Sir

Your News story “Misconduct? It's all academic...” (Nature 445, 240–241; 2007) and related articles in the same issue were too eager to find fault with the process universities use to deal with allegations of misconduct in research. As provost (S. M.) and vice president for research (C. R.) of Purdue University, we consider that you failed to give readers a true picture of the realities of these investigations.

A glaring omission is the fact that the US Department of Health and Human Services' integrity guidelines require US institutions to protect “the confidentiality of respondents, complainants, and research subjects” when investigating allegations of misconduct. This confidentiality is inconvenient for journalists. It also is extremely challenging for scientists and administrators who face the unhappy task of judging the integrity of one of their colleagues, but it is necessary if we are to prevent the ruin of good reputations through malicious or erroneous claims.

The process is not perfect. Like the US system of justice, it is frustrating, confusing and tedious, but it is the process we have, and we must follow it carefully until we find a better one. Despite its imperfections, in the end, it usually does the right thing. The process works best when those alleging misconduct document their concerns thoroughly and cooperate fully with all aspects of the inquiry — including the requirement for confidentiality.

An inquiry into a research misconduct allegation is not an inquiry into the verifiability of a research claim. Verifiability is decided by experimentation and debate, and often takes time to resolve. There can be legitimate differences of opinion regarding a laboratory observation. The Purdue administration's job is not to decide among such opinions.

Purdue University's policy on research integrity states: “The mere suspicion or allegation of wrongdoing, even if totally unjustified, is potentially damaging to a person's career. Consequently, no information about charges of a lack of integrity in research may be disclosed except to the appropriate university and federal authorities.” Any response to an allegation of misconduct at Purdue will adhere to the letter and the spirit of that principle. We believe this is true at the vast majority of universities. Readers of Nature would not understand that truth from your coverage, nor would they be likely to conclude that a successful and fair inquiry might include a finding of 'not guilty'.