Marios Richards
@mariosrichards.bsky.social
Frequently wrong. Please correct (effort involved appreciated).
Experiments with Data Visualisation:
https://github.com/MariosRichards/BES_analysis_code
https://medium.com/@mariosrichards
https://mariosrichards.substack.com/
Experiments with Data Visualisation:
https://github.com/MariosRichards/BES_analysis_code
https://medium.com/@mariosrichards
https://mariosrichards.substack.com/
Reposted by Marios Richards
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...
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
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...
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...
Reposted by Marios Richards
Der Beitrag basiert auf einem Artikel in der @pvs-journal.bsky.social, der gerade #onlinefirst und natürlich #OA erschienen ist:
👉 www.link.springer.com/article/10.1007/s11615-025-00637-3
👉 www.link.springer.com/article/10.1007/s11615-025-00637-3
Personality Traits and Vote Switching: A Tale of Open-Minded Switchers and Agreeable Populism Resisters - Politische Vierteljahresschrift
In increasingly individualised Western societies, personality is becoming a key factor in understanding political behaviour. While some voting patterns are well researched, this article examines a les...
www.link.springer.com
November 6, 2025 at 3:43 PM
Der Beitrag basiert auf einem Artikel in der @pvs-journal.bsky.social, der gerade #onlinefirst und natürlich #OA erschienen ist:
👉 www.link.springer.com/article/10.1007/s11615-025-00637-3
👉 www.link.springer.com/article/10.1007/s11615-025-00637-3
So I mostly think this is just a side effect of
(i) polling where you give people lots of options other than broad income tax bump
(ii) political cueing (public support for tax rises for was at/near an all-time high until 2023*)
* curious what drives that 2023+ jump in 'reduce taxes' ...
(i) polling where you give people lots of options other than broad income tax bump
(ii) political cueing (public support for tax rises for was at/near an all-time high until 2023*)
* curious what drives that 2023+ jump in 'reduce taxes' ...
November 3, 2025 at 11:20 AM
So I mostly think this is just a side effect of
(i) polling where you give people lots of options other than broad income tax bump
(ii) political cueing (public support for tax rises for was at/near an all-time high until 2023*)
* curious what drives that 2023+ jump in 'reduce taxes' ...
(i) polling where you give people lots of options other than broad income tax bump
(ii) political cueing (public support for tax rises for was at/near an all-time high until 2023*)
* curious what drives that 2023+ jump in 'reduce taxes' ...
Very cool - I'm particularly interested in the algorithms to infer 'community structure' (LIB/RWA/ALT 'factors' connected to the various items). Would be interesting to see the results on a dataset with more economic values (at first glance this looks like 17 auth-correlated items and Lower Taxes).
1/2 I am delighted that, after a very long gestation, my paper on political ideologies as mass belief systems, using network psychometrics, has been published: journals.plos.org/plosone/arti...
The structure of mass political belief systems: A network approach to understanding the left-right spectrum
Many socially consequential beliefs, notably political and religious ideologies, consist not of single propositions in isolation from others but as systems of many propositions. Philip Converse, one o...
journals.plos.org
November 1, 2025 at 2:37 PM
Very cool - I'm particularly interested in the algorithms to infer 'community structure' (LIB/RWA/ALT 'factors' connected to the various items). Would be interesting to see the results on a dataset with more economic values (at first glance this looks like 17 auth-correlated items and Lower Taxes).
British Election Study 1997 asks more or less the same question to the same respondents - do you like the Conservative Party? - on a 5pt scale and an 11pt scale ...
... and the correlation of the results is only .77.
9 point "Strongly Against" them ... who *also rated them 10/10 for liking them!*.
... and the correlation of the results is only .77.
9 point "Strongly Against" them ... who *also rated them 10/10 for liking them!*.
October 31, 2025 at 10:29 PM
British Election Study 1997 asks more or less the same question to the same respondents - do you like the Conservative Party? - on a 5pt scale and an 11pt scale ...
... and the correlation of the results is only .77.
9 point "Strongly Against" them ... who *also rated them 10/10 for liking them!*.
... and the correlation of the results is only .77.
9 point "Strongly Against" them ... who *also rated them 10/10 for liking them!*.
Reposted by Marios Richards
A majority of voters, and a plurality of white voters, think that Reform is a racist party with racist policies.
October 28, 2025 at 2:03 PM
A majority of voters, and a plurality of white voters, think that Reform is a racist party with racist policies.
Good to see that Silicon Valley is following Neon Genesis Evangelion both in content and logical coherence.
October 28, 2025 at 1:45 PM
Good to see that Silicon Valley is following Neon Genesis Evangelion both in content and logical coherence.
