This shouldn’t really affect any of you, but I’m moving most of the images on my site to Flickr. I wasn’t very smart in managing the size of my images and already racked up over 1.5gb of bandwidth on my server. I figure moving the images offsite will help a lot.

Over at Mental Floss they have a listing of the 10 Trailblazing Scientists About to Change Your Future. I certainly agree with their assessment that these scientists are important to our future, but where are the psychologists?

Just as an example, Dan Gilbert over at Harvard, is helping us better understand what it is that makes us happy. In his book, Stumbling on Happiness, he tries to show that what we think will make us happy is often far from what actually does make us happy. And while these other scientists may be revolutionizing their respective fields, psychologists like Dan are revolutionizing the way we think about ourselves.

Some credit for his and others’ work would be nice.

I learned a valuable lesson the other day about ordering things online: read the fine print.

I’m an avid web-shopper, so I feel particularly stupid for this one, but I suppose it can get the best of us. I needed a new calendar for my office and figured I would just get one off of Amazon. Five minutes of searching later I found a nice one called “Beaches of the Caribbean” (see above). It was $6.99 + $3.00 shipping so I went for it. A couple of days later I get a surprisingly small box from Amazon. I open it up and what do I find? A calendar designed for an elf.

At that point I didn’t feel like dealing with returning it so now I’m stuck with a tiny calendar for a year. Not a big deal, but had I read the fine print (see below) I likely would have bought something else.

Lesson learned.

Not to harp on the Super Bowl ads to much, but USA Today has an excellent system set up showing the quality of all the Super Bowl ads. They even have online/continuous data for each ad by age, gender, and income. That’s pretty impressive.

Have a look.

For as long as there has been science, there have been ‘bad’ scientists. Now I don’t mean ‘bad’ in the sense that their work was meaningless (though there have been, and certainly are, plenty of those), but rather that these scientists committed the worst scientific crime possible: they fabricated their data.

The essence of science is slow, steady, and incremental discovery and disconfirmation. By fabricating data, scientist steer others in wrong directions and break the chain of discovery. And yet, as anyone who has the faintest understanding of human psychology knows, there are vast pressures on all of us to succeed. This is particularly true in the scientific community. Graduate students vie for jobs that are becoming harder and harder to fine. Junior faculty vie for promotions and tenure. And everyone is out for the next grant. The pressure is immense.

One doesn’t have to look too far for examples of how this pressure gets to even the best and the brightest. “In 1999, Victor Ninov (above) of Lawrence Berkeley (Calif.) National Laboratory and his colleagues claimed to have made the first atoms of elements 118 and 116.” When no one could replicate these findings suspicions arose and questions were asked. Turns out it the results were fake.

Jon Sudbø (above) of the University of Oslo’s Norwegian Radium Hospital fabricated data related to oral cancer research. “Sudbø has acknowledged that he invented some data, and a five-person investigative panel led by Anders Ekbom of the Karolinska Institute in Stockholm last week issued a report saying the bulk of his work was invalid.” (1)

The famous “hockey stick” chart (above) depicting the drastic rise in climate temperatures made famous in 2001 by Michael Mann (above) in his climate report, has been sharply criticized. Some claim his data were fabricated (http://www.climateaudit.org/).

More recently, and perhaps more notoriously, Hwang Woo-Suk (above), the Korean stem cell scientist, admitted in court that he fabricated data.

And finally, even some of the greatest scientists of all time succumbed to this great error in judgment. The famous astronomer Johannes Kepler (above) fabricated data to advance his heliocentric theory.

The point is that despite the ivory tower that science sits in, it is not immune to the flaws of human nature. And this worries me.

While this is a visible problem in the hard sciences, I am particularly worried when it comes to behavioral research (my domain). During pre-historic times, when behavioral scientists collected the majority of their data using paper-and-pencil questionnaires, there were audit trails. Now, when many experiments are conducted on computers, this audit trail is gone. In my department, at least half of all research is done via computer programs, with the only person having access to the data being the researcher conducting the experiment. In fact, I collect a great deal of data over the web. I program the experiments, post them, and collect and analyze the data. If I felt the urge to, changing participants’ responses would be trivial. Thankfully, I don’t. But others might.

I’m not suggesting that we stop collecting data with the aid of computers (far from it), but I am trying to make it apparent just how easy it would be to falsify data given that there is no paper trail and only a single set of eyes on the data.

My worries stems from two places. My first concern is that science is already being criticized for being elitist. By adding the label of fraud, these ‘bad’ scientists are not helping the cause. My second concern is a bit more selfish. I worry that I will waste time attempting to expand someone else’s ideas and failing to do so because the information I have is fraudulent. Imagine coming up with an extension to someone else’s work, only to realize, several failed experiments and several months later, that those original ideas were fraudulent. Yah, it’s a problem.

If nothing else, it should be abundantly clear how important it is to scrutinize scientific findings. While the vast majority of findings are likely legitimate, those few that aren’t should be found and, to be dramatic, destroyed. In science, and I suspect elsewhere to, there is nothing worse than false data.

(1) Couzin, J. (2006), “Fake Data, but Could the Idea Still Be Right?” Science, 313 (5784), 154.