Different studies of the same effect sometimes produce markedly different results, a phenomenon dubbed the replication crisis. Significance testing and sampling variation provide simple explanations for much of the apparent crisis, whereas compatibility and Bayesian interpretations of confidence intervals identify real replication failures. These failures are quantified as heterogeneity in random-effect meta-analysis, which can apportion at least part of the heterogeneity to the modifying effects of subject characteristics and study methodologies. Meta-analysis can also identify and discount heterogeneity arising from publication bias and sometimes from scientific fraud. Any remaining unexplained heterogeneity does not constitute a replication crisis.
Please register for this Masterclass at ECSS Glasgow 2024 via the congress app as soon as it is available. Further information will be announced in good time before the congress.
Date: Tuesday, 2 July
Time: 09:00 – 10:30
Session room: Carron 2
Prof. Will Hopkins
Internet Society for Sport Science
Auckland, New Zealand
This email address is being protected from spambots. You need JavaScript enabled to view it.