Overprecision in Estimates of Physical Constants

Science depends on accurate estimates—and on honest assessments of how uncertain those estimates are. This study shows that scientists often express too much confidence in their reported values of physical constants, such as the charge of an electron or the speed of light. By analyzing 159 constants over nearly 50 years, the authors find that many published confidence intervals are too narrow, since the truth falls outside the published confidence intervals more often than it should. This widespread “overprecision” has remained persistent over time. The findings suggest that scientists, like everyone else, are prone to overconfidence—and that science would benefit from more cautious and well-calibrated communication about uncertainty.