Testing an interaction term is not the same as testing the simple effects

Popularity Index: 3 Updated at 2026-10-11 14:37 UTC

Sites

Related Topics

Non-inferiority margin: choosing it from historical data instead of convenience Splitting the sample so the hypothesis is tested on data the model never saw A one-sided test is defensible only if the direction was fixed before the data arrived Fisher exact vs chi-square when a cell count drops below five How much pilot data is needed before predicting an effect size is honest Removing an outlier after seeing the result turns a test into a decision Sequential testing: why peeking at the data changes the error rate Permutation tests as the escape route when the normality assumption fails Should the hypothesis predict a direction or only that some difference exists Cook's distance flags the single observation driving the regression line

Hot Categories

Featured Categories