Banner by: Jill McDonnell
Sensationalized Science & the Media
by Cassidy Power
Great news! Eating one square of dark chocolate every day will make you live longer! Honey is a natural antibiotic! The cure for cancer has been discovered! Everyone reads headlines like this which are beyond misleading. It’s a common trope that one day something will be a cure-all superfood and the next day it will be cited as the leading cause of cancer. But is it science or the media that is the the culprit of this incessant flip-flopping and is that flip-flopping a detriment to society?
The relationship between science and society is fraught with miscommunication and mismatched expectations. Even the languages they speak are different. Science is quick to blame the media for the proliferation of exaggerated or false information, but science has a lot of work to do as well.
When scientific research is quite simply bad, journalists without scientific training are hardly to blame for the proliferation of terrible science. Considering most published research is false, it’s no surprise that headlines conflict day-to-day. Getting published is integral to any scientists career, resulting on an emphasis on quantity over quality. This “publish or perish” environment has been shown to decrease innovative research by making it more of a risk. Innovation takes time, and with the pressure to constantly publish, very few researchers have that kind of time. Wolfgang Pauli, a theoretical physicist, criticized this phenomenon and is quoted as saying “I don’t mind your thinking slowly. I mind your publishing faster than you can think.”
The peer-review system is designed to act as quality control, but this system has some major flaws. The first is that researchers are not paid for their time spent reviewing others’ work, and it takes valuable time away from their own labs. Additionally, there is little data on the peer-review system, making regulations difficult. Even determining the number of articles a researcher reviews in a year proves elusive. Reviewers have been known to pass off articles onto their subordinates or give a cursory glance at the results without examining the methods. Furthermore, there was recently a scandal where researchers gamed the system to allow them to review their own work. Review rings are also popular. These are groups of researchers that recommend each other to review papers and agree to only favorably comment on the submission.
The peer-review system is hardly an effective means to control the quality of research published. This issue is compounded by science journals who are more concerned with the results than the methods. That’s why articles appear frequently about false cancer cures and causes. How do these studies get false positive results? It all comes down to statistics.
In the science world there’s all sorts of cutesy names for fudging data, manipulating statistics, and other suspect practices. Take “p-hacking,” the practice of analyzing data in different ways in order to ‘prove’ a certain outcome. P-hacking is responsible for a large number of false positives because it enables researchers to label research as statistically significant even when it may not be. The website FiveThirtyEight has a interactive tool to explain this. It focuses on the economic effects of the Republican and Democratic parties. The tool presents a variety of variables and metrics as well as ways to define a good or bad economy. Depending on the variables selected, the results “prove” Democrats or Republicans had either good or bad effects on the economy.
Not all false positives have such complicated back stories. Some of them are just the result of bad science. For example, one research team from Vanderbilt University had to retract its findings after realizing they had ordered the wrong mice for the study. Furthermore, when examining the studies that pop science articles are based, one often finds that the studies only had a handful of participants and thus are statistically insignificant. These problems are compounded by the fact that the media often misrepresents the weight or meaning of scientific studies. The American Press Institute found that six in ten Americans read headlines and nothing else. Headlines are notoriously sensationalized. John Bohannon notoriously was able to lead journalists to report on a fake study about dark chocolate helping participants lose weight. While the study itself was poor, the journalists made no effort to verify with experts if this study had any significance. Furthermore, the researchers never claimed any causation between chocolate and weight loss; this was conclusion was drawn by the journalists reporting on it.
Science, society, and the media all speak different languages. To a scientist, a theory is a near certainty while to society it’s a flimsy possibility. Scientists also are less likely to assign weight to a single study, whereas society associates anything scientific with truth. The media is supposed to be the translator but most journalists don’t have a background in science or statistics and are unable to judge the validity of various studies. Over the past few years, scientists have sought more active roles in how their research is disseminated.
The field of science communication is largely formed of these scientists, unhappy with how their life’s work has been presented. Even scientists who don’t seek this type of expanded role are often trained in how to interact with the media and lay people as part of their orientation to a particiular institution. The European Commision, Brown University, and the National Institute of Health are all among such institutions that have a guide for scientists interacting with the media.
It is equally, if not more, important for readers to understand how they should consume media. There are a few ways to avoid being duped by sensational journalism. The first is to read more than just headlines. Aside from that, readers can look at the number of participants in a study. If a study has less than 100 participants, the conclusions are essentially meaningless. Readers should also take note of the experimental design: what was actually being tested? Who were the test subjects? Was the study carried out on humans or mice? Finally, did the scientists themselves draw any conclusions?
It’s easy to blame the media for the proliferation of false scientific claims but scientists and readers must shoulder some of the blame. At the most basic level, the science world needs to innovate a means of self-regulating. The media must hold itself accountable and verify any claims made in headlines and throughout an article. Readers too have to make conscious decisions to critically analyze the media they consume. Without intentional changes from all parties, sensational journalism will continue to reign supreme.