Jon Mellon
@jonmellon.bsky.social
Co-director British Election Study. Political Scientist and Data Scientist. Political science methods/political behavior/causal inference. Posts do not represent employer.
Pinned
Jon Mellon
@jonmellon.bsky.social
· Nov 14
Out now open access at
@ajpseditor.bsky.social. 194 potential exclusion-restriction violations for studies using weather as an instrumental variable onlinelibrary.wiley.com/doi/full/10....
@ajpseditor.bsky.social. 194 potential exclusion-restriction violations for studies using weather as an instrumental variable onlinelibrary.wiley.com/doi/full/10....
Reposted by Jon Mellon
I am on paternity leave, so not promoting this properly, but lots of Britons have stable but idiosyncratic views , as shown in this book doi.org/10.1093/9780...
Validate User
doi.org
February 3, 2026 at 4:22 PM
I am on paternity leave, so not promoting this properly, but lots of Britons have stable but idiosyncratic views , as shown in this book doi.org/10.1093/9780...
Reposted by Jon Mellon
Yes that’s definitely true. I have a lot of cites to my “only use this method if you run a validation for your use case” papers that read “Mellon (2014) shows that this method is valid for my use case”
January 30, 2026 at 2:36 PM
Yes that’s definitely true. I have a lot of cites to my “only use this method if you run a validation for your use case” papers that read “Mellon (2014) shows that this method is valid for my use case”
Yes that’s definitely true. I have a lot of cites to my “only use this method if you run a validation for your use case” papers that read “Mellon (2014) shows that this method is valid for my use case”
January 30, 2026 at 2:36 PM
Yes that’s definitely true. I have a lot of cites to my “only use this method if you run a validation for your use case” papers that read “Mellon (2014) shows that this method is valid for my use case”
I think the LLM hallucinated citations problem is going away (because it’s fundamentally easy to check if a paper exists using an API) but the much harder problem (especially for faculty reviewing student work) will be hallucinating the content of citations
January 30, 2026 at 1:18 PM
I think the LLM hallucinated citations problem is going away (because it’s fundamentally easy to check if a paper exists using an API) but the much harder problem (especially for faculty reviewing student work) will be hallucinating the content of citations
I’d want to see a very convincing null in a placebo test on pro government attitudes before I’d buy this wasn’t an issue
January 28, 2026 at 9:53 PM
I’d want to see a very convincing null in a placebo test on pro government attitudes before I’d buy this wasn’t an issue
have a look at a flexible form (e.g. spline) of the first stage relationship and make sure you pick a formulation that satisfies monotonicity at the very least
January 28, 2026 at 9:40 PM
have a look at a flexible form (e.g. spline) of the first stage relationship and make sure you pick a formulation that satisfies monotonicity at the very least
i.e. it's more about the source of the money ("government's doing good things for me!") than having the money ("I'm not going to be evicted!")
January 28, 2026 at 9:37 PM
i.e. it's more about the source of the money ("government's doing good things for me!") than having the money ("I'm not going to be evicted!")
exclusion restriction violations: do payments from government (or anticipation of imminent payment from government) form a plausible alternative pathway to political outcomes (I'd argue yes)
January 28, 2026 at 9:36 PM
exclusion restriction violations: do payments from government (or anticipation of imminent payment from government) form a plausible alternative pathway to political outcomes (I'd argue yes)
These are more about the estimand you're targeting than invalidating the method
January 28, 2026 at 9:36 PM
These are more about the estimand you're targeting than invalidating the method
Relatedly, do we think that the LATE is relevant to the ATE here? The kind of insecurity you're worried about when you're looking down the barrel of a rough couple of weeks might be different to that caused by longer-term insecurity.
January 28, 2026 at 9:36 PM
Relatedly, do we think that the LATE is relevant to the ATE here? The kind of insecurity you're worried about when you're looking down the barrel of a rough couple of weeks might be different to that caused by longer-term insecurity.
One challenge is the length of time that we think political beliefs take to form. i.e. do we expect an instantaneous response to financial security or are we doing some kind of cumulative exposure update?
