What is HARKing?
HARKing (Hypothesizing After the Results are Known) is the practice of presenting post-hoc hypotheses as if they were formulated before data collection. It is a form of research misconduct that inflates false positive rates and undermines the credibility of published findings.
How it works
A researcher who tests 20 hypotheses, finds that one is statistically significant by chance, and then writes the paper as if that hypothesis was the only one tested is HARKing. The practice is widespread because academic incentives reward novel, significant findings. HARKing transforms exploratory findings (which need replication) into seemingly confirmatory results (which appear established), misleading readers about the strength of evidence. Pre-registration of hypotheses before data collection is the primary countermeasure.
Applied example
A psychology researcher analyzes 15 outcome measures, finds that one shows a statistically significant effect, and writes the paper reporting only that measure with a narrative explaining why it was the predicted outcome. The 14 non-significant outcomes are never mentioned.
Why it matters
HARKing is one of the questionable research practices that contributed to the replication crisis, producing a scientific literature contaminated with false positives dressed as theoretical predictions.
