Thu, 20 September 2018
Meta-analyses—structured analyses of many studies on the same topic—were once seen as objective and definitive projects that helped sort out conflicts amongst smaller studies. These days, thousands of meta-analyses are published every year—many either redundant or contrary to earlier metaworks. Host Sarah Crespi talks to freelance science journalist Jop de Vrieze about ongoing meta-analysis wars in which opposing research teams churn out conflicting metastudies around important public health questions such as links between violent video games and school shootings and the effects of antidepressants. They also talk about what clues to look for when trying to evaluate the quality of a meta-analysis.
Sarah also talked with three other contributors to our “Research on Research” special issue. Pierre Azoulay of the Massachusetts Institute of Technology (MIT) in Cambridge, Ben Jones of Northwestern University in Evanston, Illinois, and MIT’s Heidi Williams discuss the evidence for some hoary old scientific home truths. See if you can guess who originally made these claims and how right or wrong they were:
Read the rest of the package on science under scrutiny here.
This week’s episode was edited by Podigy.
[Image: Davide Bonazzi/@SalzmanArt; Show music: Jeffrey Cook; additional music: Nguyen Khoi Nguyen]