Very happy to be a member of the CESifo network. I had a great time participating in the Area Conference on Economics of Education last month and presenting my work (w/ Mathias Mørk) on the estimation of teacher value-added. Working paper coming soon (hopefully)!
📢 Call for papers:
🚀 4th CESifo Junior Workshop on the Economics of Education 🤩
30-31 March 2026, Munich
Keynote: Michela Carlana (Harvard)
PhD students & early postdocs, please apply!
www.ifo.de/en/cesifo/ev...
Deadline: 11 Jan 2026
@caterinapavese.bsky.social @mtotarelli.bsky.social
Forthcoming in EJ: ‘Genetic and Socioeconomic Achievement Gaps in Elementary School’ by Mikkel Aagaard Houmark, Victor Ronda, Esben Agerbo, Preben Bo Mortensen, Michael Rosholm doi.org/10.1093/ej/ueaf033 @mikkelhoumark.bsky.social @michaelrosholm.bsky.social @resmedia.bsky.social
👏Congrats @anaisfabre.bsky.social on winning the 𝗖𝗘𝗦𝗶𝗳𝗼 𝗬𝗼𝘂𝗻𝗴 𝗔𝗳𝗳𝗶𝗹𝗶𝗮𝘁𝗲 𝗔𝘄𝗮𝗿𝗱 𝗶𝗻 𝗘𝗰𝗼𝗻𝗼𝗺𝗶𝗰𝘀 𝗼𝗳 𝗘𝗱𝘂𝗰𝗮𝘁𝗶𝗼𝗻 𝟮𝟬𝟮𝟱 for her paper "The Geography of Higher Education and Spatial Inequalities" and to runners-up @mikkelhoumark.bsky.social &
Genia Rachkovski for great presentations!
@ifoeducation.bsky.social
The following presentation still on social science genomics (yeah!! We're finally taking over entire sessions in econ conferences)
@mikkelhoumark.bsky.social estimating the technology of skill formation disentangling the differences by gender, genes, and parental SES
Super cool! And you didn't even mention the part that was most interesting/surprising to me: The apparent lack of shared environmental confounding. Great work.
Extremely excited to share the first effort of the Revived Genomics of Personality Consortium: A highly-powered, comprehensive GWAS of the Big Five personality traits in 1.14 million participants from 46 cohorts. www.biorxiv.org/content/10.1...
Hvorimod hvis det er measurement error, der er forklaringen, så bør drengene stadig få lavere eksamenskarakterer betinget på at have samme årskarakterer.
Har I prøvet at køre regressionen hvor I bytter om på års- og eksamenskarakterer? Ud fra logikken i analysen så bør man vel så finde at der er en positiv effekt af fx at være dreng - for hvis en dreng og en pige har samme årskarakter (som er biased imod drenge), så må drengen vel være dygtigere?
Og selv hvis vi antager at eksamen er et unbiased mål for det faglige niveau og at to elevgrupper med samme faglige niveau også bør få samme årskarakter, så er det ikke nødvendigvis et problem at der stadig er forskel når I kontrollerer for eksamenskarakteren - af de to statistiske grunde jeg nævner
Mon ikke der også sniger sig adfærd ind i eksamenskarakteren? Din personlighed påvirker jo fx, hvor meget du forbereder dig til eksamen, og måske også selve din præstation gennem nervøsitet osv. Så det kan godt være at idealet er at karakteren kun skal afspejle det faglige, men tror vi på det?
🚨 REPLICATION REPORT UPDATE: One year ago, a tweet by John Holbein alerted me, @ollefolke.bsky.social, and @jopieboy.bsky.social to a paper with a shocking result about Sweden’s law criminalizing the purchase of sex.🧵
Endelig så kan forskellige mål jo være uperfekte på forskellig vis, og selvom de sigter efter at bedømme samme faglighed, så kan de jo være påvirket af andre forhold på forskellige måder. Så måske supplerer de bare hinanden? Eller også er det eksamenskaraktererne der er biased imod piger osv.?
For det andet så er eksamenskarakterer jo påvirket af om du er fx dreng eller pige, dvs. I kontrollerer for et outcome, hvilket kan forårsage alle mulige korrelationer med andre variable (collider bias). Det gør det ligeledes vanskeligt at se at grupperne "bør" få samme årskarakter.
For det første er en eksamen jo et uperfekt mål for dygtighed, så hvis du sammenligner fx en dreng og en pige der har fået samme eksamenskarakter, så vil det bedste bud stadig være at pigen er dygtigere (fordi der er en gennemsnitlig forskel på grupperne). Dvs. at de ikke "bør" få samme årskarakter.
Hej Anders. Jeg synes godt nok, det er en voldsom konklusion på baggrund af den analyse. Et par overvejelser:
Ja, det er rigtigt. Så man kan jo næsten tale om at sociale medier sørger for at korrigere den fejlopfattelse. Men det virker så også plausibelt at der er tale om en overkorrektion, måske drevet af, som du foreslår, at man følger populære brugere frem for bare sine venner. Interessant uanset hvad!
Det er faktisk forventeligt at de fleste vil være "mindre populære" (eller i hvert fald have færre venner) end deres venner. Se: en.wikipedia.org/wiki/Friends.... Men ikke desto mindre virker det som en voldsom forskel, I finder!
Men hvis man nu antager at omkostningerne er tæt på 0, ligesom i dokumentaren, så er det jo faktisk næsten ingenting!
Quite striking! But I guess that correlation also includes common experiences that we do not typically think of as family background ((often) having the same teacher, same peers, in general growing up at the same time) - for regular siblings, this would largely be unshared environments, right?
Do genetic effects on education work through social mechanisms? And does the social environment moderate these effects? I’m addressing these RQs with brilliant @gaiaghirardi.bsky.social and @mikkelhoumark.bsky.social in a new project funded by the Independent Research Fund Denmark!
In my reading they present the model as equivalent (in terms of bias) to a family fixed effects model. And that model also solves the issue of cor(g_ij, e_ij). But then we agree that they are only equivalent in a world with cor(g_ij, e_j) but no cor(g_ij, e_ij)?
The idea is that "Because the environmental effect, e_j, does not vary within families, it is mechanically uncorrelated with the
sibling difference in phenotypes, y_1j − y_2j". But if there was an individual-specific environment, e_ij, it would be correlated with g_ij and y_1j − y_2j, i guess?
Interesting! But it seems that this method works only when there is no individual-specific ("nonshared") environment, which is a very strong assumption - or am I missing something?
🔊Next conference of the European Social Science Genetics Network (ESSGN) in Bristol, 22-23 May.
✅Keynote by @michellemeyer.bsky.social
✅Panel on the current frontier of MR w/ George Davey-Smith (Bristol), Dalton Conley (Princeton), and Alice Carter (Novo-Nordisk).
essgn.org/essgn-confer...
Hi #EconSky,
Not sure if the Starter Pack party's over, but I've made an Econ Starter Pack of Starter Packs! 😄
It's a work in progress, so I may have missed some. Let me know if there's anything to add—DMs are open!
Thanks for support @economista.bsky.social!
docs.google.com/spreadsheets...
BÆM is back! Come and join us in beautiful Bristol for the 2nd edition of the Bristol Applied Economics Meetings:
Development Economics: May 6-7
Gender, Diversity, and Human Capital: May 8-9
Submit your paper by Feb 28. Travel support is available. baem.info. 1/4
Great list - would love to be on it!
Would like to be added!
Hi! Here to see what the fuss is about. #EconSky