NYTimes | Cheating
in scientific and academic papers is a longstanding problem, but it is
hard to read recent headlines and not conclude that it has gotten worse.
Falsified or erroneous results have forced authors and editors to
retract papers from journals that let themselves be duped into
publishing them. Researchers at some of the nation’s most prestigious
universities have been implicated. There are ways to minimize this kind
of fraud, but it will require changing the process, from how scientists
share their data to how their peers review it and who is allowed to
enforce academic standards.
In the latest well-publicized case, the journal Science on Thursday retracted
an article, published in December, which had purported to show that gay
political canvassers could change conservative voters’ views on
same-sex marriage in brief face-to-face conversations. The researchers
could not produce the original data necessary to resolve questions about the work.
Cheating
is thought to contaminate only a small portion of all the research in
this country, but no one knows for sure. Cases that have emerged from
the shadows are tracked by Retraction Watch, an independent blog that covers research. In an op-ed article in
The Times on May 23, the blog’s co-founders noted that, in each of the
past few years, the Office of Research Integrity, part of the Department
of Health and Human Services, has sanctioned a dozen or so scientists
for misconduct ranging from plagiarism to fabrication of results.
Clearly,
this has not been a good year. The journal Environmental Science &
Technology corrected a March paper on fracking because the lead
scientist failed to disclose funding from an energy company. In May, The
Journal of Clinical Investigation retracted a paper on cancer genetics
from a young researcher at the National Cancer Institute because the
data was fabricated.
How
could this happen? Often a young researcher, driven by the academic
imperative to “publish or perish,” fudges the data. In many cases, a
senior scientist who is supposed to be monitoring the research pays
little attention, content to be listed as one of the authors.
In
theory, a journal’s peer reviewers are supposed to detect errors, but
they often do not have the critical data needed to check the findings,
nor the time to do so, particularly since they are seldom paid.
Sometimes the cases only come to light when a whistle-blower, perhaps a
student or researcher in the lab where the cheating occurs, points the
finger. The scientific community clearly needs to build a better safety
net.
