Workshop in Biostatistics

M112 Alway Building, Medical Center

(next to the Dean's courtyard)

DATE: April 20, 2017
TIME: 1:30 - 2:50 pm
TITLE: Measuring scientific bias across disciplines and countries
SPEAKER: Daniele Fanelli
Senior Research Scientist
Meta-Research Innovation Center at Stanford (METRICS)

 

Abstract:

Science is said to be suffering from a reproducibility crisis caused by growing research and publication biases. How common are these problems in different fields and countries, and have they actually worsened over time? In this workshop I will present past and recent studies that tried to answer this question. I will describe, in particular, two different methods that I have yielded informative results: one based on a very simple and flexible proxy of scientific bias, and one based on second-order analysis of meta-analytical data. I will illustrate the advantages and limitations of each method and I will discuss what the findings obtained using each method can tell us about the integrity of contemporary science.

Suggested readings:

Daniele Fanelli, Rodrigo Costas, and John PA Ioannidis.  Meta-assessment of bias in science.  Proceedings of the National Academy of Sciences, USA.  (2017)  Vol. 114 no. 14, p3714-3719.

Daniele Fanelli.  Positive results increase down the hierarchy of the sciences.  PLOS one, published April 7, 2010.