Show newer

We looked at > 100 papers on flaws of specific predictive optimization applications. Seven recurring clusters emerged. Given the technical architecture of predictive optimization (vs other forms of automated decision making) we show that this is exactly what we should expect. That made us ask: does *every* predictive optimization application have these flaws? We picked 8 case studies based on prevalence and impact on people. Among these, the answer seems to be yes. predictive-optimization.cs.pri

Show thread

Great news to start the week! The collaboration of #ODISSEI and #CLARIAH and 15 national partners has been awarded a #NWO Large-scale Research Infrastructure Grant of €15.2 million to build digital infrastructure for the social sciences and humanities. Read more on odissei-data.nl/en/2023/02/nwo

WAHNSINN 🙌- Hattrick für die Grundrechte!! Wir haben auch das #EqualPay-Verfahren vor dem BAG gewonnen! ⚖️Gleiche Bezahlung ist nicht verhandelbar! Bye bye “Er hat halt besser verhandelt”!! Ein Riesendurchbruch für die #Entgeltgleichheit💪Glückwunsch an die Klägerin!

Understanding Society @usociety launches 12th annual Insights report, showcasing research on health, work, and the cost of living. 2023 themes:
- economic insecurity and the cost of living
- long-term inequalities and health
- improving work and health.
understandingsociety.ac.uk/202

It's International Love Data Week!

JSSAM loves data that state if a study has been preregistered. When submitting to JSSAM, authors using empirical data should include a statement in the acknowledgements about whether the study or analysis plan was preregistered.

#LoveData23 #LoveDataWeek

📢 Job alert! Come work with me @ on the Social Cohesion Panel.

Coordinate a panel survey, conduct your own research, decide flexibly on when & where you work (within Germany). We strongly value ! Apply by Feb 28!

@surveystatisticsandmethodology
@sociology
@dvpw
@politicalscience
@economics

diw.de/sixcms/detail.php?id=di

📢 Das Wählerverzeichnis zur @ratswd Wahl schließt in ⏰zwei Wochen. Also schnell noch eintragen bis zum 27. Februar! www.ratswd.de/wahl

Wer ist durch den Wegfall der bislang kostenlosen Scientific API von Twitter betroffen?
Ich suche Beispiele für Bericht im DLF, wo dies Einfluss auf die Forschung hat.
#FediCampus #ScienceMastodon

New article alert! Chayut Wongkamthong & Olanrewaju Akande on "A Comparative Study of Imputation Methods for Multivariate Ordinal Data"

"The results highlight several important conclusions about the performances of MI methods for ordinal data as well as several implementation decisions analysts will need to make in applying the methods ... [and] illustrate some potential issues with uncertainty quantification when using some of the MI methods."
#Research #Statistics
doi.org/10.1093/jssam/smab028

New article alert! Mehdi Dagdoug, Camelia Goga & David Haziza on "Imputation Procedures in Surveys Using Nonparametric and Machine Learning Methods: an Empirical Comparison"

"We conduct an extensive empirical investigation that compares a number of imputation procedures in terms of bias and efficiency in a wide variety of settings ... The results suggest that a number of machine learning procedures perform very well in terms of bias and efficiency."
#Research #Statistics
doi.org/10.1093/jssam/smab004

New paper out w/ @j_adriaans & @csauer11 (on 🐦) in Social Psychology Quarterly!

Using German employee data, we clarify that pay justice and pay satisfaction are closely related yet distinct evaluations

doi.org/10.1177/01902725231151
1/4

Vom 13.-17. 2.2023 findet die Love Data Week statt mit spannenden Veranstaltungen zum Mitmachen und Mitlernen rund um #Forschungsdaten, #Datenmanagement, #DataLiteracy & #OpenScience. An dem internationalen Programm beteiligt sich auch die @UniKoeln: fdm.uni-koeln.de/netzwerkuzk/l
#LoveData23 #LoveDataWeek #C3RDM @UniCologne_D7

New article alert! James Wagner, Brady West, Mick Couper, Shiyu Zhang, Rebecca Gatward, Raphael Nishimura & Htay-Wah Saw on "An Experimental Evaluation of Two Approaches for Improving Response to HH Screening Efforts in Nat'l Mail/Web Surveys"

"We found that the bilingual approach did not increase the [RR] of high-likelihood Spanish-speaking addresses ... [and that] a priority mailing with a $5 incentive ... [increased] the screening survey [RR]"
#Research #Statistics
doi.org/10.1093/jssam/smac024

New article alert!

"Sequential and Concurrent Internet-Telephone Mixed-Mode Designs in Sexual Health Behavior Research" by Stéphane Legleye and Géraldine Charrance

"Sequential telephone-Internet mixed-mode protocols nested in a probability telephone survey may be a good way to improve survey research on sensitive behaviors."

#Research #Statistics #Survey

doi.org/10.1093/jssam/smab026

"Challenges of Virtual RDS for Recruitment of Sexual Minority Women for a Behavioral Health Study" by Middleton, Drabble, Krug, Karriker-Jaffe, Mericle, Hughes, Iachan, & Trocki

"The RDS involved using virtual ... rather than in-person processes for orienting, screening, and scheduling potential participants for a telephone survey, and inviting “seeds” from a probability sample to recruit participants from their social networks."

#statistics #ResearchMethods

doi.org/10.1093/jssam/smab039

Professional news: I have just signed my new contract and am now Head of Innovation Sample & Social Cohesion Panel at ! I'm thrilled to be leading both those amazing panel survey studies from now on & will be hiring a postdoc for my team very soon!

Und wir freuen uns erst: Vorläufige Akkreditierung @ratswd – wichtiger Schritt in der Entwicklung unserer #Dateninfrastruktur zur Analyse des gesellschaftlichen Zusammenhalts #OpenScience #FAIRData (#Findable #Accessible #Interoperable #ReUsable)

---
RT @ratswd
Wir freuen uns, die vorläufige Akkreditierung eines neuen FDZ verkünden zu können! Das FDZ-FGZ @fgz_risc stellt Daten zu Fragen des gesellschaftlichen Zusammenhalts zur Verfügung. …
twitter.com/ratswd/status/1615

Show older
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.