Hey JJ,

Great article; it outlined the problems with surveys and sampling specific populations. The issues with the projections likely stemmed from exclusion; the type of person who takes a survey is very different from those who don't. Therefore they cant be accurately represented. It would have been interesting for you to expand on the statistical miscalculations that could have led to such skewed data. Maybe it's taking a sample from each state or those who don't use Facebook, I'm not sure what would've been best, but it would be nice to hear your opinion.