[Originally posted 5/20/2007]
If integrity is defined as adhering to a moral or ethical code, what does it mean when applied to science? Is it any different than the moral or ethical code we apply in our daily lives? Yes and no. It certainly does mean the usual definitions of honesty and all the virtues. But in science it also means more than that.
There is no rule book for scientific methodology that a graduate student needs to memorize to become a scientist. Nevertheless, one of the most critical lessons a grad student must learn is how to do science properly. Each discipline and each problem demands its own approach, making it impossible to write a single "user's manual" of how to do science. Following rigorous methods is critical to having one's results accepted by a critical, scrutinizing community of peers. The methodology therefore is developed as a means of assuring that all possible subjective biases have been removed or minimized.
In high-energy particle physics, the stakes are so high, with so many people and so much money involved, that it is not uncommon to apply secret offsets to the data before analysis. This prevents the scientists who are doing the analysis from subconsciously skewing the analysis to achieve the desired result. After the analysis, the offsets are removed and the correct answer can be obtained.
Not all disciplines require such a technique but everyone must show rigor in methodology that ensures any critic that the results are credible. The methods used in the research must be reported accurately to such an extent that not only is it clear that subjective elements have been offset but that anyone skilled in the art could reproduce the work.
Careful work is valued in all vocations. In science it is doubly important that integrity is demonstrated at every step. Any lapse in integrity leads to a loss of credibility from the rest of the community, usually devastating in a career.