Psychology · Research Methods & the Replication Crisis
A topic within Research Methods & the Replication Crisis, itself one of 10 topics in that field and part of Psychology.
2 works
Geoff Cumming
Argues psychology should replace single-study null-hypothesis significance testing with estimation, confidence intervals and cumulative meta-analysis across studies, because a single p-value cannot say how big an effect is or how much to trust it.
link checked 17 Sept 2026Notes on this copyGene V. Glass
Coins the term 'meta-analysis' for the statistical combination of results across many separate studies, arguing this integrative analysis is a distinct and necessary third stage of research beyond the original data collection and its secondary re-analysis.