Will sociology ever accept 'big data's move from causality to correlation? Now that we have enough data to make really good correlational models, will causality still be studied as intensely?
What are the main barriers to using data visualisation before deciding on a statistical model? To me, this seems the perfect way to work out what you're looking at, and indeed for. (I have in fact just done this at length for a Sociology class)
Not really a question, but I'd really like some discussion of 3D data representation where the three dimensions are x,y, and TIME instead of z. (I personally think that every layer of interaction you can cohesively add is almost as valuable as another axis).