Ever heard the phrase, “correlation does not imply causation”? It’s a favorite of statisticians, who bristle at people’s tendency to draw errant conclusions about data because they conflate the two terms. Here’s one particularly offensive, oft-misconstrued example of that mistake: crime rates are higher in African American communities than in white or Latino populations. Does that, therefore, mean that people with black skin are more criminally inclined than people with skin of a different hue? Of course not. Income level, racism, population density and geography, educational opportunity, history, and police reporting/monitoring standards are all central factors in crime statistics—none of which are revealed in a simple chart of crime-rates-by-race. In other words, correlation (a relationship between race and crime) does not prove causation (that race determines crime rate).
Thate same numbers may be employed to prove opposing points is at the heart of a current debate about the efficacy of OSHA. A provocative infographic on Compliance and Safety asks the question: Is OSHA a Wasteful Regulatory Nightmare or Common Sense that Saves Lives? Proving these two utterly different conclusions are statistics about workplace deaths in recent decades, companies’ worker compensation costs prior to and after OSHA inspections, the impact of fines on company behavior, and the like.
Presumably, all the data is true. Nonetheless, it is used to draw strikingly different conclusions. For example, a graph of workplace fatalities between 1940 and 2011 reveals a steady decline, from 34,000 deaths per 100,000 employees in the early 1940s to under 5,000 per 100,000 in the current decade—a trend that remained steady after the launch of OSHA in 1971. The conclusion: “OSHA does not have a measurable impact on workplace fatalities.” (If that’s true, then I have egg on my face: in last week’s post, I noted that “[OSHA’s] work has contributed to a 62 percent decrease in workplace fatality rate and 42 percent occupational injury rate since its founding in 1971. Is it coincidental that over the same period, American labor productivity has also increased markedly? We think not.”)
In the opposing column of Compliance and Safety’s infographic, under the heading “OSHA Regulation Saves Lives,” a bulleted list demonstrates specific realms in which OSHA regulation and legislation has significantly reduced such incidents as brown-lung disease, grain-bin explosions, and Hepatitis B infections in healthcare workers. It additionally cites a 2012 study by Harvard University that estimated that universal OSHA inspections at American companies would save us a total of $20 billion (per year? Total? Not sure—another reason to question statistics when they’re thrown around like this).
So is OSHA a net good or a waste of government money and managerial manpower? Given that the numbers hold few truths, let’s look at what OSHA (which enjoyed a budget of $583,386,000 in 2012–a time when the U.S. government has been slashing spending) has to say for itself. The organization’s webpage provides safety resources for the following issues: classification and labeling of chemicals, safety and health solutions to winter storms, flu season; it compiles comprehensive data on everything from employment- and ethnicity-specific workplace-injury rates to employers that have violated different OSHA regulations.
That alone seems like a pretty good resource for Americans to have at their disposal. So what’s OSHA’s conclusion about itself? “In four decades, OSHA…[has] had a dramatic effect on workplace safety….Worker deaths in America are down….Worker injuries and illnesses are down….,” the organization says. “OSHA is making a difference.”