Plenary: Michael Jennions, A Drunkard’s Walk: from negative results to reverse P-hacking

Plenary presentation at the 2021 conference of the Society for Open, Reliable, and Transparent Ecology and Evolutionary Biology (SORTEE)
www.sortee.org/
Host: Shinichi Nakagawa
When: 13 July 2021 0100-0230 UTC
Download: osf.io/723ba/
Abstract: I have no formal training in statistics, meta-analysis, meta-science, open science or philosophy of science. None. My publications on these issues have always been tangential to my own research. But I have engaged with these topics over the years because of deep frustration with the publication process. It is an on-going challenge to be ethical while trying to get research published; and uneasy pacts have to be brokered between the devils and the angels to be a good PhD mentor in a field where numerous publications are an entry level requirement for an academic career. I can’t offer a deep analysis of the problems we face as ecologists and evolutionary biologists. Instead I will take the scenic route and describe my personal experiences over three decades. I will touch on publication bias, meta-analysis and P-hacking. I hope that this will, at least, give hints as to where real progress has been made, and to where reforms are still needed. It might even cause you to be more sympathetic to the failings of your older colleagues (i.e. my generation).
Brief bio: I grew up in South Africa. I studied for an BSc and MSc at the University of the Witwatersrand in Johannesburg where I studied the mating behaviour of frogs. Next I did a PhD at Oxford University (UK), working on sexual selection and fluctuating asymmetry (or sweet FA as a colleague once called it) in birds, crabs, damselflies, frogs and flowers. This was followed by a post-doc in Panama at the Smithsonian Tropical Research Institute working on cichlids, livebearer fish and crabs. I moved to the Australian National University in 2001, where I have mainly worked on crickets and mosquitofish. My broader interests in science have been indulged through participation in a working group on meta-analysis at NCEAS (Santa Barbara, USA) and a sabbatical year at the Wissenschaftskolleg zu Berlin (‘Wiko’) in Germany. I have co-authored many papers, and a decent number of book chapters. But I’d happily trade them all for writing a bestselling popular science book that allows people to see the world differently.