How to Avoid Lies and Damned Lies: Pitfalls of Data Analysis

There seems to be a pervasive notion that "you can prove anything with statistics." This is only true if you use them improperly. In this paper we'll discuss things that people often overlook in their data analysis, and ways people sometimes "bend the rules" of statistics to support their viewpoint. We will also discuss ways you can make sure your own statistics are clear and accurate. I will include examples from medicine, education, and industry.