January 28, 2026 at 9:36 PM
One challenge is the length of time that we think political beliefs take to form. i.e. do we expect an instantaneous response to financial security or are we doing some kind of cumulative exposure update?
It's a bit annoying that the default python setup for datascience (jupyter notebooks) is a bad fit for git version control (you end up with inconsequential chunk output merge issues). I wish they'd gone with a stricter separation of code and output like rmd.
January 28, 2026 at 9:28 PM
It's a bit annoying that the default python setup for datascience (jupyter notebooks) is a bad fit for git version control (you end up with inconsequential chunk output merge issues). I wish they'd gone with a stricter separation of code and output like rmd.
Reposted by Jon Mellon
If it makes you feel better seahorses are more closely related to horses than sharks
January 26, 2026 at 1:00 PM
If it makes you feel better seahorses are more closely related to horses than sharks
If it makes you feel better seahorses are more closely related to horses than sharks
January 26, 2026 at 1:00 PM
If it makes you feel better seahorses are more closely related to horses than sharks
Interesting. Any chance this is a cultural specific tendency against choosing extreme survey answer options?
January 18, 2026 at 7:50 PM
Interesting. Any chance this is a cultural specific tendency against choosing extreme survey answer options?
So are the differences for Asian women compared to white women?
January 18, 2026 at 6:49 PM
So are the differences for Asian women compared to white women?
What’s the reference groups for these differences?
January 18, 2026 at 4:41 PM
What’s the reference groups for these differences?
Reposted by Jon Mellon
Psychologists stop trying to do psychological science
no wait
no wait
Economists stop trying to do political science
Physicists stop trying to do economics
January 15, 2026 at 2:06 PM
Psychologists stop trying to do psychological science
no wait
no wait
We’re going to end up at the Guinness brewery aren’t we?
January 15, 2026 at 2:04 PM
We’re going to end up at the Guinness brewery aren’t we?
Reposted by Jon Mellon
When you stop borrowing our methodologies
January 15, 2026 at 1:59 PM
When you stop borrowing our methodologies
Not actually my view. Disciplinary boundaries are overrated. My main complaint is when people don’t read the discipline they’re working in and miss obvious things
January 15, 2026 at 1:48 PM
Not actually my view. Disciplinary boundaries are overrated. My main complaint is when people don’t read the discipline they’re working in and miss obvious things
Economists stop trying to do political science
Physicists stop trying to do economics
January 15, 2026 at 1:45 PM
Economists stop trying to do political science
This is probably overly conservative but if I'm treating the proof as a black box I think I have to be
January 15, 2026 at 2:14 AM
This is probably overly conservative but if I'm treating the proof as a black box I think I have to be
A couple of senses:
1) not in the assumptions.lean file I told the LLM to use
2) axioms, sorry, etc
3) the LLMs at least think it's possible to sneak assumptions "in the proof term but not in the theorem statement" but I'll admit I'm not totally clear on how this works.
1) not in the assumptions.lean file I told the LLM to use
2) axioms, sorry, etc
3) the LLMs at least think it's possible to sneak assumptions "in the proof term but not in the theorem statement" but I'll admit I'm not totally clear on how this works.
January 15, 2026 at 2:13 AM
A couple of senses:
1) not in the assumptions.lean file I told the LLM to use
2) axioms, sorry, etc
3) the LLMs at least think it's possible to sneak assumptions "in the proof term but not in the theorem statement" but I'll admit I'm not totally clear on how this works.
1) not in the assumptions.lean file I told the LLM to use
2) axioms, sorry, etc
3) the LLMs at least think it's possible to sneak assumptions "in the proof term but not in the theorem statement" but I'll admit I'm not totally clear on how this works.
I'm currently using this script to try and identify any hidden assumptions and I think my annotations are correct (and therefore I know what's being assumed)
github.com/jon-mellon/C...
github.com/jon-mellon/C...
January 15, 2026 at 1:51 AM
I'm currently using this script to try and identify any hidden assumptions and I think my annotations are correct (and therefore I know what's being assumed)
github.com/jon-mellon/C...
github.com/jon-mellon/C...