Data mining is indeed a very mediocre scientific activity. Correlation on itself means nothing at all. If you want to proof something the correlation should be 100% and you should be able to explain why the correlation exists and replicate it in controlled experiments. The problem is that those slam dunk scientific discoveries are all or mostly allready found. And nowadays the poor scientists need to find something to bolster their path to glory.
Good science could be: find a correlation an proof the causality. But a lot of studies stop at the correlation. That's what fills newspapers nowadays. 'You get fat from diet coke since most people that drink diet coke are fat'.
Some scientist try to eliminate all other reasons and then decide that their causality is the only one that explains the correlation. But in effect they say: those things correlate and I 'the superintelligent scientist with multiple PhD's' cannot find another explanation and that is why my explanation must be true.
For background you should listen regularly to 'more or less: behind the stats' http://www.bbc.co.uk/programmes/b006qshd
You can listen to podcasts that are interesting and fun to listen to. And some of the older ones are absolutely great. They gave me great insight in the workings of media (and science).
The only downside is that if your girlfriend tells you something she heard on the radio and you answer her: correlation is not causality, she gets upset.