Statistical analysis does not come from statistical software

This talk was presented to the residents and fellows of the Department of Ophthalmology (11 June 2009) at Yonsei University in Seoul, Korea. The main point of this presentation is that using statistical software correctly is not sufficient to produce a valid statistical analysis. The best way to mitigate the dangers of an invalid analysis is to understand the data set sufficiently to recognize when an analysis does not describe the data. This is demonstrated with an example that derives from my dissertation. The presentation is intended for an audience that does research, but does not have a statistical background. It includes a list of free software packages (all but EpiInfo are FOSS) that can be used to investigate data sets.

[Slides for "Statistical analysis does not come from statistical software" talk (204 KB)]

This document is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.