The project from early 2005 that got me back into the saddle, statistically speaking, involved a large amount of methodological research, data collection, analysis and report writing. It's the single largest client project I've undertaken and resulted in a 20,000-word report.
The nice part of it was the range of analysis techniques I was able to employ, from simple descriptive and summary statistics (means, standard deviations, counts etc) to cross-tabulations and on to multi-variate analysis using k-means clustering, principal components and MANOVA tests.
To be honest I think the client on the receiving end of the report was a little overwhelmed by that part of the document, which is why I had buried it in an appendix.
Anyway, it blew the cobwebs off my statistical calculator, textbooks and probability tables!!
2 days ago