Why the Media Has Lost Faith in Science

We see are seeing increasing numbers of media stories about problems with peer-reviewed scientific research. Whether it is a story about a Harvard professor falsifying data, or debates about what is scientific fact versus fiction, science seems to be embroiled in controversy. Last week I saw an article about how peer-review has failed to identify scientific fraud. Such cases make for great headlines, and may explain why the media has lost faith in science. But they are just the tip of the iceberg when it comes to the integrity of science.

The Scientific Method

Science is a means of using systematically collected information to draw conclusions. Scientists design research studies, collect data, and use the data to reach conclusions. It begins with a research question–an issue that can be addressed with data. Perhaps there are hypotheses–the scientist’s best guess about what the results will be. Ideally, scientists conduct their studies in good faith. They plan their methods in advance, do their best to follow the plan, and are open to any outcome. The answer to the question might be ‘yes’, but it might also be ‘no’. Sometimes the answer is unclear, which leads to the conclusion that more research is needed.

The methods of science are designed to reduce subjectivity, but scientists are people and they have biases, both conscious and unconscious. This is why our methods are preplanned and systematic. For example, drug trials use “blind” procedures where a sample of patients are randomly assigned to either get the new experimental drug or a placebo. The people who administer the drugs don’t know which patients are getting the new drug and which are receiving the placebo. This minimizes the bias that might be introduced had they known which patients got the new drug. Unfortunately, too often the methods become corrupted by the motives of individuals who engage in a variety of unfortunate practices.

Follow the Money

Science is a business, and scientists have skin in the game. Whether they are professors working at universities, or scientists working in industry, there is tremendous pressure to achieve interesting, marketable, and publishable results. Some of the motives of scientists include.

• Activism: Many scientists care deeply about social issues, and they want their science to help the cause. There is no doubt that science can play a vital role in social policy, as it provides critical information about what approaches are effective versus ineffective, and what unintended consequences might be of specific actions. Too often, however, in their enthusiasm, scientists can allow the activism to overwhelm the science. Research can get turned inside out by beginning with the conclusion and then cherry-picking results that support it. I once read a paper dealing with a social issue that claimed research support for the author’s position from a single study. What went unremarked was that there were 24 other studies addressing the issue, and all found the opposite results. So much for an objective weighing of the scientific evidence.
• Money: It has become increasingly clear in recent years that businesses sometimes play fast and loose with the science. This is particularly true in healthcare where drugs and medical treatments are expensive to create, and must have scientific evidence to get approval for use. A professor I know who does medical research received a grant from a drug company to conduct a drug trial. His company contact told him that if he got good results, there would be more money coming. Perhaps the contact merely meant that if he was conscientious in his methods and stayed on schedule, they would continue to use him. But that is not what he heard.
• Career Survival: Professors, particularly at research-oriented universities, live in the publish or perish world. This is a hyper-competitive environment where there is tremendous pressure to achieve good results and publish in prestigious peer-reviewed journals. A paper in Nature or Science can make a career. It is so easy to rationalize questionable practices like HARKing (doing science backwards by basing hypotheses on your results rather than using results to test hypotheses) and p-hacking (conducting a series of manipulations of the data until positive results emerge). These practices are so wide spread, and often encouraged by peer reviewers, that it is easy to ignore that they distort the science.

Why the Media Has Lost Faith in Science

It is hard to fault the media for giving the impression of having lost faith in science. Whether it is outright fraud, such as data fabrication and plagiarism, or sloppy scientific procedures, it is the role of media to call us out on our failings. Of course, negative stories about science fraud and the politicization of science have undermined the public’s faith. But perhaps that is not entirely a bad thing. We should all be cautious in blindly accepting scientific conclusions. Many of our conclusions, methods, and theories are controversial even within the scientific community. Often people cherry-pick the science to support their positions so they can convince us to believe what they want us to believe. Sometimes the motives are pure and sometimes not.

The media attention on misconduct of scientists has perhaps had a positive impact. It has made universities and funding agencies more aware of the need to oversee scientific integrity. At my university there is an entire department devoted to policing bad behavior. I learned this first hand when I was asked to serve on a peer review committee to investigate a plagiarism case against a professor. The committee consisted of a university administrator, an attorney, and several senior faculty. We found no evidence for wrong doing in this case, but if we had, the professor likely would have been fired. It was serious business.

Technology has helped in the effort to maintain research integrity. The widespread availability of plagiarism checkers are used by journals to check submitted papers so the very public cases we see in the news are unlikely to occur moving forward. My university requires every submitted master’s thesis and doctoral dissertation be checked.

Some researchers have developed ways to detect falsified data by identifying highly unlikely patterns in results. Faked data often has tell-tale signs that have led to people being caught. There are also efforts to replicate results of published studies. Although failure to replicate findings doesn’t necessarily mean the data were fabricated, it allows for scientific self-correction–finding that previous conclusions might be wrong.

Conclusions in science should be based on a weighing of evidence from a variety of studies using different methodologies. No single study is conclusive, and needs to be independently replicated before we accept the conclusion. Science itself is always tentative. The next study might show us that our past conclusions were incorrect and our methods were flawed. It is not unreasonable to be skeptical of scientific results, As scientists we need to do our best to be as objective as possible in our methods and conclusions, but we should not be surprised that others may have lost faith in science as always providing the right answer.

Image generated by DALL-E 4.

SUBSCRIBE TO PAUL’S BLOG: Enter your e-mail and click SUBSCRIBE