Statistics - friend or foe?

I took psychology in university and have had a love/hate relationship with statistic ever since.  In one way I am fascinated by the idea of creating meaningful patterns and developing theories simply by looking at numbers.  On the other hand I have sat through a plethora of graduate student meetings in which they discussed their research and then endlessly debated the best statistical theory to apply in order to get meaningful results.  I never studied stats at a graduate level so perhaps I am speaking purely out of ignorance, but I never quite understood how you could analyze data one way, not have anything significant show up, analyze it a different way and BAM!  There you have it, a valid correlation.  It all seemed a bit blurry around the edges to me.  (As a side note, I must mention that I have endless criticism about how the majority of psychological experiments are conducted so maybe I am just too biased to be writing about this.)

The reason I started thinking about this however, is because I just finished reading Super Freakonomics by Steven D. Levitt and Stephen J. Dubner, both of whom are economists and both of whom consider stats to be their bread and butter.  There has been a proliferation of books like this one in the past ten years or so, which mostly examine human behaviour and make it clear that we, as humans, are pretty irrational in our decision-making and thought processes.  It is all well and good and interesting, but most of these books are based on results, which come from psychology experiments, which are based on stats.  In fact, some reviews that I have read of Super Freakonomics argue that the authors' own conclusions, or the conclusions of the studies they mention, are faulty in that they twist the numbers and the wording to fit a certain model.

So where does this leave us? Especially those of us who may not know or understand the intricacies of data collection and analysis.  I would argue that a highly skeptical and critical-thinking crowd is better than one that takes things at face value, but do we need to be critical to the extent that we cannot read something published by two highly educated economists without furrowing our brows and questioning each argument and piece of evidence they present?  Seems like a pretty infuriatingly complex way to read something for enjoyment.  I don't know where the answer lies but I can say this much - it's hard not to feel like a schmuck when you read something that you find interesting and well-researched, only to discover that there is more than a dozen people out there who can poke holes in every single premise you just accepted.  I suppose that may be the real reason behind why I could never succeed in academia; sometimes reading something that makes the world make a little more sense is better left alone in my opinion.

"There are three types of lies -- lies, damn lies, and statistics."

— Benjamin Disraeli