Publish or perish

Once her paper is written, polished and submitted to a journal, it enters what is, to her, a black hole. When the paper arrives at a journal, an editor, who is either a full-time professional editor, as is the case for most top biology journals, or an editor who is also a researcher, usually at a lower tier journal, will assess the novelty of the work and the quality of the data presented in it and decide whether or not it is going to be sent out to peer review. The great majority of papers are rejected by top journals without being sent for review by scientists in the community. Up to 80% of the papers never make it past the editor. The leading journals typically reject close to 90% of all submitted papers.

Professional editors are former scientists with PhDs who have left the research lab to take on a career in publishing. These editors actually play an important part in science, even though they will, most of the time, never actually do another experiment for the rest of their careers. The editor must be able to assess science from a wide range of fields and determine whether the work presented in the paper is a significant advance or not. This is easier said than done, and an editor relies heavily upon the peer review process to make the more difficult decisions. Peer review is the process whereby scientists not involved with the work presented in the paper anonymously evaluate the quality and significance of the work. A research paper that is sent out to review is typically seen by three referees, who provide both confidential comments to the editors and comments for the authors of the paper. It takes several hours to write a full review of a paper. The goal of the review is for referees to provide advice on whether the scientist has presented enough data to support the conclusions provided at the end of the paper and to let the editors know whether they think the work is novel enough to publish in a prestigious journal or whether it was an incremental advance;simple enough. But here's the kicker: the reviewers do it for free. In addition to running their lab, writing grant applications and research papers, teaching, counseling their lab members, attending faculty meetings, and traveling to science conferences, researchers give their time to the peer review process. It is widely understood that reviewing research papers and grants is an integral and important part of the scientific process.

Like everything in science, not all is as it would appear. The peer review system can be corrupted by researchers holding a grudge against an author, who are friends with an author, or who are competing with the author. This is why there are multiple reviewers, but in the end the voice of the pickiest reviewer is often heard over the protests of the authors. It's not a perfect system, but no one has really come up with a better one that promotes competition and protects the authors.

Research papers are rarely accepted for publication after a single review. More often than not, the scientists are asked to do further experiments and significantly revise their text before publication. After the reviews are returned to the editor, he or she will confer with other editorial colleagues and decide whether the paper is going to be rejected or accepted, or whether the researcher will be given the opportunity to revise the manuscript and resubmit it. But what is to keep reviewers from making claims or asking for unreasonable amounts of work?

Analogy time: let's say that you are a scientist reviewing a paper and realize that the results are very similar to results that your lab has produced, but which you have not yet published. Let's say that the authors of the paper have figured out a detail that you have not. Do you have your student or post-doc do the key experiments with the knowledge you have gained so you can avoid being scooped? Do you ask for experiments that would delay publication until you can catch up? Do you ask for them to do an experiment that would actually require them to collaborate with you? Do you make such harsh comments that the paper might get rejected? After all, delaying their publication could seriously help you. The reality is that scientists in this position could excuse themselves when they realize that there is a conflict of interest, but many do not. Remember: these are anonymous reviews, so you can say what you like as long as it appears legit. The idea of having three reviewers is to have a balanced view of the paper, so no single reviewer can be unreasonable. The only thing keeping reviewers honest is their own integrity (and sometimes a diligent editor), but it really is a lot to ask of someone who is by nature highly competitive. There is little that anyone could do in any system to completely prevent unscrupulous behavior.

For many scientists, receiving a rejection letter is an enraging experience. This is particularly true when the reason given is that the work does not represent a significant conceptual advance or that it is not novel enough. Talk about a big ol' slice of humble pie. An editor or reviewer doesn't think your work is interesting. After working your ass off for several years, your work is deemed insignificant. The reaction to this can range from rage to depression, but is rarely quiet acceptance. Often a scientist will appeal the decision. However, a paper is rarely published after it has been rejected, and the appeal letter often only serves as a means to vent frustration or anger rather than actually achieving anything.

The pressure to publish can be so great that it leaves a young scientist to do the unspeakable: commit fraud. Like speaking Voldemort's name, scientists cringe at the mere mention of someone forging results. It is widely considered the worst thing a scientist can do. A survey came out in 2003 that indicated that the majority of scientists knew of a colleague who had done something that they would consider unscrupulous. Yet the incidence of reporting scientific fraud is very low. The reason? It is extremely difficult to prove that someone has committed fraud unless they do something very stupid or confess when confronted. However, it is so horrible to imply that a scientist has committed fraud that it is exceedingly rare to have someone be confronted and then confess. Not only that, there is virtually no oversight by the NIH to try to detect it. There are of course cases where people do bad Photoshop jobs to try to forge a result, but mostly the data is just manufactured wholesale and this type of fraud is next to impossible to detect. Since science works by piling results on top of one another, building a foundation from which to get a larger picture of (say) biology, researchers who forge data are destined to be discovered - unless, for some reason, they happen to be correct in the story they made up. But by the nature of science research, data that cannot be reproduced could just mean that the experiment was done wrongly or differently, which makes it even harder to detect fraud.

