A01 A02 A03 A04 A05 A06 A07 A09 A10 A11 A12 A13 A14 A18 A19 A21 F01 F02 INF Ö

ReproducibiliTEA: “Three-Sided Testing: An Interval Hypothesis Approach to Analyzing Your Data”

Null hypothesis significance tests (NHST) only allow the null hypothesis to be rejected without ever addressing practical significance. Three-sided tests (TST) can solve this problem—they make it possible to formally test whether results are practically significant, practically zero, or inconclusive. No statistical magic, no hype. Just a better, more transparent framework.

Join us in discussing how TST works, where it outperforms classical methods, and why specifying a smallest effect size of interest is so important.

The following paper provides the basis for the discussion:

Isager, P. & Fitzgerald, J. (2024). Three-Sided Testing to Establish Practical Significance: A Tutorial. Preprint. https://doi.org/10.31234/osf.io/8y925_v2

It starts on Tuesday, 15.07.2025 as usual at 14:00, either in IB/6-127 at the RUB or online via zoom.

The entire program is available here: >>

And further information on the Journal Club project page at the Center for Open Science: >>