Carsten Bergenholtz

@Bergenholtz
5 Followers
13 Following
9 Posts
I study complex problem solving. Currently Associate Professor at Aarhus University.
Show threadCarsten BergenholtzDec 26, 2022
@danluu However, that my teaching routines and aims will be fundamentally disrupted seems obviously true. I am very excited about a future of teaching with ChatGPT (and similar AI driven tools).
Show threadCarsten BergenholtzDec 26, 2022
@danluu Interesting to see the same range of opinions from various experts. Some key contingencies, I think, are 1) topic: I teach the somewhat more peripheral topic of phil of science at a business school (bachelor level), and on my 25 last exam questions it passed about 6-7, and received a C on 1 of them (none better). 2) Nature of the questions also matters. I expect deep reading of curriculum, citations and linking across texts. For now it is a tool to assess if a possible question is good.
Carsten BergenholtzNov 25, 2022
I teach a large bachelor level Phil of Science course at a business school. A substantial part of the course addresses how science is difficult (e.g. covering topics such as the replication crisis, human biases, why social phenomena are often difficult to measure/construct, causality etc.) I also want to rebuild trust in science, e.g. presenting topis such as science as a collective process, falsification etc. As an accessible take on this I use https://fivethirtyeight.com/features/science-isnt-broken/), Any (other) suggestions?
Science Isn’t Broken

If you follow the headlines, your confidence in science may have taken a hit lately. Peer review? More like self-review. An investigation in November uncovered a scam in which researchers were rubber-stamping their own work, circumventing peer review at five high-profile publishers.

FiveThirtyEight
Show threadCarsten BergenholtzNov 22, 2022
@randomwalker If I don't see all "tweets", it is more difficult to track the quality of their assessments - also who is willing to acknowledge mistakes, e.g. So, I've created my own "algorithm" of how to get diverse input, allowing me to rely less (!) on any external algorithm.
Show threadCarsten BergenholtzNov 22, 2022
@randomwalker I think it matters what kind of content one follows. If research, the order of "tweets" is probably less important, since speed (measured in hours) is not essential. However, if one is tracking people identifying- discussing current events, it is really useful to see the chronological order -> helps me identify who is able to not only be accurate + perceptive, but also early. In the long run, this helps me create a list of trusted followers, from whom I want to see all "tweets".1/2
Show threadCarsten BergenholtzNov 21, 2022
@Mhicmhuirich @hildabast Thank you, that fixed it!
Show threadCarsten BergenholtzNov 18, 2022
@hildabast Is it only in my feed that the pictures in this thread are flagged as "sensitive content", and I have to click to see them?
Show threadCarsten BergenholtzNov 3, 2022
@lakens I look forward to what a real test would look like, and in particular what "error control" implies here! Not sure the field of management is ready to accept such connotations, but tbf we are the younger sibling that is usually a bit behind psych, econ, sociology etc. In any case, I don't think the paper provides sufficient advice on how to avoid including nonsense assumptions.
Show threadCarsten BergenholtzNov 1, 2022
@lakens @dsquintana In case you haven't seen this paper, here is how some folks in the field of management approach the multiverse analysis, with a focus on the particular challenges that the field of management poses https://people.bu.edu/tsimcoe/documents/published/testimony.pdf