#harking

2025-01-03

Ersetzt man nach Erhalt frischer #Daten die ursprüngliche Ausgangs-#Hypothese durch eine "passendere", betreibt man sogenanntes #HARKing. Damit führt man die Kollegen an der Nase herum – und nicht selten auch sich selbst … — In unserer "Freitags-Kolumne": laborjournal.de/editorials/315

Zeichnung eines jungen Mannes, der erklärend einen Arm ausstreckt. Er wirft einen langen Schatten hinter sich, dessen Nase "Pinocchio-artig" verlängert ist. Rechts daneben Überschrift "Wirklich nur kleine Lügen?" des im Post verlinkten Artikels, gefolgt von dessen Vorspann.
2024-06-13

Looking for material on #harking in #qualitative #politicalScience research. I don't know of any papers that directly discuss the issue. My uninformed hunch is that much qualitative research poses as hypothesis-testing (because that's fashionable and gets you published), even when the research started out much more inductive. Is this is as much of a problem as it would be in quant research? #research_methods

Bjørn Sætreviksatrevik@fediscience.org
2024-01-17

Table 1 of the "False-positive psychology" paper (Simmons, Nelson & Simonsohn 2011, papers.ssrn.com/sol3/papers.cf) estimate the false-positive rate of some questionable #ResearchPractices, both alone and in combination (see attached figure).

I remember reading somewhere that the authors later stated that the numbers were inaccurate and should have been somewhat higher. Does anyone have a reference for this? #FalsePositivePsychology #QRP #phacking #HARKing #ReproducibiliTea

A visual representation of Table 1 from the "False-positive psychology" paper (Simmons, Nelson & Simonsohn 2011). 

Researcher degrees of freedom	p < .05
Situation A: two dependent variables (r = .50) 	10 %
Situation B: option to add 10 more observations per cell	8 %
Situation C: controlling for gender or interaction of gender	13 %
Situation D: dropping (or not) one of three conditions	13 %
	
Combine Situations A and B	14 %
Combine Situations A, B, and C	31 %
Combine Situations A, B, C, and D	61 %
2023-11-28

What's your favorite resource explaining #p_hacking and #HARKing ?

Needed for the @digiresacademy advent calendar 🎄

Research Network Digi-Oek.chDigiOekCH@social.tchncs.de
2023-06-24

[en] Cheating in Science: Harvard "Honesty Scholar" May Have Been Caught in Dishonesty

"... dishonesty can lead to creativity" - an interesting and somewhat amusing read.

The New York Times: "Questions about a widely cited paper are the latest to be raised about methods used in #behavioral research."

datacolada.org/111

#ResearchHighlights #honesty #dishonesty #phacking #harking #dredging #gino #harvard #fraud #cheating #academic #datacolada

I just had a long discussion with a colleague who is about to start a complex and burdensome #experiment whose design imho is flawed. I'm well aware my reasoning might be wrong. What troubles me is the handwaving way specific experimental design questions, e.g.:
- What are you going to measure as result?
- X and Y frequency
- How will you interpret if Y is say 20%?
- We could theorise, but we'll do the experiment instead, we'll see when we get the results...
#harking #reproducibility #science

2022-12-24

@tedpavlic I'm sure it's not the intention - indeed they add a cautionary note - but this kind of advice leads to HARKing & other questionable research practices that massage research into a neat narrative.

It's the antithesis of #OpenScience practices like #RegisteredReports. You should start with a research question, a summary of what is known, & then the methods. Post-hoc theorising must be declared. It keeps you honest.

#WritingTips #ResearchWriting #HARKing #QRPs #PublicationEthics

Client Info

Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst