iiinova — where your digital self keeps evolving

Reposted by Jakob Schwerter

Jeff Greene @jeffgreene.bsky.social · 15/06/2026

A great view of Tübingen with @jakobschwerter.bsky.social Allan Wigfield and @mattbernacki.bsky.social. Always great to visit the Hector Research Institute of Education Sciences and Psychology at @unituebingen.bsky.social.

Jakob Schwerter, Allan Wigfield, Matt Bernacki and me smiling for a selfie at the top of a hill looking down on Tübingen, Germany.

051

Reposted by Jakob Schwerter

Jeff Greene @jeffgreene.bsky.social · 08/11/2025

...Thus, at the moment, I remain skeptical about the benefits of #GenAI in learning, outside of expert-in-the-loop uses (e.g., a teacher using it to craft personalized versions of lesson materials, which the teacher can critically evaluate and refine because of their expertise)... (6/n)

112

Jakob Schwerter @jakobschwerter.bsky.social · 13/04/2025

Desto schwächer ist Bayern gegen den BvB?

000

Reposted by Jakob Schwerter

IFS TU Dortmund @ifs-tudortmund.bsky.social · 13/03/2025

New research video: Discover the importance of students’ social resource factors for academic achievement and well-being in elementary school by @jakobschwerter.bsky.social on YouTube! #Education #Learning #EducationResearch #ElementarySchool www.youtube.com/watch?v=OLEE...

042

Jakob Schwerter @jakobschwerter.bsky.social · 07/12/2024

Bundesliga Topspiel bis jetzt schon eher langweilig, oder? #bmgbvb

000

Reposted by Jakob Schwerter

Steve Haroz @steveharoz.com · 30/11/2024

This is really handy. #stats #CausalInference

An explanation of confound, collider, and mediator

28424

Jakob Schwerter @jakobschwerter.bsky.social · 29/11/2024

Ich finde es viel spannender und mich interessieren die Ergebnisse der anderen viel mehr. Hab vorher aber auch nur BvB-Spiele geschaut und ist jetzt immer noch so.

010

Jakob Schwerter @jakobschwerter.bsky.social · 24/11/2024

Sahin has not lost a game at home. Would be a perfect story if BVB would win against Bayern at home as well. But most of the times when I thought we would have a chance, we lost. So 🤷

000

Reposted by Jakob Schwerter

Jakob Wegmann @jakobwegmann.bsky.social · 23/11/2024

Hello #EconSky! 📢Excited to share my #JMP 📢 It provides a missing piece to understand how people react to income taxation. Tldr: By changing tax pre-payments governments can - with almost no costs - increase perceived work incentives of secondary earners and reduce the gender gap. Thread below👇🧵

To collect income taxes, almost all countries require employers to withhold monthly tax prepayments which are then fully credited against the final income tax liabilities of their employees. Despite being a fundamental component of income taxation systems worldwide, the impact of these withholding taxes on labor supply is poorly understood. We investigate their importance in the context of married couples in Germany where the withholding tax liability can be redistributed between spouses. We exploit a reform that reduced the withholding tax for some married women more than for others, while inducing no differences in income taxes. Using administrative data for the full population of German taxpayers, we estimate an elasticity of labor income with respect to the withholding tax eight years after the reform of 0.14. Additional evidence from a self-conducted survey suggests imperfect understanding of the tax system and limited pooling of resources within the household as the main mechanisms. As the majority of couples shift parts of the withholding tax liability from the husband to the wife, our results suggest that the increased withholding tax liability of married women contributes to their low labor supply. This highlights the need for governments to be aware of the distortion of labor supply incentives when the design of withholding taxes does not match actual income tax incentives.

26827

Jakob Schwerter @jakobschwerter.bsky.social · 23/11/2024

So similar in many ways, but so different in others.

000

Jakob Schwerter @jakobschwerter.bsky.social · 23/11/2024

I am collecting data in university courses over 2-3 cohorts. So good to see that learning behaviour is mostly similar across cohorts with a similar effect on exam performance (stay tuned for specific results). But what a pain to explain differences in taking the first or second exam, etc.

100

Jakob Schwerter @jakobschwerter.bsky.social · 21/11/2024

Is this only happening in econometrics or does something similar exist in #education #psychology ? I would be happy to try out something like this.

