Studies accumulate over time and meta-analyses are mainly retrospective. These two characteristics introduce dependencies between the analysis time, at which a series of studies is up for meta-analysis, and results within the series. Dependencies introduce bias —Accumulation Bias— and invalidate the sampling distribution assumed for p-value tests, thus inflating type-I errors. But dependencies are also inevitable, since for science to accumulate efficiently, new research needs to be informed by past results.

In this KEBB webinar, I would like to discuss why it is important to think about accumulation in meta-analysis. And explain how a new approach to statistics that we call 'safe' and 'any-time valid', could help us handle the problem of accumulation bias, thanks to its connection to the likelihood paradigm and game-theory.

Additional Files
30339.pdf Presentation , 1mb