The long-respected system of peer review is
increasingly under attack. According to many scientists and academics,
peer review is ineffective and time consuming at best and a tempting
venue for favoritism and plagiarism at worst. The main problems with
peer review stem from two basic issues: inconsistency in the reviewing
process itself and too many submissions to top-tier journals. According
to Drummond Rennie, executive editor of the
Journal of the American Medical Association, “If peer review was a drug, it would never be allowed on the market.”
Problems with the Reviewing Process
Peer review is a hazy concept. Every journal and grant-giving
foundation has a different definition of what constitutes peer review
but most require it before an article is even considered for
publication. Most publishers agree that peer review has something to do
with a third party reviewing a manuscript or grant proposal, but after
that the details are often fuzzy. Major points of variation in peer
review include how many peers review an article, who those peers are and
what the conditions of the review are.
Articles are usually reviewed by between two and four people who are
chosen by an editor to recommend whether an article is fit for
publication. These peers are generally—but not always—experts on the
article’s subject and are asked to read for consistency, mistakes,
misinformation and evidence of fraud. Some journals prefer for these
peers to be anonymous and some have a fully transparent review process
but either way, the review process usually does not accomplish its
goals.
Peers miss an amazing amount of mistakes, inconsistencies and
outright fraud. Generally unpaid, peers have little motivation to give
articles the thorough dissection they need. Peer reviewers are often
asked to review far more articles than their schedules allow, a
situation documented by Daniel Myers in a recent essay about his
experiences as a
peer reviewer.
As a result, peers tend to breeze through an article, decide whether it
sort of makes sense or not and pass it along to the editor with a
recommendation based more on what they had for breakfast than the
integrity and quality of the research.
The recently discredited work of Woo-Suk Hwang at Seoul National University is an excellent
example.
Woo-Suk Hwang’s work went through a rigorous peer review process that
entirely failed to catch the researcher’s massive amounts of fraud.
Peers caught up in today’s publish-or-perish academic environment are
often too overwhelmed with their own department’s writing demands to
devote the correct amount of time and attention to the articles of their
fellow scientists.
Submission Increases and Stylistic Changes
Authors working in academia are currently under intense pressure to
publish in top-tier journals. In the past, researchers were free to
submit articles to the most suitable journals instead of simply the most
famous ones. Less-than-groundbreaking work could find a home in second
and third-tier journals and authors had a low rate of rejected articles.
Today, editors at top journals are overwhelmed with submissions that
are unsuitable for their publications and authors are constantly
frustrated by the rejections of well researched and thoughtfully written
articles that simply aren’t appropriate for the most widely read
publications. Much as not every short story is ideal for Harpers and the
New York Times only has room for a certain number of editorials,
top-tier journals are similarly limited by space and focus.
The editor-in-chief of Science, Donald Kennedy, states that his publication
rejects
at least 6,000 papers every year and submissions are increasing
steadily. With so many rejections, editorial staff members are required
to use more of their already over-stuffed workdays addressing author
complaints and sorting through claims of impartiality and plagiarism.
Authors are going to increasingly desperate measures to get editors
to publish their work. In recent years, there has been a serious uptick
in the amount of time scientists spend networking with editors in
attempt to make a personal connection that will make rejection less
likely. Another tactic is to exaggerate the results of a study, creating
a flashier article at the expense of good science. Articles that have
any link to human diseases are always more popular and writers have been
known to stress very tenuous links between their research and human
health just to appeal to editors.
Possible Solutions
In theory, peer review is a wonderful idea. The system of allowing an
author’s cohort to review her work for flaws is sound and worked
reasonably well for a number of years. When universities began foisting
ridiculous publishing demands on their researchers, the system began to
break down. Academics are now generally required to publish at least
five articles per year, one of which is in a major journal. Considering
how many more academics than top-tier journals there are, this is
obviously unreasonable. Peer reviewers are often these same
academics—constantly struggling to find enough time for their own work,
let alone a colleagues article. To improve the peer review process,
universities first need to relax their researchers’ publishing quotas.
The peer review system has never been standardized and that would be
an excellent next step toward saving it. Creating a basic training
program for reviewers, a universal form for reviews and a system to deal
with accusations of favoritism and plagiarism in the review process
would be useful. Deciding whether reviews for individual publications
would be anonymous or transparent would also be a smart move as would
limiting the number of reviews a peer can do per year.
The peer review system is deeply flawed but salvageable. Authors and
publishers are losing confidence in the whole system for good reason,
and many authors are now pushing for its abandonment. Peer reviews is an
essential control for maintaining the integrity and quality of research
and those interested in saving it need to make changes sooner than
later to prevent the breakdown it’s headed toward.
0 comentarios:
Post a Comment