The Results in Table 1 Don't Seem to Correspond to Those in Figure 2

About the Talk

Presenter

Assoc. Prof. Mine Çetinkaya-Rundel

Abstract

For a data analysis to be reproducible, the data and code should be assembled in a way such that results (tables and figures) can be re-created.

This talk addresses the challenges of reproducing scientific results: data that is not available, missing dependencies, outdated software. It presents good, better and best workflows for reproducibility, touching on everything from data storage and cleaning, through analysis, to the communication of final results.

Date: August 20th, 2019 – 17:00 (Istanbul)