MetaCausal is a site created to evaluate, criticize, discuss, and occasionally intervene on issues of quality of scientific evidence in both the academic and public settings. We are a loosely organized team of statisticians, economists, epidemiologists, and journalists working on everything related to statistics, science, and people, particularly when it has to do with health. The site is obsessed with the idea that health-related data science should be candid, transparent, scientifically rigorous, translatable, open, and useful, with a strong emphasis on robust causal inference.
MetaCausal was started in 2018 by Noah Haber alongside the Causal language and strength of inference in academic and media articles shared in social media (CLAIMS) systematic review study. The CLAIMS team systematically evaluated the strength and implied strength of causal inference in both the academic journal sources and media reporting of the most widely distributed health news articles of 2015, in partnership with NewsWhip™. We looked at some of the most-shared health articles of 2015 and found that the causal strength in the original research was often pretty low, but the researchers overstated it in their academic articles, and then media exaggerated it even further.
If you wonder why you keep seeing that chocolate makes you live forever while also giving you Alzheimers, why people make poor or illogical choices, how people misunderstand statistics, and why people distrust science and statistics so much, so do we.
This site is not formally associated with any academic, industry, or media organization.
This site is run by Noah Haber. Noah completed his doctorate in Health Economics in the Department of Global Health and Population department at the Harvard T.H. Chan School of Public Health in 2017. His work spans several areas, including the meta-scientific work discussed on this site and work on HIV/AIDS in South Africa.