100

Jakob Schwerter @jakobschwerter.bsky.social · 19/11/2024

Apart from daily participation here, is there a website to practice these DAGs? Like guessthecorrelation.com?

guessthecorrelation.co

150

Jakob Schwerter @jakobschwerter.bsky.social · 14/11/2024

😂 waren leider sehr heiße Hot-Takes.

000

Reposted by Jakob Schwerter

Kou Murayama @koumurayama.bsky.social · 12/11/2024

My first postdoc @hkjach.bsky.social's main work has been finally published!! We assessed lots of curiosity personality traits and information-seeking task behaviors and examined the relationship. See what we found... www.pnas.org/doi/10.1073/...

pnas.org

0116

Jakob Schwerter @jakobschwerter.bsky.social · 12/11/2024

What is the Name of the R-Package you mentioned?

100

Reposted by Jakob Schwerter

Tim (Till) Fütterer @tillfuetterer.bsky.social · 10/11/2024

Exciting news! The @EARLI has funded our Emerging Field Group on “Automated Assessment of Teaching Effectiveness Using Multimodal Data.” We’re now entering the work phase and look forward to sharing insights. Check out our group: www.earli.org/efg-13 #EdResearch #AIinEducation

earli.org

EFG - 13 - EARLI

061

Jakob Schwerter @jakobschwerter.bsky.social · 21/10/2024

Hi Learning Scholars! Which publication is currently everyone’s fav to link log-data to SRL behavior? Or that shows which log-data help to measure SRL behavior? Or an overview of which variables generated from clicks reflect SRL behavior? #SRL #Clicks #LearningScholars

011

Reposted by Jakob Schwerter

Jeff Greene @jeffgreene.bsky.social · 18/10/2024

I don't think anyone has started a Learning Scholars starter pack, so here we go! Let me know if you'd like to be added, removed, etc. Please share! go.bsky.app/CBPm1dV

105525

Reposted by Jakob Schwerter

Rasenfunk @rasenfunk.social · 19/08/2024

Verteidigt Leverkusen seinen Titel? Stürzt Stuttgart ab? Rafft sich der FC Bayern? Beginnt beim BVB eine Ära? Wir blicken mit @evabohle.bsky.social und @kalorineki.bsky.social auf die Saison voraus – inklusive Tabellentipp: rasenfunk.de/bundesliga-m/468 #bundesliga

Tabellentipp von Eva, Karo, Max und den Hörer*innen. Gibt es im letzten Kapitel der verlinkten Folge auch zum Hören.

55014

Jakob Schwerter @jakobschwerter.bsky.social · 17/03/2024

Und in der Session 6-03 „Selbsttests, Lernverhalten und Leistung in der Hochschulbildung. Einblick aus digitalen Verhaltensspuren und Selbstberichten.“ präsentiere ich aktuelle Ergebnisse zu Selbsttest im Hochschulstudium

000

Jakob Schwerter @jakobschwerter.bsky.social · 17/03/2024

Interesse an #MachineLearning? In der Session 3-11 gibt es einen Einblick in die Verwendung von baumbasierten Methoden für die Imputation von fehlenden Werten und der Datenanalyse, sowie die Auswertung von Open-Ended.

100

Reposted by Jakob Schwerter

Matt Bernacki @mattbernacki.bsky.social · 12/02/2024

It's out! 🔬📐🏗️🏦🔭👩‍⚕️ 6+ years of partnership building, STEM Career media production, design & full experiments: "The [+] effect of an intervention personalizing mathematics to students’ career and popular culture interests on mathematics interest and learning." psycnet.apa.org/record/2024-...

162

Jakob Schwerter @jakobschwerter.bsky.social · 10/02/2024

Vor allem konnte man mit Can im 3er Aufbau seine Schwächen kaschieren. Maatsen funktioniert da schon echt gut.

