Ian Hussey

@ianhussey.mmmdata.io

Meta-scientist and psychologist. Senior lecturer @unibe.ch‬. Chief recommender @error.reviews. "Jumped up punk who hasn't earned his stripes." All views a product of my learning history. If behaviorism did not exist, it would be necessary to invent it.

@rmcelreath.bsky.socialReposted by @ianhussey.mmmdata.io

An extraordinary example of a paper that cannot logically address its estimand: effect of restricting under-18 playtime. But sample has no age data, and most of the sample is adults. Reviewer #3 even noted this, and authors agreed. See link in quoted for details.

You said: “It’s not clear from the text why the authors believe that the youth effect is underestimated. If the youth population is about 6% of the sample, this means the first stage effects of the policy change should affect the tails of the distribution. So any estimates of effects on the full population which is primarily adults do not necessarily reflect what is happening with the youth. Perhaps clarifying what is meant with “underestimated” is sufficient, or perhaps the authors meant that the increases in adult utilization is underestimated. As it stands, I think “underestimated” is incorrect because the youth effects could be negative, positive, or neutral and we have no idea what the magnitude is.” Our response: Your point is extremely well-taken: We have removed the offending text from the manuscript (i.e. we have removed ‘Thus, the figures generated here will naturally underestimate changes to the prevalence of heavy playtime amongst young people.’). This is in line with the more general revision of language throughout the text to tend to be more conservative.
@floriannaudet.bsky.socialReposted by @ianhussey.mmmdata.io

🍫 or 💩? About the effect size reported in a trial that shaped the estimated effect of fluoxetine in meta-analyses of pediatric depression, my stellar co-author @ploederl.bsky.social says: “It’s the kind of effect size you only get if you ask people whether they prefer chocolate or faeces.” 👇👇👇

@ianhussey.mmmdata.io

100%. When I was in that world, I heard the term “functional analysis” thrown around daily by CBS folk, and never seen a single functional analysis performed or published. Iwata, Skinner etc would have things to say.

@ianhussey.mmmdata.io

So, again, but in public this time: the idionomic approach folk would greatly benefit from engaging the literature on these approaches to reduce research waste, avoid rediscovering the wheel (but worse), etc.

@ianhussey.mmmdata.io

As for the 'but worse' bit - my recollection is that none of the effects reported in that paper would survive multiple testing corrections, which arguably should have been applied.

@ianhussey.mmmdata.io

Around 2019 I spoke to several folk in the ACT community about how their interest in so-called idionomic analyses was just rediscovering EMA studies, multilevel models and temporal network analysis. No one was interested - they implication was that they 'needed' a bespoke CBS branded approach.

@ianhussey.mmmdata.io

Hayes has been banging this drum about idionomic analyses for 6+ years, with folk like @jamiecummins.bsky.social pointing out that its reinventing-the-wheel-but-worse for just as long.

My experience is that the ACT community is insufficiently interested and aware of existing methods+stats work.

@jamiecummins.bsky.socialReposted by @ianhussey.mmmdata.io

My lukewarm take is that when people disagree about whether LLMs can do X (be it reasoning, creativity, etc), the substantive disagreement is really about what X means, rather than about LLMs themselves. Aka Wittgenstein was right

Older posts