What is “Forensic Meta-Science” and Why Should I Care?

Presenter: Fabio Molo | UZH, Center for Reproducible Science and Research Synthesis

Open science practices — including the sharing of data, code, and methods — are promoted as means to reduce questionable research practices like p-hacking and "hypothesizing after the results are known". If the whole research process is transparent, questionable practices and fraud will easily be detected by peer reviewers and readers -- or so is the hope.

This talk will present examples where "data sleuths" have indeed detected inconsistencies in prominently published research and open research data, thereby limiting the further spread of false results. But we cannot leave the detection of errors in published research to a few idealistic researchers working nightshifts. The talk suggests that more readers of scientific publications complement their critical mindset with a few simple techniques to detect red flags for questionable, unreliable research — that is, to routinely conduct "forensic meta-science".
The talk will introduce the INSPECT-SR forensic checks for systematic health reviews along with the GRIM test and other simple techniques that enable readers to check reported statistics, such as mean and standard deviation, for numeric consistency. The talk will also introduce the broader meta-research and research synthesis activities and services of the Center for Reproducible Science and Research Synthesis.

Researchers at UZH and other institutions