Jan Vanhove
@janhove.bsky.social
Senior lecturer at the Department of Multilingualism at https://unifr.ch/. Master's student statistics at https://unibe.ch/. Saxophonist with https://beatmoustache.ch/. Aficionado of Danish ska.
Blog & website: https://janhove.github.io
Blog & website: https://janhove.github.io
Thanks for the suggestion, I'm looking into it.
December 18, 2025 at 12:13 PM
Thanks for the suggestion, I'm looking into it.
Could be integer overflow!
December 17, 2025 at 12:43 PM
Could be integer overflow!
Den ena norskkurs jag tog för tjugo år sedan räcker tyvärr inte till - även om jag tycker om att läsä norsk litteratur 😬
December 17, 2025 at 12:02 PM
Den ena norskkurs jag tog för tjugo år sedan räcker tyvärr inte till - även om jag tycker om att läsä norsk litteratur 😬
Lugnt! Blistering barnacles.
December 17, 2025 at 11:02 AM
Lugnt! Blistering barnacles.
Yeah, and with glm in particular, things can get pretty weird when adding or removing predictors in this way.
December 17, 2025 at 10:52 AM
Yeah, and with glm in particular, things can get pretty weird when adding or removing predictors in this way.
I agree. I'm doing the analysis the authors think they did and find different results than they did. I don't think these analyses are *good*.
December 17, 2025 at 10:48 AM
I agree. I'm doing the analysis the authors think they did and find different results than they did. I don't think these analyses are *good*.
Yeah. So now I'm wondering if I ought to send this to the outlets that covered the original study.
December 17, 2025 at 10:25 AM
Yeah. So now I'm wondering if I ought to send this to the outlets that covered the original study.
Im Zweifelsfall ska man fyra oftare en een beetje vroeger.
December 17, 2025 at 10:23 AM
Im Zweifelsfall ska man fyra oftare en een beetje vroeger.
They make me frown, too! The context here is that the authors wanted to show that their results are 'robust' even when controlling for per-capita GDP etc. In actual fact, they controlled for GDP (not per-capita) instead, so I merely redid the analysis they wanted to do but didn't.
December 17, 2025 at 10:03 AM
They make me frown, too! The context here is that the authors wanted to show that their results are 'robust' even when controlling for per-capita GDP etc. In actual fact, they controlled for GDP (not per-capita) instead, so I merely redid the analysis they wanted to do but didn't.
Danke 😊 danke dir fürs teilen, hilft enorm!
December 17, 2025 at 7:45 AM
Danke 😊 danke dir fürs teilen, hilft enorm!
not to mention a bunch of other stuff: janhove.github.io/posts/2025-1...
Jan Vanhove :: Blog - Does multilingualism really protect against accelerated ageing? Some critical comments
janhove.github.io
December 17, 2025 at 7:03 AM
not to mention a bunch of other stuff: janhove.github.io/posts/2025-1...
it was a python session ;) a few other problems with this article, too: janhove.github.io/posts/2025-1...
Jan Vanhove :: Blog - Does multilingualism really protect against accelerated ageing? Some critical comments
janhove.github.io
December 17, 2025 at 7:03 AM
it was a python session ;) a few other problems with this article, too: janhove.github.io/posts/2025-1...
My hunch: it's not the # of those samples that are effective, but what effectively the sample size is. (Large nr. of people vs. Nr. of large people.)
November 25, 2025 at 4:37 PM
My hunch: it's not the # of those samples that are effective, but what effectively the sample size is. (Large nr. of people vs. Nr. of large people.)
Ha, never knew the German mnemonic! (I use "Maak voort, aardig meisje! Jan spuit u nat! Plons!")
November 24, 2025 at 6:20 PM
Ha, never knew the German mnemonic! (I use "Maak voort, aardig meisje! Jan spuit u nat! Plons!")
Of course. I just think standardised ESs are rarely a good idea. :)
November 20, 2025 at 9:26 PM
Of course. I just think standardised ESs are rarely a good idea. :)
In maths, sets can be open, closed, both ("clopen", "abgeschloffen"), or neither.
November 8, 2025 at 4:37 AM
In maths, sets can be open, closed, both ("clopen", "abgeschloffen"), or neither.