LightNews
Jean-Baptiste
@jbber.bsky.social
42 followers 130 following 110 posts
Data analyst, social sciences. I have a particular interest in data visualization and survey design.
Posts Media Videos Starter Packs
Reposted by Jean-Baptiste
rmcelreath.bsky.social
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 1h
Looks like a good guide - the general data cleaning part is a lean intro to some very common issues in all sorts of data. Would be great if every phd who touches raw data was offered a short course in these basics (in R or Python or whatever HipsterScript) cleaning-data-r.ala.org.au/2_general-cl...
1 5 11
Jean-Baptiste @jbber.bsky.social · 1h
The giveaway on the homepage was that you never put a period at the end of a title.
Jean-Baptiste @jbber.bsky.social · 3h
A more pessimistic hypothesis could that in countries having this kind of specific law (while it could be covered by more general laws), the objective is to target some minorities. But this is a wild guess, again.
Jean-Baptiste @jbber.bsky.social · 3h
Surprising this isn’t already covered by anti-fraud laws, as in other countries appearing in grey on the map. Just a wild guess, but maybe a specific law exist because this type of fraud is harder to fight under general anti-fraud laws?
1
Jean-Baptiste @jbber.bsky.social · 20h
il y a un cartographe dans les ITA, pas vraiment administratif
1
Jean-Baptiste @jbber.bsky.social · 20h
Une suggestion de lecture sur le sujet : shs.cairn.info/la-comedie-h...
La comédie humaine du travail
Sociologie clinique
shs.cairn.info
1
Jean-Baptiste @jbber.bsky.social · 20h
"alors que les gens du CNRS sont normalement formés au management, aux risques psycho-sociaux et la communication non-violente". Dans cette citation, le "alors que" est surprenant: toutes ces choses sont instrumentalisées pour présenter les conflits organisationnels comme une problématique psy.
1 1
Jean-Baptiste @jbber.bsky.social · 1d
Once I had a low-level management position at a university. They wanted me to take a course in management, that used pseudo-scientific concepts (e.g. en.wikipedia.org/wiki/Neuro-l...). A thing I remember is the instructor implying it may be OK to lie to employees. I left the position soon after.
Neuro-linguistic programming - Wikipedia
en.wikipedia.org
1 3
Reposted by Jean-Baptiste
dem8z.bsky.social
Deborah McDowell @dem8z.bsky.social · 1d
PLEASE SPREAD THE WORD!!

The University of Texas-Austin is beginning a process to eliminate the Black Studies, Latino Studies, and Gender Studies departments in the College of Liberal Arts. This is a grave threat to the educational liberty of students, faculty, staff, and the people of Texas. 1/
26 1K 1.3K
Reposted by Jean-Baptiste
georgiatomova.bsky.social
Georgia Tomova @georgiatomova.bsky.social · 5d
We should do a study on how much of the funded applied research suffers from problems that the unfunded methods research could have helped prevent or resolve
2 5 19
Jean-Baptiste @jbber.bsky.social · 5d
(that being said, perhaps this is the best solution)
2
Jean-Baptiste @jbber.bsky.social · 5d
If it works for your current problem, it's not dumb, it's creative :) Now, it might not work if we reuse it in situations where we have data that makes it break, but sometimes we just need a quick fix before finding a better long-term solution for future occurrences of the problem.
1 2
Jean-Baptiste @jbber.bsky.social · 6d
Educational context, to show how bad it is?
3
Reposted by Jean-Baptiste
pwgtennant.bsky.social
Peter Tennant @pwgtennant.bsky.social · 11d
"Uncooperative statistician": the term used (typically by a senior clinician) to describe a well-trained and knowledgeable statistician who refuses to conduct flawed or fraudulent research.
3 14 53
Jean-Baptiste @jbber.bsky.social · 13d
now, it doesn't solve the issue for some blog posts, social media, etc.
1
Jean-Baptiste @jbber.bsky.social · 13d
Just thought of it: style guides usually require (or should require) authors to spell out completely acronyms the first time they occur in text writing.wisc.edu/handbook/doc... This solves the ambiguity of "CI" for published papers (except perhaps in papers where the two kinds of CI are mentioned)
APA Usage and Style Guidelines
Abbreviations APA rules for abbreviations state: Acronyms and abbreviations must be spelled out completely on initial appearance in text. Use only if abbreviation is conventional, is apt to be familia...
writing.wisc.edu
1 1
Jean-Baptiste @jbber.bsky.social · 13d
This is a good observation, but unfortunately it also makes me think of this comic strip xkcd.com/927/
Standards
xkcd.com
1 1
Jean-Baptiste @jbber.bsky.social · 13d
Let's expect to see more things like this: www.justice.gov/usao-nh/pr/e...
Eight Defendants Indicted in International Conspiracy to Bill $10 Million for Fraudulent Market Survey Data
www.justice.gov
2
Jean-Baptiste @jbber.bsky.social · 13d
This is yet another risk adding up to the issue of survey response falsification: www.aapor.org/wp-content/u...
www.aapor.org
1
Reposted by Jean-Baptiste
dingdingpeng.the100.ci
Julia M. Rohrer @dingdingpeng.the100.ci · 14d
A lot of psych is already conducted with online convenience samples & ppl are probably excited about silicon samples bc it would allow them to crank out more studies for even less 💸

How about we reconsider the idea that sciencey science involves collecting own data.
www.science.org/content/arti...
AI-generated ‘participants’ can lead social science experiments astray, study finds
Data produced by “silicon samples” depends on researchers’ exact choice of models, prompts, and settings
www.science.org
13 71 230
Jean-Baptiste @jbber.bsky.social · 13d
In addition, how are people supposed to respond to the plethora of surveys they are subjected to on a daily basis? Respondents should be treated with respect and their attention as a limited resource, which is something at odds with the idea that all this should be a rite of passage for researchers.
1
Jean-Baptiste @jbber.bsky.social · 14d
effectivement, on peut difficilement tirer de conclusions à partir de ce sondage dans le cas d'élections avec un élément de variabilité spatiale.
Jean-Baptiste @jbber.bsky.social · 14d
First step before that: generate your questionnaire and questions with chatGPT, and don't revise and test it properly, because who cares.
1
Jean-Baptiste @jbber.bsky.social · 14d
donc grosso modo, un peu moins de gens détestent le RN comparé aux autres partis, mais les gens qui le détestent le détestent très fortement.
1
Jean-Baptiste @jbber.bsky.social · 14d
c'est un des partis avec le plus haut % de "très mauvaise opinion" dans ce sondage (avec les autres partis d'extrême-droite et la France Insoumise), mais il fait globalement mieux que les autres partis mainstream si on regroupe ensemble "très mauvaise opinion" et "assez mauvaise opinion".
1