predictive Modelling Shows Demographics Do Not Predict Vote Choice at the Individual Level'

DOI: doi.org/10.31235/osf...

Thread below
November 20, 2025 at 8:09 AM
It's not even the case that Down's model suggested the median voter would be vote-optimising. It was only such if the single ideological dimension of public preferences is normally distributed.
Fascinating how often political commentary in the UK still refers to the median voter. In a multi-dimensional space with salience endogenous to positions, it‘s unclear to me who that should be. In a multi-party system, winning the median voter is of course not necessarily a vote-maximizing strategy.
November 19, 2025 at 7:57 PM
Morally repugnant.
The Sun has been told Home Secretary Shabana Mahmood will on Monday propose confiscating jewellery, watches, necklaces from asylum seekers to meet asylum costs

This reflects the most controversial aspect of the Danish scheme - the Jewellery Law. The toughest Labour MPs thought this was OTT
November 17, 2025 at 7:48 AM
Reposting this for I suspect not the last time in the lifetime of this parliament: renewal.org.uk/blog/public-...
November 14, 2025 at 11:49 AM
Outsourcing an ML results interpretation problem:

You are comparing classifiers A and B in terms of accuracy on a small number of datasets (say, 15) with small test sets in the range of 100-200 obs. (1/3)
November 12, 2025 at 1:40 PM
I've written a blog post about using metric unfolding to extract latent dimensions from data, using the ANES feeling thermometers & BES 'like' scales as examples.

It's a high-level intro to unfolding and using the smacof R package. These plots are the results.
philswatton.github.io/2025/11/05/e...
November 7, 2025 at 11:55 AM
Reposted by Phil Swatton
"If the person is a U.S. citizen or otherwise lawfully in the United States, that individual will be free to go after the brief encounter."

--Brett Kavanaugh, Noem v. Vasquez Perdomo

www.chicagotribune.com/2025/11/03/c...
November 3, 2025 at 4:20 PM
Great piece, very much looking forward to part 2. Excellent critiques both of various right-wing ideologies around tech and of left-wing disinterest in tech.
October 25, 2025 at 6:14 PM
The UK's liberal democracy is not being defended by its mainstream politicians in Labour or the Conservatives. Only third parties seem alive to the threat. Unless something changes, and soon, we will fibd ourselves experiencing our own version of Trump's US.
It'd be nice of Downing Street would wholeheartedly condemn the Lam mass-deportation stuff, but it's also an enormous failure of the centre-right to maintain a cordon sanitaire against this slide.
October 22, 2025 at 10:53 AM
Reposted by Phil Swatton
🚨NEW BLOGPOST🚨

Since the government can't seem to see it, I explain how and why a simple act of Parliament can bring the Andrew affair to the end - and why they should do it.

joxleywrites.substack.com/p/parliament...
October 20, 2025 at 10:28 AM
Reposted by Phil Swatton
One of my issues with Politico is that its jaunty, gossipy tone jars with the dangerous turn in UK politics.

A "mostly clanger free conference"? A recording of the Shadow Justice Secretary complaining about seeing "no white faces" surfaced and was the talking point in interviews and coverage...
October 8, 2025 at 6:45 AM
Just caught up on this fantastic essay by @sundersays.bsky.social in @renewaljournal.bsky.social. Brilliant, punchy, to the point on Labour's lack of serious engagement with multiculturalism, and a much-welcome rebuttal to those who advocate for Labour to accommodate Reform's radical right populism.
October 5, 2025 at 6:46 PM
I wrote a slightly longer response to the NS piece on my substack. You can read it here: dysfunctionalprogramming.substack.com/p/some-comme...
October 3, 2025 at 7:28 PM
The thing I most hate about this kind of reasoning is that it regards the question of whether or not Farage is a racist as unimportant or incidental. All that matters is what an imagined median voter likes right now. Politics as anything other than triangulation regarded as a misstep.
October 2, 2025 at 6:03 AM
Reposted by Phil Swatton
Short article covering the psychometrics and LLMs work we published a few months back. Hopefully, more readable (i.e. less maths!) :)

@philswatton.bsky.social

cetas.turing.ac.uk/publications...
Patterns, Not People: Personality Structures in LLM-powered Persona Agents
Expert Analysis
cetas.turing.ac.uk
October 1, 2025 at 2:02 PM
Reposted by Phil Swatton
Quite. So many Labour people on this website visibly have no idea of just how bad things have got for ethnic minorities since they came on, and how their visible indifference lands.
Also frankly I am quite desperate! A party that’s at best ambivalent on the question of whether I’m English and wants to deport my friends is 9 points ahead in the polls and the current leader is floundering and visibly out of his depth!
September 24, 2025 at 11:47 PM
Reposted by Phil Swatton
Brilliantly lucid piece from Phil - well worth a read.
New piece by me. I wrote about what the political methodology literature tells us about the uses and abuses of surveys, why a polling-first political strategy is a mistake, and why a clear set of values and a clear political vision is essential for reconciling contradictory pressures on policy.
"public opinion should only be one consideration among many in policymaking and party strategy, and never the prime mover of politics."

@philswatton.bsky.social on the pitfalls of poll-driven politics and the importance of political leadership.
renewal.org.uk/blog/public-...
September 10, 2025 at 12:57 PM
New piece by me. I wrote about what the political methodology literature tells us about the uses and abuses of surveys, why a polling-first political strategy is a mistake, and why a clear set of values and a clear political vision is essential for reconciling contradictory pressures on policy.
September 8, 2025 at 5:20 PM
It's good to see these MPs continuing to stand up for the role of deliberation in democracy and for the scrutinising role of parliament. If Farage does become PM in the future, a concentration of power in the executive will prove a catastrophe.
September 4, 2025 at 11:56 AM