Let There Be Light: Statistics and Law
Updated: Apr 15, 2020
Statistics and law have more in common than one might think. For instance, the public often carries a dim view of each. Mark Twain’s powers of observation concluded that “[t]o succeed in other trades, capacity must be shown; in the law, concealment of it will do,” and made famous the statement “There are three kinds of lies: lies, damned lies, and statistics.” As one practicing both disciplines, I’ve been treated to variations on these themes for years. I’ve observed that nobody likes lawyer jokes as attorneys do, and thankfully statisticians have a sense of humor.
Besides these overcast views, law and statistics have another aspect in common: shedding light in dark places. In law, truth-seeking is fundamental. Civil and criminal procedures aim to bring wrongdoing to light–to find truth. They establish sets of rules and principles for investigation, for bringing forth witnesses, for exposing
conspiracy, and finding wrongdoing. A critic would call these idealistic phrases, as law frequently functions as a bludgeon, but in its nobler moments law brings truth out of dark places and then sets it right. There’s nothing quite like diving deep into a case, seeing the patterns, and setting the puzzle pieces into an organized picture.
Statistics functions the same way. Statistics as a discipline is about finding light in unorganized data, revealing trends, and summarizing hidden patterns. It goes to extensive lengths to determine when things happen by chance or according to causal mechanisms. It is a wondrous experience to take a chaotic dataset and see underlying structures emerge through careful analysis, computer programming, and justified statistical assumptions. There’s nothing quite like diving deep into data, seeing the patterns, and setting the puzzle pieces into an organized picture.
When law and statistics combine, as they so often do in modern litigation, one might worry that Twain’s dismal views of each would synergize into a deep swamp indeed. On the contrary, statistics in litigation can shed remarkable light on wrongs, allowing decision and action where none was possible prior to analysis.
Consider just one case. In Bouaphakeo v. Tyson Foods, 577 US __ (2016) the plaintiffs alleged that Tyson Foods had violated the Fair Labor Standards Act with regards to time spent donning and doffing protective gear in a meat-processing plant. No precise records had been kept of these times, and so it was impossible to tell precisely how long each individual plaintiff had spent in these activities. Yet, an industrial relations expert came to the plant and after studying the workers, produced statistical evidence of how long the activities took. Using this statistical study along with records of when employees had worked in various departments, it was possible to shed at least a partial light on the potential damages suffered by class members. As there was no other possible way to recover individual times, statistics were essential in their potential recovery.
I have often joked with friends that nobody likes an attorney until they need one. The same could be said about statisticians. But when needed, these two remarkable disciplines have the tools to enlighten the dark in ways others can only dream of.