psforscher.bsky.social
Director of the CREME developmental meta-research team at Busara, a non-profit that does behavioral science in service of poverty alleviation. https://patrickforscher.com/
371 posts
1,531 followers
275 following
Prolific Poster
Conversation Starter
comment in response to
post
5. This is the moment of truth.
The metascience community could tout this as a win and fundraise off of it, cementing their legacy as useful idiots.
Or they can side with science, boldly speaking out about the bad-faith ways the Trump administration is co-opting their language and momentum.
comment in response to
post
This post was inspired by #NairobiEvidenceWeek, a five-day event that @busaracenter.bsky.social is co-organizing alongside CEGA and NIERA.
This may seem an odd time to host an event on evidence-based policymaking & development -- yet moments of crisis are when community can foster resilience
comment in response to
post
I wish it were six months ago!
comment in response to
post
thanks, that's helpful
comment in response to
post
i don't think this analysis works because you can use it on anything that's bad -- it's always possible to find something worse. the relevant question is whether the costs (environmental or whatever) justify the benefits
comment in response to
post
Yeah this analysis does not include any of the costs of training the model, nor does it analyze actual usage other than some guesswork.
Quite bad analysis imo
comment in response to
post
This analysis is quite bad. The real concern is about the entire environmental cost of training and running a model, not the cost of submitting one query
comment in response to
post
The environmental concern is mostly with the training costs though …
comment in response to
post
I am so stoked that this working paper is finally out. This method has a lot of potential to help improve upon some long-standing problems in development, such as de-contextualized interventions & entrenched power hierarchies within research teams
comment in response to
post
Some generalizable principles are absolutely important as starting points to solve a practical problem but not all valuable knowledge is of that form
comment in response to
post
I agree. But I think psychologists have made “generalizability” one of their central goals in an effort to seem more like a hard science. In my view this is one of the main obstacles that prevents psychologists from delving into the contextual details necessary to solve practical problems
comment in response to
post
That makes sense. There is a broad sense of complacency this time rather than of crisis
comment in response to
post
Do these interventions need to be generalizable? No.
But then I would also say that most current psychologists working to solve practical problems do not necessarily need to concern themselves with generalizability
comment in response to
post
For example, the main payment system in Kenya, M-Pesa, is basically a way to move payments to your phone carrier (mobile money). Gambling is a big problem on M-Pesa, so Kenyan research psychologist might well choose to study how to reduce gambling over M-Pesa
comment in response to
post
Yes they definitely intersect, though I would say that the outcomes and interventions we choose to spend our time on are a function of the populations we choose to study and who is doing that studying — so generalizability need not enter into the discussion
comment in response to
post
This would have been an interesting point to raise.
Here are a few articles on the rise of applied behavioral science
www.psychologicalscience.org/observer/gs-...
behavioralscientist.org/who-is-doing...
comment in response to
post
So, even though I agree that concerns about relevance have not been a huge part of the crisis narrative (one exception that I wrote, here: www.psychologicalscience.org/observer/ent...), the applications are happening without the input or even knowledge of psychologists