I've come up with - yet another - plot. This time combining mini pie charts and the political compass plot.
Each pie chart reflects how voters have shifted preference from their 2024GE Labour vote to their vote intention in May 2025 for the value subgroup.
Each pie chart reflects how voters have shifted preference from their 2024GE Labour vote to their vote intention in May 2025 for the value subgroup.
October 27, 2025 at 4:36 PM
I've come up with - yet another - plot. This time combining mini pie charts and the political compass plot.
Each pie chart reflects how voters have shifted preference from their 2024GE Labour vote to their vote intention in May 2025 for the value subgroup.
Each pie chart reflects how voters have shifted preference from their 2024GE Labour vote to their vote intention in May 2025 for the value subgroup.
I love("") that you can just read the timeline and watch the guy 'politically age' like it's a sped up nature scene.
Like watching someone segue smoothly "Haha, it almost seems like my belt is shrinking!" to "My belt really is shrinking each year":
en.wikipedia.org/wiki/Larry_S...
Like watching someone segue smoothly "Haha, it almost seems like my belt is shrinking!" to "My belt really is shrinking each year":
en.wikipedia.org/wiki/Larry_S...
October 25, 2025 at 4:13 PM
I love("") that you can just read the timeline and watch the guy 'politically age' like it's a sped up nature scene.
Like watching someone segue smoothly "Haha, it almost seems like my belt is shrinking!" to "My belt really is shrinking each year":
en.wikipedia.org/wiki/Larry_S...
Like watching someone segue smoothly "Haha, it almost seems like my belt is shrinking!" to "My belt really is shrinking each year":
en.wikipedia.org/wiki/Larry_S...
I may have buried the lede - here's a toy model where you can manually change party position *and see how moving towards another party can(!) reduce rather than increase* your voteshare.
There's nothing clever happening under the hood (no salience/legitimacy mechanics) just a dumb spatial model.
There's nothing clever happening under the hood (no salience/legitimacy mechanics) just a dumb spatial model.
Here is the - distinctly alpha version - Interactive Political Compass Dominance Map:
py.cafe/app/MariosRi...
Ever wanted to know what would happen if a party shifted position/changed perceived competence? Now you can!
py.cafe/app/MariosRi...
Ever wanted to know what would happen if a party shifted position/changed perceived competence? Now you can!
PyCafe - Panel - Interactive Political Compass Dominance Map
Create & Share Streamlit, Dash and Python Apps Online.
py.cafe
October 24, 2025 at 9:49 AM
I may have buried the lede - here's a toy model where you can manually change party position *and see how moving towards another party can(!) reduce rather than increase* your voteshare.
There's nothing clever happening under the hood (no salience/legitimacy mechanics) just a dumb spatial model.
There's nothing clever happening under the hood (no salience/legitimacy mechanics) just a dumb spatial model.
Here is the - distinctly alpha version - Interactive Political Compass Dominance Map:
py.cafe/app/MariosRi...
Ever wanted to know what would happen if a party shifted position/changed perceived competence? Now you can!
py.cafe/app/MariosRi...
Ever wanted to know what would happen if a party shifted position/changed perceived competence? Now you can!
PyCafe - Panel - Interactive Political Compass Dominance Map
Create & Share Streamlit, Dash and Python Apps Online.
py.cafe
October 24, 2025 at 9:34 AM
Here is the - distinctly alpha version - Interactive Political Compass Dominance Map:
py.cafe/app/MariosRi...
Ever wanted to know what would happen if a party shifted position/changed perceived competence? Now you can!
py.cafe/app/MariosRi...
Ever wanted to know what would happen if a party shifted position/changed perceived competence? Now you can!
A policy *supported by a smaller % of the population* than it would seek to deport!
Stephen Bush on the extraordinary draft legislation which Shadow Home Secretary Chris Philp, Matt Vickers, Katie Lam and backbench colleagues have proposed in parliament
It is a proposal that would seek to deport around 5% of the resident population, including over a quarter of a million with ILR
It is a proposal that would seek to deport around 5% of the resident population, including over a quarter of a million with ILR
October 23, 2025 at 11:26 AM
A policy *supported by a smaller % of the population* than it would seek to deport!
What are we supposed to do, Grandad ... "Build" "Houses" in any of the parts of the undeveloped parts of the UK?
No - we must conserve what is essential to Hungary.
No - we must conserve what is essential to Hungary.
October 22, 2025 at 12:18 PM
What are we supposed to do, Grandad ... "Build" "Houses" in any of the parts of the undeveloped parts of the UK?
No - we must conserve what is essential to Hungary.
No - we must conserve what is essential to Hungary.
Great post, but it doesn't go too far enough.
