Growing up on a healthy diet of Microsoft Office products, I am well versed in Word, Excel and Powerpoint.  As I have transitioned into the research world, these products still have their place, however, I sometimes find that the habits I developed for organizing data doesn’t necessarily transfer to statistical analysis.  Recently, I ran into a situation where I was evaluating the performance of solutions in multiple different environments.  Organizing this data appeared straightforward to me at first, I would simply group the different environments into one row grouped by the id of the individual.  My data then looked something like this:

Continue reading →