Eiko Fried Profile picture
Jan 23 5 tweets 2 min read
1/5 This review by Gonthier (2022) tackles a crucial topic: are non-verbal intelligence tests culture fair? This is important because you often see the reasoning "ethnicity/race 1 has lower IQ than ethnicity/race 2, & it must be genetic because non-verbal tests are culture fair".
2/5 This takes ugly extremes such as that there is "some genetic component in Black–White differences in mean IQ" (Rushton & Jensen 2005) etc. So the review here really matters to address threats to such conclusions and can set the record straight.
3/5 Gonthier investigates numerous sources of evidence, from controlled lab experiments in the US to n=1 qualitative reports from ethnologists dating many decadesback, concluding that there is substantial evidence that non-verbal tests are *not* culture fair.
4/5 This is discussed in great detail in the paper, and summarized in this table. Some of these stories were a real learning experience for me as a scholar not working on intelligence—highly recommended.
5/5 Overall, Gonthier concludes that culture is deeply engrained in visuo-spatial reasoning tests, which "makes it impossible to draw clear-cut conclusions from average score differences between ethnic groups".

Here is the #openaccess paper:

…itiveresearchjournal.springeropen.com/articles/10.11…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Eiko Fried

Eiko Fried Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @EikoFried

Nov 30, 2022
Excited about @AnnaWysocki3 's talk on "Statistical Control Requires Causal Justification"

journals.sagepub.com/doi/full/10.11… Image
Highly recommended if you always wanted to know what colliders are or do—and if you ever "added x and y as covariates" to e.g. a linear regression because that is "what your field does".
Main take-home message: Image
Read 6 tweets
Nov 20, 2022
1/8 Valentin, I'll respond with a quote RT because then I can write a proper thread here (rather than having to add 1 tweet at a time in my response).

So my "record" is the 2017 paper on 52 symptoms of depression, which was rejected 8 times.
2/ This is in my part my 'fault' bc I kept submitting it to applied journals, but I really didn't want this paper in a "journal specialized on measurement" (quote from 6 rejection letters): I wanted to reach clinicians & applied researchers.
3/ But I had submitted several papers that year, so didn't feel too bad abt waiting. Today, paper has ~400 citations & spawned little mini-literature of folks doing similar analyses w other scales. You can find a bit of a summary in this tweet here:

Read 9 tweets
Nov 17, 2022
1/ So @UniLeiden has now "lost" a second Prof in a short period of time. In my reading of the news, he is no longer allowed on uni premises due to ‘extremely undesirable behaviour’, but keeps salary & title.
What did he do? See screenshot below from uni executive board. Image
2/ The other Prof we "lost" 3 yrs ago had committed fraud, tampered w data & grant applications, taken blood samples w/o ethics approval, fabricated experiments, removed participants, dropped & added authors (and was then hired by TU Dresden for .. I don't know exactly for what).
3/ Both cases reveal highly problematic practices that went on for years without uni doing something. And there is just no way *some* people in power weren't aware.
When practices did come to light, it is bc (often female & junior) folks spoke up, at their own peril.
Read 7 tweets
Oct 24, 2022
Lots of new followers in the last few weeks, so here's a short thread introducing you to some of the work we conducted since 2020.
Broadly speaking, our work tackles how to best 1⃣understand,2⃣measure, and3⃣model mental health problems.

🧵
1/

Pillar 1: 𝗧𝗵𝗲𝗼𝗿𝘆 𝗯𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗮𝗻𝗱 𝘁𝗲𝘀𝘁𝗶𝗻𝗴.

Don Robinaugh has led fanastic work on this topic. Our first recent paper I recommend is conceptual work on the importance of having clear theories.

journals.sagepub.com/doi/full/10.11…
2/

Another project led by Don is our work on a formalized theory for panic disorder, sort of walking the theory walk instead of just talking the theory talk ;).

This is still in preprint stage, but there'll be some updates and news on this soon.

psyarxiv.com/km37w/
Read 18 tweets
Oct 21, 2022
This thread has turned into neat lists of
(1) things that can go wrong with Qualtrics surveys (and how to circumvent that), and
(2) tricks of statistically interrogating data for anomalities that may cause unexpected findings.

Thx to all for contributing!
(in our case, the zero correlation was due to a Qualtrics export error that was language specific—it was present in only half the sample, and not in the other, leading to group-level data that looked fine overall and behaved appropriately, but didn't correlate w other measures)
( @oscar_olvera100 promised to let you know the outcome! )
Read 9 tweets
Oct 18, 2022
🚨New paper alert🚨
"Studying Mental Health Problems as systems, not syndromes"
is now published #OpenAccess in Current Directions.

I'll summarize the paper briefly here. ⬇️

journals.sagepub.com/doi/10.1177/09… Image
In the paper, I describe 2 barriers to progress that our field has not sufficiently grappled with.
Barrier (1): diagnostic literalism, i.e. mistaking mental health (MH) problems a person has with the diagnosis a person receives.

In short: MH problems ≠ diagnoses.
This becomes obvious when you look at the history of the DSM, which I briefly sketch in the paper. Diagnoses were meant as rough clinical proxies. They are not the kind of things that lend themselves well to e.g. biomarker discovery, or one-size-fits-all treatments.
Read 18 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(