000

Jakob Schwerter @jakobschwerter.bsky.social · 23/01/2024

A big shout out to my amazing co-authors Ketevan Gurtskaia, Andrés Romero, Birgit Zeyer-Gliozzo, and Markus Pauly for their invaluable contributions to this work! 🌟🤝

000

Jakob Schwerter @jakobschwerter.bsky.social · 23/01/2024

Our preprint provides a comprehensive view, guiding empirical researchers to choose the most effective method based on their specific data context. 📊👩‍🔬 #ResearchInsights arxiv.org/abs/2401.09602

100

Jakob Schwerter @jakobschwerter.bsky.social · 23/01/2024

Our findings? Tree-based methods, especially MICE RF and missRanger with PMM, often outperformed standard MICE PMM in most scenarios. This could be a game changer for dealing with complex datasets! 🚀 #MachineLearning

100

Jakob Schwerter @jakobschwerter.bsky.social · 23/01/2024

In our study, we look at the performance of tree-based imputation methods such as MICE with Random Forest (RF), #missRanger, and #mixGB and compare them to the traditional MICE PMM. 🌳📈 #StatisticalAnalysis

100

Jakob Schwerter @jakobschwerter.bsky.social · 23/01/2024

Working with multiple imputation and wondering how tree-based methods compare to MICE PMM? Researchers from FAIR (fair.tu-dortmund.de/en/) and the RCTrust (rc-trust.ai) have just published a preprint that sheds light on this! 🌟 #DataScience #Statistics arxiv.org/abs/2401.09602

arxiv.org

Evaluating tree-based imputation methods as an alternative to MICE...

Dealing with missing data is an important problem in statistical analysis that is often addressed with imputation procedures. The performance and validity of such methods are of great importance...

121

Jakob Schwerter @jakobschwerter.bsky.social · 13/01/2024

Can in der IV und Öczan auf der 6 anstatt Süle in der IV?!

000

Jakob Schwerter @jakobschwerter.bsky.social · 22/12/2023

I guess if you like a manuscript you submit your review faster than if you don’t like it?

010

Jakob Schwerter @jakobschwerter.bsky.social · 22/11/2023

Die schwachen Ergebnisse von IGLU kamen vor, oder? Zumindest wurden die medial diskutiert.

000

Reposted by Jakob Schwerter

Nick Huntington-Klein @nickchk.com · 16/11/2023

not at all meaning this as a dunk, it's truly interesting how different empirical problems and methods can shift between uninterestingly basic in one field to novel and barely-appreciated in another. i saw this so many many many times in researching my book

A december 2021 article in Methods in Psychology called "Omitted Variable Bias: A threat to Estimating Causal Relationships"

We aim to raise awareness of the omitted variable bias (i.e., one special form of endogeneity) and highlight its severity for causal claims. Firstly, we demonstrate via analytic proof that omitting a relevant variable from a model which explains the independent and dependent variable leads to biased estimates. Secondly, we offer an easy-to-understand visualization for the problem. Finally, we discuss two remedies, diminishing the risk of the omitted variable bias, namely the instrument variable or two-stage least squares estimator and the regression discontinuity design. We hope that our review will motivate researchers to use them more often in future research.

65714

Reposted by Jakob Schwerter

Brit Williams, Ph.D. @drbritwilliams.bsky.social · 11/11/2023

This one time on twitter years ago, @robertkelchen.com made me chuckle, BC he said he doesn’t deal with AERA much BC this distance between submission & presentation is so far the paper could just go under review. And I just had a paper accepted by a journal but not AERA and I can’t stop laughing.

271

Reposted by Jakob Schwerter

Viola Angelini 🟥 @violaangelini.bsky.social · 08/11/2023

📣📢 We are hiring an assistant professor (research career track) in #econometrics! Expertise in machine learning is especially welcome. Groningen is a great place to live and work! Sharing is caring. #EconSky 📈📉 inomics.com/job/career-t...

inomics.com

Career Track Assistant Professorship in Econometrics (1.0 FTE) | INOMICS

The Department of Economics, Econometrics and Finance of the University of Groningen is looking for a Career Track Assistant Professor in Econometrics.

097

Jakob Schwerter @jakobschwerter.bsky.social · 08/11/2023

Nicht anzeigen, weil die Person wenigstens gezeigt hat das sie weiß, dass das Verhalten falsch war?

100

Jakob Schwerter @jakobschwerter.bsky.social · 04/11/2023

Wenn nicht jetzt, wann dann? Gefühlt gab es in solchen Situationen aber deutliche Niederlagen😅

220

Jakob Schwerter @jakobschwerter.bsky.social · 04/11/2023

Unpopular opinion, I guess, but I really like the touch bar of the MacBook and will miss it. Why is it so disliked?

000