I still daydream often, while walking, while sitting outside in the Ann Arbor sun, while swimming laps and many other activities. While I sometimes think about my present situation, my mind usually drifts to the future: what I hope to accomplish, where I will live in twenty years, what my hobbies will be and many other ideas. I have wanted to be a researcher since fifth grade, and I even at that age, I looked at the Nobel Prize and publishing in Science, Nature, the New England Journal of Medicine and the Journal of Neuroscience as the peak of a science career. I know, I know, it’s shallow, but I can’t help myself. I dream about the moment, receiving a message saying that my paper has been accepted, or that I have won the Nobel Prize. Though these are huge goals, I know that the chance of a Nobel Prize is low. In spite of this, it could be even lower with the direction the science community is taking today. The state of science is one that the Nobel committee should be concerned about because the system isn’t as perfect as it might seem.
Problems have been developing within the world science community, especially in the United States, for the past fifty years. These issues have stemmed from environments at universities and research institutions where the cutthroat settings incentivizes flashy results to improve career opportunities. Additionally, the science publication system has turned into a system remarkably similar to that of Wall Street, where high ranking journals have scientists by the throat, often consolidating the wealth of their organizations and using the science for financial gain, not to benefit the human condition. It is within these confines that science today operates. Science should be about its content and aid to the public, not its monetary impetus.
Many people praise science because it is one of the most successful processes that humans have ever discovered to make sense of the world. A major aspect of this success comes from its ability to remove confirmation bias from the process. Confirmation bias is a tendency to search for or interpret information in a way that confirms one’s preconceptions. In the last fifty years, the confirmation bias has shifted immensely, mostly turning into something called publication bias. This is demonstrated as the withholding of negative results from review boards and publishers. Though all of this can be countered through plenty of scrutiny through peer review and self-correction, these methods require many rounds of replication studies. Due to the nature of these institutions, this scrutiny is not attractive and occurring on a regular basis. For this reason, we are already beginning to see the problems of our system.
Just to demonstrate the problem, here are a few examples. Several meta-analyses in the past few years has shown that the value of antidepressants has been significantly overestimated. In addition, when drug companies Bayer and Amgen reviewed their own studies on cancer, they found that only 25 and 11 percent, respectively, of their own claims could be replicated. Scarily enough, many of the reviewers at Bayer and Amgen were the original researchers themselves. The biggest problem has been in the field of psychology. In 2015, Brian Nosek, a social psychologist and head of the Center for Open Science in Charlottesville, Virginia conducted a review study in which he reassessed, along with 269 other researchers, the findings of 98 widely accepted psychology studies. It was found that 39 of the 100 replication attempts were successful. Even more frightening is that this number could be larger. It has been estimated that possibly 80 percent of all major psychology studies are unable to be replicated. At the moment, there is virtually no funding for replication of scientific research, which can have dire consequences for the trust in the scientific process.
Replication in science, especially for the past fifty years, has been decreasing significantly. Why? Well, publishers and research institutions often pressure scientists to find new and exciting data, which doesn’t often include meta-analyses or replication of previous studies. Not only that, but this has increased the large amount of publication bias, in which scientists remove or analyze their data to lead to exciting new outcomes. To understand publication bias, readers should understand that scientific statistical analysis exists largely on the concept of significance.
A set significance level, depending on the field, is set before the study begins. In psychology for example, it is usually at 5 percent, so that means there is a one in twenty chance that the result of the study is due to chance. Sadly, a 2011 survey of scientists found that 70 percent of those surveyed had stopped their studies halfway through once it reached the 5 percent error, where it is deemed “acceptable.” Though highly unethical, it should not be surprising that this result occurs because of the cutthroat culture for surprising and new information in science.
A major reason why this occurs are journals themselves. Though I might dream about being published in an elite journal, much of the business has become highly lucrative and monetary based. As Randy Schekman, a Nobel Prize-winning cell biologist from University of California, Berkeley, explained, “These luxury journals are supposed to be the epitome of quality, publishing only the best research … (Nevertheless,) these journals aggressively curate their brands, in ways more conducive to selling subscriptions than to stimulating the most important research.” For this reason, flashy but inaccurate papers have been published, with the number of retractions significantly increasing over the last two decades. This is only perpetuated by the fact that institutions incentivize publishing in these journals. The professional reward for being published is alluring, and funding and appointment panels often use the name of the publication as a measure for quality of the individual’s science. Appearing in Science, Nature, Cell, JAMA or the New England Journal of Medicine often leads to large grants and professorships.
To make matters worse, journals often are funded through donations and federal dollars; yet, these donors are often never are able to access the research he or she is funding, along with large swaths of the public. In addition, extra fees are put on the scientists to submit their work, more than is necessary for it to be published. This does not include the massive amount of fees required by the public to even access the papers online or in print. Moreover, the highly controlled access to the content further monetizes science.
If we want people to trust science and reduce the incentives for publication, a new approach must be taken. One bright area has been in open science, a new push by the science community for decreased pressure on scientists for career and financial incentives, as well as the monetary gain of publishers. The Center for Open Science works to “change in the culture and incentives that drive researchers’ behavior, the infrastructure that supports their research, and the business models that dominate scholarly communication.” In this area, their mission is to respect uncertainty, open greater access for scientists and the general public to read the science, push for institutions to evaluate researchers based on the content of the science, and most importantly, researchers will receive scholarly credit by journals for getting their science right rather than getting it published. All of these are the necessary future for science, and our system will only be successful if these steps are taken.
David Kamper can be reached at email@example.com