So, why do some graduate students and post-doctoral fellows forge data? The answer is unfortunately predictable: the pressure to succeed is too high. For the same reason that some premedical students cheat in undergraduate classes, some scientists will resort to making up their own success to meet the demands of an advisor who is putting pressure on them to get results that fit into their model or a committee that is putting pressure on them to get results in order to graduate, or to avoid the fate of having their funding run out with no papers to show for their efforts. The sense of helplessness that all scientists eventually face is almost never addressed. Forging data is seen as the domain of the unscrupulous and weak.

Let's say that our graduate student has to go back and do some more experiments. She resubmits her paper and it is accepted for publication: the ideal situation. She gets the green light to write her thesis. At the same time, she is deciding if and where to do a post-doctoral fellowship. Once you have your PhD, you are considered inexperienced and unqualified to run your own lab. A postdoctoral position is held up by most scientists as your only reasonable option. It is only recently that students have been exposed to alternative careers after graduate school. The fact is that only around 14% of all PhDs in the biological sciences will ever run their own academic labs. This statistic is never revealed to people applying for graduate school or incoming graduate students;not because faculty are unaware of the fact that most don't get faculty posi tions, but that it would seem to be part of the frame of mind required to be a scientist. This means that most of her colleagues will not make it to the coveted faculty position. Instead, they will remain post-docs indefinitely, become teachers, or take a job in the biotech industry.

For a long time (until as recently as the early 1990s) it was considered selling out if a scientist took a job in the biotech industry. The change corresponded directly with the glut of post-docs who just couldn't find faculty jobs and the growth of the biotech industry. However, in the eyes of university professors it meant that you had failed - you weren't good enough to make it. This seems ludicrous today. Now there is sometimes as much competition for a good industry position as there are for faculty positions. Increasingly, there is a large number of scientists going into 'alternative careers,' becoming editors, intellectual property attorneys, policy makers, and biotech business consultants. Indeed, it is now common to have people giving seminars on other career options in the same institutions where professors consider their graduate students to be failures if they don't become a post-doc. This reflects a more realistic trend, and in the end it will benefit all. If there is not a significant increase in the available faculty positions, then fewer and fewer people will go into science as the dirty little secret about job availability becomes more widely known amongst undergraduates. The dedication that it takes to become faculty at a research institute is so strong that their students and post-docs not getting what they aim for really doesn't occur to many professors.

Our student is one of these people who does not consider the possibility of failure. Graduation and the thesis defense go smoothly, and after revising the text she can officially be called 'doctor'. Now she cleans up some loose ends and is off to become a post-doc. Becoming a scientist is a more difficult process than most would have imagined. So what! Wait until you hear what a post-doc goes through. Most correctly estimate that getting a PhD is hard - you are earning a doctorate after all. But grad school is child's play compared to doing a post-doc. This is where despair and depression are layered like a birthday cake topped with a generous icing of pressure and doubt. Yummy!

Post-docs are usually required to get funding to support their projects. If they don't have any first author papers from graduate school, then they are pretty well screwed. Also, if the new lab they are working in has not yet had publications in the area they are proposing to do work in, they will have a hard time. Many lab heads will require that their post-doc comes in with independent funding before they start in the lab. This means that students start writing grant applications as soon as they are done with their PhD, or even while they are still writing their papers and thesis required for graduation.

It is widely assumed from early in the education of scientists that they don't have to be proficient, clear writers to succeed. The truth could not be any different. While there certainly are a lot of scientists who are as awkward with the English language as they are with the opposite sex, one of the most important tools that scientists can have is the ability to communicate what they have achieved and how they achieved it. The language of science around the world is English, and foreign researchers are well aware of it. What is remarkable is the fact that few graduate programs ever teach their students how to write a grant application or research paper.

Let's assume that our budding scientist has been able to navigate this process and get a grant. (We have thus far assumed that she is really good, so why stop now?) The fact is that the NIH awards post-doc grants to approximately 30% of applicants. Other agencies are a little better or worse, but 30% is normal, necessitating application for several grants to secure funding. Our post-doc will now be getting about $28,000 a year to do science. This is the standard salary paid by an NIH grant to a first-year post-doc. On aver age, she will spend about five years as a post-doc and typically switch labs and institutions once. That means that she will have spent approximately 11 years becoming a scientist before she is applying for jobs at universities to run her own lab.

Was this article helpful?

0 0

Post a comment