TUCSON, Ariz., March 12, 2025 (GLOBE NEWSWIRE) — Governments issue regulations with tremendous impact on the economy and on health practices. These regulations are based on published scientific research. But there is a crisis of irreproducibility (falseness) of research claims in science, write S. Stanley Young, Ph.D., and Warren Kindzierski, Ph.D., in the spring issue of the Journal of American Physicians and Surgeons.
The authors take test statistics from meta-analyses and use a graphical method to independently confirm, disprove, or identify ambiguity (uncertainty) in the reported findings and research claims. They also estimate the SearchSpace to determine the number of possible hypotheses tested.
If a researcher tests 1,000 hypotheses, he can expect to find 50 (5 percent) false positive results in the same set of data that are “statistically significant” to the magical P<.05. This is called “data dredging” or “fishing expeditions,” the authors state. It can produce large numbers of unreliable published studies.
The authors demonstrate that most research claims regarding the hazards of small particulate matter (PM2.5), health outcomes of eating red or processed meat, or the benefits of public masking to prevent respiratory illness are false, uncertain, or unproven. Their analysis does show the validity of claims that lockdowns were associated with increased domestic violence. Claims concerning implicit racial or gender bias were not validated by their methods.
“Bad science has crowded out good science in the literature,” the authors conclude. “Bad science permits governments and policymakers to develop policies and regulations in many areas—including the environment, health, and the workplace—with no public benefit or with actual harm.” Their methods also help to identify and report good science.
The Journal of American Physicians and Surgeons is published by the Association of American Physicians and Surgeons (AAPS), a national organization representing physicians in all specialties since 1943.
Contact: Jane M. Orient, M.D., (520) 323-3110, janeorientmd@gmail.com
