Apparently I’m late on the bandwagon for spotting this one, but twitter is atwitter with news that it’s cold in Washington D.C. right now. A propose of my last post, Chairman the Senate Environment and Social Works Committee and Amateur scientist, James Inhofe (R-OK), brought a snowball into the Senate yesterday as data against the notion that Climate Change is occurring.  “We hear the perpetual headline that 2014 has been the warmest year on record,” he said.

Apparently he meant to say that climate can change, but that humans have no ability to effect this kind of change directly. As he states on his webpage:

“The climate is changing, as it always has been changing, and always will, there’s archeological evidence of that, there’s biblical evidence of that, there’s historic evidence of that, it will always change,”Inhofe said.

But more interesting to me is his data gathering.

datapoint 1: Snowball in Washington D.C. February 27

datapoint 2: “It’s very, very cold out.”

conclusion: It is unseasonably warm.

Senator Inhofe should spend some time with Darrell Huff. Huff’s classic text, How to Lie with Statistics is one of my favorite books. It’s a bit out of date with its examples, but still a fun, easy read. And more importantly, a good look at how data can be manipulated to appear to represent a picture different from reality.

It all comes down to data collection and interpretation. As Senator Inhofe did, I could easily gather some quick data from my immediate surroundings, ‘ 100% of the people I had breakfast with live in my house’ and derive a quick conclusion: ‘everyone in the world lives in my house.’

Or, I could go more subtly and use data closer to Sen Inhofe. Right down the hall in the other house of congress sit enough data to clearly demonstrate that employment numbers are climbing steadily with no sign of this so called ‘great recession.’ In fact, no sitting member of Congress has ever been unemployed while serving in congress.

Actually, the important point to make is that science can be done well, but it can also be done poorly. When it’s done poorly, it can be due to a number of reasons, such as poorly designed experiments, lack of appropriate controls, inaccurate data collection or poor choice of samples. All of these can happen without any intent to deceive, however, any experiment done with a conclusion already in hand is inherently partisan and is bound to produce unrepresentative results. Science should not be politicized, doing so generates flawed data and erodes public confidence in both our lawmakers and our scientists.

I have no problem with science being presented in congress, but I do think that it needs to be done responsibly and in careful consultation with those who understand the entire scientific process.

Does your congress member have a scientific advisor?

I’ve just written to my Senators and Representative to ask that question directly and I hope to hear back from them soon. I’ll be sure to post their answer here.