They would never try to make the data fit some preconceived idea, would they?
Except in the movies (where a large percentage are "mad"), scientists are generally thought to be among the most principled and unbiased people in society. I am not sure where this idea comes from, but realistically, one would expect the population of scientists to reflect the morals of the general (educated) population. This page is not an exhaustive look at bias among scientists, but, rather, an example of how some of these "unbiased" people can fall into a trap of supporting some dogma or paradigm despite data to the contrary.
Oops! inadvertently corrupted data
The story began when I had gone on vacation for a week. Not being there, an experiment was performed by another researcher, who used one of my Microsoft Excel templates to analyze the data. The templates allow one to take new data, put it into the worksheet and, with little effort and time, produce a completely labeled graph of the new data. It turned out that some of the data was out of order, so the other researcher copied the appropriate values and pasted them directly into the cells that produced the graph. Normally, all the cells that are used to generate the graph are equations that point to other values. However, the person was not very proficient at entering equations, so putting the values in directly was "quick and dirty." When I returned from vacation, I used the template to analyze my experiments. However, I did not notice that some of the equations had been changed to absolute numeric values. A couple dozen experiments were analyzed with the corrupted worksheet, resulting in an entire series of results that were invalid.
The paradigm takes precedence, not the data
Based upon this incorrect data, an abstract was written to an upcoming scientific meeting. The error was discovered a couple months before the meeting was to take place, but well after the abstract could be corrected. Upon proper analysis, the data showed absolutely no effect in the treatment group. Therefore, a large part of the abstract was incorrect. I recommended that the abstract be withdrawn. The only other alternative that I saw was to present the data that contradicted statements made in the abstract. However, the abstract was chosen as one of the best at the meeting and was scheduled to be in a symposium, instead of just being at the poster session. The lure of prestige was too great and the acknowledgement of error was too embarrassing to withdraw the abstract. Instead, the other investigators wanted to categorize the data into subgroups and see if there were any differences. The largest subgroup showed that about half the samples exhibited the reported effect, while the other half showed the opposite effect or no effect at all. One of the smaller groups showed no effect in any of the samples, while in the third group, one out of three showed the reported effect. The other investigators wanted to show examples of the reported effect from the samples that showed it, instead of reporting the averages (which showed no difference at all). I protested, saying that this was dishonest and deceptive. They insisted that this be done, so I took my name off the title slide - not wanting to be associated with the obviously biased data.
Pressure to compromise and "go along"
During the next few weeks of slide preparation, pressure was put on me to put my name back on the presentation, with compromises made to the slides to make them less deceptive. I heard from those putting the pressure on me that everybody within in the group disagreed with my position and wouldn't have made such a big deal about the slides. However, I had already determined not to compromise my principles to go along, no matter how large the opposition. I was to learn much later that most of the people in the group had supported me, though not publicly. Ultimately, more experiments were performed, and it was finally acknowledged that the data showed no difference. However, this occurred well after the deceptive data had been presented at the meeting.
Implications for the evolutionary paradigm
The purpose of this page is not to bash scientists or indicate that they are all dishonest. However, as humans, we scientists are sinners as much as anyone else, and don't want to make waves or look stupid in front of our peers. I believe that, to a large degree, acceptance of evolution by scientists is more of a function of going with the dogma rather than examining the data for what it actually shows. The numerous examples of evolutionary data that don't fit the evolutionary paradigm are simply ignored, or evolution is credited with more power for change than it could possibly possess. Few scientists want to rock the boat or take a stand that could be thought of as siding with the "stupid" creationists. In other words, it is not "cool" or popular to question evolutionary theories, since one might be classified as one of those "mad" scientists.
Last Modified June 21, 2008