I mean, isn't this the perfect time to really make a pitch for the HANG THE PAEDOS, SAVE THE NHS vote?
I mean, isn't this the perfect time to really make a pitch for the HANG THE PAEDOS, SAVE THE NHS vote?
🚨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...
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 21, 2025 at 8:28 PM
Great post, but it doesn't go too far enough.
I mean, isn't this the perfect time to really make a pitch for the HANG THE PAEDOS, SAVE THE NHS vote?
I mean, isn't this the perfect time to really make a pitch for the HANG THE PAEDOS, SAVE THE NHS vote?
Reposted by Marios Richards
When we analyzed news coverage, we found the main driver to be intra-party divisions within the center-right CDU, where candidates were vying for party leadership.
While the UN compact & far-right agitation against it certainly warranted coverage, the issue became disproportionately amplified.
While the UN compact & far-right agitation against it certainly warranted coverage, the issue became disproportionately amplified.
July 14, 2025 at 8:24 AM
When we analyzed news coverage, we found the main driver to be intra-party divisions within the center-right CDU, where candidates were vying for party leadership.
While the UN compact & far-right agitation against it certainly warranted coverage, the issue became disproportionately amplified.
While the UN compact & far-right agitation against it certainly warranted coverage, the issue became disproportionately amplified.
October 19, 2025 at 10:10 PM
Reposted by Marios Richards
Your regular reminder that scrapping ILR for people who have come here and played by the rules, which is what Lam proposes here, is a position supported by 3% of the public
Sunday Times interview Tory "rising star" Katie Lam
She is clear she wants lots of legal migrants to be told to "go home" so as "to leave a mostly but not entirely culturally coherent group of people"
(The interviewer suggests she is scrapping ILR or stripping people of it)
She is clear she wants lots of legal migrants to be told to "go home" so as "to leave a mostly but not entirely culturally coherent group of people"
(The interviewer suggests she is scrapping ILR or stripping people of it)
October 19, 2025 at 7:30 AM
Your regular reminder that scrapping ILR for people who have come here and played by the rules, which is what Lam proposes here, is a position supported by 3% of the public
@dylandifford.bsky.social @ewanhoyle.bsky.social
I've had a look at the (mean) respondent-level gaps between party Like and PTV evaluations over time
~3 clusters
(1) Parties of Govt (Lab/Con/SNP) … & Reform in 2025
(2) 3rd Parties (LD/PC/UKIP/Reform until 2025)
(3) Green Party
I've had a look at the (mean) respondent-level gaps between party Like and PTV evaluations over time
~3 clusters
(1) Parties of Govt (Lab/Con/SNP) … & Reform in 2025
(2) 3rd Parties (LD/PC/UKIP/Reform until 2025)
(3) Green Party
October 15, 2025 at 9:04 AM
@dylandifford.bsky.social @ewanhoyle.bsky.social
I've had a look at the (mean) respondent-level gaps between party Like and PTV evaluations over time
~3 clusters
(1) Parties of Govt (Lab/Con/SNP) … & Reform in 2025
(2) 3rd Parties (LD/PC/UKIP/Reform until 2025)
(3) Green Party
I've had a look at the (mean) respondent-level gaps between party Like and PTV evaluations over time
~3 clusters
(1) Parties of Govt (Lab/Con/SNP) … & Reform in 2025
(2) 3rd Parties (LD/PC/UKIP/Reform until 2025)
(3) Green Party
One of the reasons I'm skeptical about the "Lump Of Taboo" concept - that you can only invoke a taboo a finite amount so many times before it 'runs out' - is that it doesn't seem to apply to any taboos in history.
Sinatra was such a fox
Frank Sinatra on what it means to be an American in 1945
October 14, 2025 at 1:26 PM
One of the reasons I'm skeptical about the "Lump Of Taboo" concept - that you can only invoke a taboo a finite amount so many times before it 'runs out' - is that it doesn't seem to apply to any taboos in history.
Reposted by Marios Richards
1. Ecological Fallacy: just because a constituency goes from one party to another doesn't mean many individual voters made same journey
2. Availability Bias: just because you can easily picture the stereotype of a particular voter doesn't mean that they are numerous in electorate
2. Availability Bias: just because you can easily picture the stereotype of a particular voter doesn't mean that they are numerous in electorate
The Two Psychological Biases to Avoid to Understand Voting Behaviour in 2025
The ecological fallacy and availability bias
jamesbreckwoldt.substack.com
October 14, 2025 at 7:27 AM
1. Ecological Fallacy: just because a constituency goes from one party to another doesn't mean many individual voters made same journey
2. Availability Bias: just because you can easily picture the stereotype of a particular voter doesn't mean that they are numerous in electorate
2. Availability Bias: just because you can easily picture the stereotype of a particular voter doesn't mean that they are numerous in electorate
Just started The Alters ... and, as all the reviews have said, it's really good.
