2010-08-19

Experiments Always Succeed

Posted in research at 10:53 am by cori

Their findings concurred with the conclusion that the experiment had failed: The monkeys didn’t appear to react to the change in patterns.

via chronicle.com

This article exhibits two specific problems, one with the journalistic establishment and one with research overall.

The author of this piece is misrepresenting science and research, at least as a researcher who follows the scientific method would understand it, and this researcher is not only being dishonest in reporting his data, but if he believes as the journalist certainly seems to indicate that he does he isn’t following the dictates of science and research.

Here’s the crux of the matter: an experiment *only* fails if you approach it with the intent to prove a specific hypothesis. In the truest spirit of the scientific method this should never be the case – you undertake an experiment in order to form a hypothesis. If the experiment doesn’t support your conclusion then either your conclusion is incorrect, you have misinterpreted the data, or the experiment you designed isn’t testing what you thought it was. The experiment *always* does what it’s designed to do, and any failure is with the experimenter.

Why this is wrong for the researcher should, I think, be obvious – if you’re going into an experiment hoping for a particular outcome then you’re already off the rails before you even start.

Why this is wrong for the journalist is perhaps not so clear. Articles like this one set the tone for our shared expectations and assumptions about the role of research in our society. We make value decisions about research based on these assumptions, and perhaps more importantly we make funding decisions based on them as well. If we, as a culture, expect experiments to conform to predetermined expectations then we are losing out on the true value of research and experimentation, and we ourselves are creating corrupt researchers like the one the article supposes Dr. Hauser to be.

Also, to Tom Bartlett, as much as I may disagree with whatever Dr. Hauser is alleged to have done, it is not Mr. Hauser, it is Dr. Hauser.

(sent to http://cori.posterous.com via email)