was an isolated event or part of a pattern, and (4) whether it is relevant only to certain funding requests or awards or to all requests and awards of the accused.34 The burden of proof is on the agency proposing the sanctions, and the agency must prove its case by a preponderance of the evidence. 35
The NSF groups its possible sanctions into three classes, ranging from the least restrictive (such as a letter of reprimand) to the most severe (including termination of a grant and recommendation for debarment).36 Individuals subject to less severe restrictions are entitled to fewer procedural safeguards, whereas procedures for imposing debarment are strictly defined.
The PHS categories for sanctions for misconduct differ slightly from those adopted by NSF. OSIR has indicated taking a variety of actions in response to findings of misconduct in science in addition to the actions implemented by the research institutions (see Table 5.1). The OSIR's actions have included referral to the DHHS's OIG (when there have been findings of possibly criminal offenses), use of PHS sanctions (such as repayment of funds or debarment), and other institutional penalties (such as “letters of admonishment to subjects or institutions, a requirement that the employing institution send letters of reprimand to the subjects, and a requirement that the subjects of an investigation send letters of apology to the informant ” (DHHS, 1991b, p. 6).
Some misconduct investigations have revealed problems that fall short of the regulatory definitions of misconduct in science but are judged to warrant remedial actions. These problems include “scientific sloppiness, incompetence, poor laboratory management, and poor authorship practices ” (DHHS, 1991b, p. 4). Failure to implement the remedial action can result in a loss of future funding or other institutional penalties. Local institutions may also take remedial actions (such as withdrawing a research proposal), even if an inquiry results in a finding of no misconduct and no further investigation is conducted.
A particular problem arises when a government agency undertakes a review of an investigation that has been completed by a university. The university investigation is often undertaken by members of the research community who are requested by university officials