Science is a messy business (and I like it that way).
Stephen Curry, on reciprocal space, wrote a good post regarding the ex Dr X. He homed in on the “wanting it to be neat” notion as a possible reason for, well, making things up!
But, of course (as anybody who has seen a real data-set knows) science isn’t. Kind of like life (looking at dust in the corner. Deciding that, no, I don’t care, life is messy).
He then goes on to discuss his own messy results. I like that, even if I don’t understand a word of what he is saying, because, well, not my niche.
I really want an honest science. That display the variance, the heteroscedacity, the unclear results – or the road up to the results. I’ve never been good at spin, or telling the neat story (which make me prone to rambling and over qualifying and all that). Even when I was first presenting research, I would bring up the problems (and my buddies would comment that they thought it was kinda neat that I didn’t put the neat spin on it). Well, it wasn’t neat. It was exasperating.
Which is also why I want to expose my students to get all muddy with data, early and often.
Dorothy Bishop also posted a really good comment. Just the sheer laziness of not actually going out and make sure data was collected. (And, I wonder, why on earth did he not use his students? What are grad students for?).
I cited that to my two masters students, who are working hard on getting 100 people to respond to their rather lenghty project. It is in collaboration with another lab, so we will not be messing with their instruments.