Charles Driver
@charlesdriver.bsky.social
Asst Prof UZH Psychology - quant methods, dynamic systems, human development, psychology. @CharlesDriverAU
Honestly I'd be happy if we'd just name far less stuff and force people to describe what they really think they're doing...
November 8, 2025 at 1:47 PM
Honestly I'd be happy if we'd just name far less stuff and force people to describe what they really think they're doing...
Worth being aware of but also a mathy inevitability.
November 4, 2025 at 5:35 AM
Worth being aware of but also a mathy inevitability.
That's not how it works in psych stats sorry 😆
October 31, 2025 at 2:36 PM
That's not how it works in psych stats sorry 😆
Does an effect suddenly become interesting at a specific point though? I can see some value to the null hyp when there is a directional hyp that supports an interesting theory, struggle to see value in a SESOI for the kind of case described though.
October 31, 2025 at 11:23 AM
Does an effect suddenly become interesting at a specific point though? I can see some value to the null hyp when there is a directional hyp that supports an interesting theory, struggle to see value in a SESOI for the kind of case described though.
If they were bigger, would the model / story be compelling? Would alternate causal stories (eg common causes) have been considered? Knowing typical clpm research, I doubt it :/
October 31, 2025 at 6:30 AM
If they were bigger, would the model / story be compelling? Would alternate causal stories (eg common causes) have been considered? Knowing typical clpm research, I doubt it :/
You really think the problems are *inherent* to the approach? I could see that getting rid of the approach leads to a general improvement (because of low standards for interpretation etc), but also see value in the approach in ideal cases...
October 29, 2025 at 7:57 AM
You really think the problems are *inherent* to the approach? I could see that getting rid of the approach leads to a general improvement (because of low standards for interpretation etc), but also see value in the approach in ideal cases...
I guess if we let 'true' do enough work I agree with you for a while, but I mostly find it pretty difficult to find that level of expertise around...
October 26, 2025 at 1:25 PM
I guess if we let 'true' do enough work I agree with you for a while, but I mostly find it pretty difficult to find that level of expertise around...
Why? I'm not convinced you're wrong, but I can also easily imagine current forms of AI displacing GP's with a few more years development. Some level of displacement coupled with some level of collaboration seems inevitable to me.
October 26, 2025 at 12:47 PM
Why? I'm not convinced you're wrong, but I can also easily imagine current forms of AI displacing GP's with a few more years development. Some level of displacement coupled with some level of collaboration seems inevitable to me.
To complement the h-index, maybe a 'pay for play' index would be nice too? avg publishing cost per citation?
October 25, 2025 at 7:39 PM
To complement the h-index, maybe a 'pay for play' index would be nice too? avg publishing cost per citation?
Truly wonderful!
October 23, 2025 at 6:47 AM
Truly wonderful!
Favourite white noise band?
October 20, 2025 at 11:54 AM
Favourite white noise band?
I didn't see anywhere in the paper arguing to correct only when the same hypothesis was being examined, did I miss it or you just think differently? Would rather see a lower alpha across the board than alpha based on arbitrary combination of different tests because they appeared in the same paper.
October 18, 2025 at 10:24 AM
I didn't see anywhere in the paper arguing to correct only when the same hypothesis was being examined, did I miss it or you just think differently? Would rather see a lower alpha across the board than alpha based on arbitrary combination of different tests because they appeared in the same paper.
Shouldn't we correct across all articles in an issue then? Or in the entire journal? Or all journals in the field? Seems to totally miss the nuance, but maybe it's me missing the argument?
October 18, 2025 at 10:15 AM
Shouldn't we correct across all articles in an issue then? Or in the entire journal? Or all journals in the field? Seems to totally miss the nuance, but maybe it's me missing the argument?
Nice explanations. One suggestion: Not sure std error is a very useful comparison when some estimates are biased. rmse of true - estimated across n sims would be a much more interesting SEM v regression comparison for the variance side of bias v variance, I think :)
October 14, 2025 at 5:47 AM
Nice explanations. One suggestion: Not sure std error is a very useful comparison when some estimates are biased. rmse of true - estimated across n sims would be a much more interesting SEM v regression comparison for the variance side of bias v variance, I think :)
Now I want to see someone with the psych follow up: 'predictors of interface preference - tidy vs data.table'
October 14, 2025 at 5:27 AM
Now I want to see someone with the psych follow up: 'predictors of interface preference - tidy vs data.table'
Weird it didn't used to do that will fix later thanks!
October 10, 2025 at 2:23 PM
Weird it didn't used to do that will fix later thanks!
Would be interesting to see perspectives on this by age. Benefiting from pre entshittification social media was really quite nice sometimes.
October 8, 2025 at 6:54 PM
Would be interesting to see perspectives on this by age. Benefiting from pre entshittification social media was really quite nice sometimes.
Intervening centrally should be great if you can really intervene and hold it there. But is the view that symptom networks are like social networks (influence flows equally in both directions) still a thing? When I read this it mostly makes it clear how differently I think of a mental disorder.
September 29, 2025 at 8:37 PM
Intervening centrally should be great if you can really intervene and hold it there. But is the view that symptom networks are like social networks (influence flows equally in both directions) still a thing? When I read this it mostly makes it clear how differently I think of a mental disorder.
#4 Relations between depression and self esteem across 3 different time scales / datasets, with @peterhaehner.bsky.social et al. Besides the multiple time scales, extending VARish models to cross-effects that influence over different time scale is unusual I think.
psycnet.apa.org/record/2025-...
psycnet.apa.org/record/2025-...
September 25, 2025 at 9:48 AM
#4 Relations between depression and self esteem across 3 different time scales / datasets, with @peterhaehner.bsky.social et al. Besides the multiple time scales, extending VARish models to cross-effects that influence over different time scale is unusual I think.
psycnet.apa.org/record/2025-...
psycnet.apa.org/record/2025-...
#3 Examining links between and within social and cognitive development, and how these differ with behaviour problems. Some big moderated growth + fluctuations models. With @mkdeserno.bsky.social et al.
osf.io/preprints/ps...
osf.io/preprints/ps...
September 25, 2025 at 9:36 AM
#3 Examining links between and within social and cognitive development, and how these differ with behaviour problems. Some big moderated growth + fluctuations models. With @mkdeserno.bsky.social et al.
osf.io/preprints/ps...
osf.io/preprints/ps...
#2 "Personality functioning as generalized correlated changes in personality traits" -- basically a large factor model over time with some common factor nuance. With @chopwood.bsky.social et al.
osf.io/preprints/ps...
osf.io/preprints/ps...
September 25, 2025 at 9:36 AM
#2 "Personality functioning as generalized correlated changes in personality traits" -- basically a large factor model over time with some common factor nuance. With @chopwood.bsky.social et al.
osf.io/preprints/ps...
osf.io/preprints/ps...