Mathematical biologist, who escaped the Ivory Tower after getting a PhD and now works in industry. Enjoys mathematical modelling, data science, functional programming, and lots of other obscure things.
Around five million scientific articles are published every year. Because they are written by experts for other experts in the field, it is generally hard to judge their credibility as a layperson. However, there are still some general rules and heuristics that can be used to evaluate them. In this talk, I will provide a gentle introduction to the topic. As a recent PhD graduate in the field of biomathematical cancer research, my focus will be on biomedical topics; however, the general approach equally applies to all (natural) sciences.
Statistics is a core part of modern-day scientific research, helping us to draw reliable conclusions from raw data. However, due to the ever increasing competitiveness in science, there's a strong pressure to publish more and more impressive results every day. Join us on our journey through the lands of questionable research practices and learn how to torture your raw data, until it finally confesses your personal version of truth. (Alternatively, visit this talk if you want to learn some tricks for spotting misuse of statistics in published research.)