Metaresearch and Reproducibility

Fraser, H., et al., (2021).

Related to the MetaMelb lab, we study the science of how science is done. This is called metaresearch or metascience, and is important in understanding credibility, replicability and reproducibility of scientific claims.

Researchers:

Fiona Fidler, David Wilkinson, Elliot Gould, Hannah Fraser, Bonnie Wintle, Peter Vesk, Mick McCarthy, Rebecca Groenewegen, Libby Rumpff, Anca Hanea

Affiliate researchers:

Victoria Hemming, Mark Burgman

Selected publications:

Wintle, B., Mody, F., Smith, E., Hanea, A., Wilkinson, D.P., Hemming, V., Bush, M., Fraser, H., Thorn, F.S., McBride, M. and Gould, E., 2021. Predicting and reasoning about replicability using structured groups.

Fraser, H., Bush, M., Wintle, B., Mody, F., Smith, E., Hanea, A., Gould, E., Hemming, V., Hamilton, D., Rumpff, L. & Wilkinson, D.P., 2021. Predicting reliability through structured expert elicitation with repliCATS (Collaborative Assessments for Trustworthy Science).

Hemming, V., Burgman, M.A., Hanea, A.M., McBride, M.F. and Wintle, B.C., 2018. A practical guide to structured expert elicitation using the IDEA protocol. Methods in Ecology and Evolution, 9(1), pp.169-180.

Alys Young (she/her)
Alys Young (she/her)
PhD Candidate / Research Assistant