2024-02-12 09:15 UTC
Replies (6)
-
@vtraag@social.cwts.nl 2024-02-12 09:16
Without causal thinking, we are bound to interpret observations incorrectly. One case study in particular illustrates this well. Here, we know (because we simulate) that open data improves reproducibility in research. However, when looking at research published in journals, we observe that articles with open data are less likely to be reproducible. This in contrast to the known positive effect. What happened?
-
@tonyRH@mstdn.social 2024-02-12 09:26
@vtraag @tklebel Congratulations to you both - such a great paper!
-
@LudoWaltman@social.cwts.nl 2024-02-12 09:43
@vtraag @tklebel Great to see this paper is now online. I had the pleasure to provide some comments on an earlier version. I believe this is a highly important paper, essential reading for anyone struggling with challenges around causality in science studies research. Unfortunately too often we end up confusing correlation and causality. This paper will help you to do better!
-
@OpenAccessElder@hcommons.social 2024-02-12 14:05
@vtraag @tklebel Congratulations on the publication and thank you for the breakdown here! The figures are excellent as well. I'd love to see more research breaking down connections like this in other open avenues, like with open access and citations. I'm sure there would be similar outcomes re: novelty
-
@eschares@scholar.social 2024-02-12 15:39
@vtraag @tklebel Congratulations! I will give this a close read. It fits exteremely well with the Statistical Rethinking course I am currently taking with @rlmcelreath
-
@dorothearrr@fediscience.org 2024-02-12 16:48
@vtraag @tklebel I found your sessions on causal inference at the CWTS Summer School very insightful. Thank you for this accessible introduction to the topic!