Abstract
The Office of Inspector General at the U.S. National Science Foundation is studying the feasibility of statistical surveys to measure the frequency with which misconduct in science occurs. Such information would be useful in shaping federal government policy. Existing data are known to be inadequate for this purpose. A basic choice must be made between a survey in which respondents would report on their own misconduct and one in which they would report on misconduct they have observed. The methodological difficulties involved in both kinds of survey are discussed Suggestions are sought from the scientific community.
Get full access to this article
View all access options for this article.
