When Does HARKing Hurt? Identifying When Different Types of Undisclosed Post Hoc Hypothesizing Harm Scientific Progress
By Mark Rubin. 2017
129 citations recorded by the source corpus.
Hypothesizing after the results are known, or HARKing, occurs when researchers check their research results and then add or remove hypotheses on the basis of those results without acknowledging this process in their research report (Kerr, 1998). In the present article, I discuss 3 forms of HARKing: (a) using current results to construct post hoc hypotheses that are then reported as if they were a priori hypotheses; (b) retrieving hypotheses from a post hoc literature search and reporting them as a priori hypotheses; and (c) failing to report a priori hypotheses that are unsupported by the current results. These 3 types of HARKing are often characterized as being bad for science and a potential cause of the current replication crisis. In the present article, I use insights from the philosophy of science to present a more nuanced view. Specifically, I identify the conditions under which each of these 3 types of HARKing is most and least likely to be bad for science. I conclude with a brief discussion about the ethics of each type of HARKing.
View Paper: https://www.semanticscholar.org/paper/0463f34209b24ac7495f04f8c3cc3389ad843b3d
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Heikki Huttunen
- Esin Guldogan
- Oguzhan Gencoglu
- Chamin Morikawa
- Mark J. van Gils
- Mehmet Süzen
- Mathias Gruber
- Jussi Leinonen
Citations in the corpus
- Beyond intrinsic and extrinsic motivation: A meta-analysis on self-determination theory’s multidimensional conceptualization of work motivation (2021, unknown): “…may be tempted to include or exclude hypotheses and analyses based on whether or not they are supported, which may lead to the under-reporting of empirical evidence that does not align with presumed theory and stronger support for a theory than is warranted based on empirical results (Rubin, 2017).”
- HARK Side of Deep Learning - From Grad Student Descent to Automated Machine Learning (2019, background): “This applies mostly to the effect of failing to report a priori hypotheses that are unsupported by the current results [5].”