In my classes this week we are discussing ethics in research. The textbook version focuses on not lying to people who will be in your study, not doing anything to cause physical and emotional harm, and so on. It's funny how textbooks sometimes miss the real issues. Maybe it is because some events, such as the research study where a group of black men were not informed they had syphilis so that the university could study the effects, are so horrible that we want to make sure they are never repeated. I can certainly sympathize with that view point.
However, I think most ethical issues are much more mundane. Most of the past two days were spent chasing down anomalies in data. Something was wrong with the test results on a final report I was preparing for a client. For four of the five years, we found a significant improvement in the students after they had been in the program. For the fifth year, there was no difference. Since all of the results are added together, no one would know if I just said, "The hell with it" and reported the data as is. I take that back, _I_ would know.
What's the harm? Why spend two days trying to make sure the data are exactly correct when no one might know the difference? That, I think, is the true test of one's character, doing the right thing when no one is watching. My point is not that I am the most ethical person in any room because, the fact is, there were a lot of times over the past two days when I felt like writing down the results that showed up on the computer screen. There were a lot of times I heard that little voice whispering, "
Who will know? Come on, you could move on to another project and make the company more money."
Our evaluation contracts are bid at a fixed price, not by the hour, so if I spend two more days on a project, that comes out of our profit. Why did I bother? Because even if no one else would know, I understand that if there are certain discrepancies in the data it may mean that it was entered wrong, that some people are in the wrong group, that the tests were scored wrong, and the whole outcome may be very different than it appears.
When people read scientific research, it is based on trust. They trust that the researcher did the right thing as well as he or she knew. That is one reason funding agencies are very biased in favor of awarding grants and contracts to people with Ph.D.'s, they figure you know more of the pitfalls in research and how to avoid them.
As Erich says in the tagline of our new ethics course, "Unethical Behavior is Bad for Business." In some circles, business is considered 'dirty' by nature, it is assumed that people in business just care about the money and nothing else matters. Good businesses don't run that way, in my experience.
If we did cut corners, skip steps on data validation, report results that were not fully explored, we wouldn't be the only ones that did it. However, that sort of thing tends to come out eventually. As a consulting company, our reputation is our most valuable asset, and anything that damages that is bad for business. Maintaining a good reputation and, consequently, a steady flow of contracts, is not the reason we try behave ethically, though. It is just a fringe benefit.
Our contracts are an expression of trust on the part of clients. Even though they don't always understand all the technical points of programming and statistical analysis they trust us so much to do the research right that they even pay us money to do it. What kind of people would we be if we let them down?