(You crash land on alien planet, rest of the crew is dead and the only way you can survive is to crew your base with ... alternate versions of you)
store.steampowered.com/app/1601570/...
(You crash land on alien planet, rest of the crew is dead and the only way you can survive is to crew your base with ... alternate versions of you)
store.steampowered.com/app/1601570/...
The Alters on Steam
Explore an emotional sci-fi game with a unique blend of survival, adventure, and base-building elements. Help Jan Dolski, the sole survivor of an ill-fated space expedition, create alternative version...
store.steampowered.com
October 13, 2025 at 11:53 PM
Just started The Alters ... and, as all the reviews have said, it's really good.
(You crash land on alien planet, rest of the crew is dead and the only way you can survive is to crew your base with ... alternate versions of you)
store.steampowered.com/app/1601570/...
(You crash land on alien planet, rest of the crew is dead and the only way you can survive is to crew your base with ... alternate versions of you)
store.steampowered.com/app/1601570/...
Finally made it through to the end of Foundation Season 3.
It's almost precision-engineered to be the least Asimov thing possible (continuous meaningless action rather vs pre-adapted-for-radio-play / baroquely overwritten characters vs thin plot-aqueduct achetypes, sf concept overload vs clarity).
It's almost precision-engineered to be the least Asimov thing possible (continuous meaningless action rather vs pre-adapted-for-radio-play / baroquely overwritten characters vs thin plot-aqueduct achetypes, sf concept overload vs clarity).
October 11, 2025 at 10:08 PM
Finally made it through to the end of Foundation Season 3.
It's almost precision-engineered to be the least Asimov thing possible (continuous meaningless action rather vs pre-adapted-for-radio-play / baroquely overwritten characters vs thin plot-aqueduct achetypes, sf concept overload vs clarity).
It's almost precision-engineered to be the least Asimov thing possible (continuous meaningless action rather vs pre-adapted-for-radio-play / baroquely overwritten characters vs thin plot-aqueduct achetypes, sf concept overload vs clarity).
Reposted by Marios Richards
"The wealth tax debate shouldn’t be separate from the one about how we tax property and pensions – that’s where most of the country’s wealth sits." c/o @resfoundation.bsky.social Bet it will continue to be, though. 🙄
October 10, 2025 at 11:59 AM
"The wealth tax debate shouldn’t be separate from the one about how we tax property and pensions – that’s where most of the country’s wealth sits." c/o @resfoundation.bsky.social Bet it will continue to be, though. 🙄
This is just generally very good (new data, new analysis!).
Lots of fun stuff - but the thing that jumped out to me is that Labour's rhetoric on AI *actually reflects their real electorate's positivity on automation* - in contrast to Labour's imaginary electorate (skilled manual workers).
Lots of fun stuff - but the thing that jumped out to me is that Labour's rhetoric on AI *actually reflects their real electorate's positivity on automation* - in contrast to Labour's imaginary electorate (skilled manual workers).
What do the public actually think about economic growth, technological change and "abundance"? I’ve done an original survey looking at just this!
There actually is an “anti-growth coalition” and “pro-growth coalition” out there, but the make up of both of these creates some very strange bedfellows…
There actually is an “anti-growth coalition” and “pro-growth coalition” out there, but the make up of both of these creates some very strange bedfellows…
What Do The Public Actually Think About Economic Growth, Technological Change and "Abundance"?
There's an "anti-growth coalition" and a “pro-growth coalition” out there, it’s just not necessarily the people you think it is
jamesbreckwoldt.substack.com
October 8, 2025 at 11:01 AM
This is just generally very good (new data, new analysis!).
Lots of fun stuff - but the thing that jumped out to me is that Labour's rhetoric on AI *actually reflects their real electorate's positivity on automation* - in contrast to Labour's imaginary electorate (skilled manual workers).
Lots of fun stuff - but the thing that jumped out to me is that Labour's rhetoric on AI *actually reflects their real electorate's positivity on automation* - in contrast to Labour's imaginary electorate (skilled manual workers).
I'm working on a new model/visualisation - ideally eventually interactive - where you can look at the political compass in terms of where voters 'perceive' parties to be in value space (arrows)/how 'competent' they perceive them to be and how that turns into voting intention/behaviour.
October 7, 2025 at 11:33 AM
I'm working on a new model/visualisation - ideally eventually interactive - where you can look at the political compass in terms of where voters 'perceive' parties to be in value space (arrows)/how 'competent' they perceive them to be and how that turns into voting intention/behaviour.