Washington. -- A big splash of attention always awaits reports that scientists cheat.
The professionals in lab coats are, of course, members of the same species as savings-and-loan hustlers, tax dodgers and other miscreants whose sins must rise to the spectacular to draw attention. But scientists present themselves as paragons of purity, and therefore even their minor delinquencies elicit oohs and ahs from a naively trustful public.
The latest example comes from a survey of 4,000 faculty members and Ph.D. candidates, reported in the current issue of the American Scientist, and widely echoed in news accounts. Located in 99 departments of chemistry, civil engineering, microbiology and sociology, the teachers and students reported witnessing many practices that are ethically out of bounds in the world of research.
For example, 22 percent of the faculty members reported sloppiness in use of research data by colleagues, and 15 percent said they personally knew of instances in which scientists ignored lab findings that inconveniently contradicted their previous research.
The faculty reported a high incidence of plagiarism among the graduate students, and a lesser, though not insignificant, amount among faculty colleagues.
About 20 percent of the faculty claimed personal knowledge of colleagues ignoring regulations governing research on humans and animals.
Students and faculty reported encountering many violations of university research policies, including misuse of research funds and failure to make required disclosure of financial involvements with private firms. Half the students and a quarter of the faculty members said they would expect retaliation if they reported scientific sinning by a faculty colleague.
The authors of the report said their findings ''raise grave concerns about the willingness and ability of members of academic research communities to govern the conduct of their peers and their students.''
The chieftains of science, however, insist that their community is self-policing and clean, and they tend to regard the public's fascination with scientific fraud as uninformed and misguided. The classic assertion of complacency was made several years ago by Daniel E. Koshland Jr., the editor of Science magazine, the central journal of the science establishment, who pontifically declared that ''99.9999 percent'' of scientific reports ''are accurate and truthful.''
That assertion is mocked by the latest and several previous surveys of scientific ethics, plus a small but steady flow of admissions of misconduct prompted by investigations at universities and other sites of research.
What the scientific establishment fails to realize is that its ho-hum confidence about its own purity risks the loss of a golden status in public perceptions. The popular fascination with scientific fraud arises from faith and hope in science, both of which remain strong, despite continuing revelations about serious misdeeds and corner-cutting.
Public trust, however, is not infinite. And though its level cannot be measured with any precision, it probably has declined in recent years.
Science's capacity for minding its ethics sank even lower recently with the virtual collapse of one of the principal government agencies in this field, the Office of Research Integrity in the Department of Health and Human Services.
The office, which has jurisdiction over the multi-billion-dollar National Institutes of Health, is the successor to previous arrangements for investigating and ruling on allegations of scientific misconduct, and therefore deserves something of a grace period. But so far, it has lost two cases in which its findings of misconduct were appealed.
Fearful of losing again, it has dropped two others, including the long-running case against Robert Gallo, the renowned AIDS researcher, whom it deemed guilty of misrepresenting his accomplishments and impeding the research of other AIDS scientists.
The public gasps about scientific delinquencies because science, in the public mind, is still an edifice of ethical purity. To sustain that faith, the scientific community must demonstrate a deep distaste for misconduct and a willingness to crack down on it. At present, neither is evident.
Daniel S. Greenberg is a syndicated columnist specializing in the politics of science and health.