Lots of people are throwing around statistics about COVID-19, cases, deaths, ICUs, ventilators, even percentages, a challenge for those who forgot long division. And then there are the software/programming types and computer models that project what yesterday’s and today’s numbers will mean for tomorrows.
Keep in mind though that in country as large and opinionated as the United States, tabulating numbers (which depend on people keeping records) is a challenge. For instance:
In his effort to narrow the stunning divide between the U.S. and other industrialized Western nations on the rates at which they incarcerate their respective citizens, Latzer’s argument is limited to people housed in prisons. His analysis thus quite explicitly ignores the almost 800,000 people who are detained, at any given point in time, in city and county jails across America housing half as many prisoners as state and federal prisons do.
Excluding jail inmates in a debate concerning over-incarceration, however, is just as illogical as ignoring foreign-manufactured vehicles from an assessment of the number of automobiles on the road. The place where prisoners are housed (prison versus jail) is just as irrelevant as the place of manufacture of automobiles in arguments over whether we have too many inmates and cars in the United States. To have any real-world significance, each normative inquiry must begin with an accurate appraisal of the empirical phenomenon being assessed. In the case of over-incarceration, that means factoring in the hundreds of thousands of people locked up in jails along with their counterparts in federal and state prisons.
According to the U.S. Department of Justice, “County and city jails in the United States reported a total confined population of 745,200 inmates at midyear 2017.” In that same year, the government reported that prisons contained 1,489,363 inmates. Despite the fact that non-lawyers often use the terms “prison” and “jail” interchangeably, prisons are reserved for persons convicted of felonies and sentenced to more than a year behind bars. Jails, by contrast, house less serious convicts serving custodial sentences of one year or less and persons being held on pending charges not yet adjudicated. People in jails and prisons alike are incarcerated and thus directly relevant to debates regarding over-incarceration.
At a time when we are trying to sort through critical cases of COVID-19, ordinary ones, and asymptomatic carriers, remembering that Americans also have trouble counting people who commit felonies may provide comfort. You can always juke the stats.
6 thoughts on “The Nation’s Math Crisis”
A book that was first published in 1954, and as far as I know is still in print, is “How to Lie with Statistics,” by Darrell Huff. I acquired my copy as an undergrad back in the 60’s and it still sits on the bookshelf in my office. It’s content is still valid and should be a must read for everyone, especially nowadays.
The corresponding book for language is Less Than Words Can Say by Richard Mitchell.
Yep. Remember those natural science lab courses where one did an experiment and then wrote a report? One learned the meaning of “dry lab”. If one knew the answer and knew the way to get that answer from data, one could create the data (bump the scale, misread the yardstick, cross out and change the tabulated record of measurements) to get the answer. One also learned about “error analysis”, where the lab report discussed how come why for one did not get the right answer.
Currently we see the results of folks never having learned from such experiences. The various places supplying stats about CCP-Bat virus don’t have a common agreed upon definition of what counts for infection, how they decided what part of their population is infected, what part of the infected people had which other disease, even to what they will attribute any deaths (the virus or the comorbidity). Furthermore, some of those places have great reason to “cook the data”. Then, even tho they knew the data had a significant range of contradictions and they had no way to distinguish how the supplied data errs, others used it to make predictions. The range of ‘expert’ projections hints at the root problem. Then, based on those predictions, people with power are making policies with catastrophic potential.
What could go wrong? Answer: idols fall. Sports, entertainment, wealth, economy,….
Who says God does not have a snarky sense of humor.
I thought Ben Castle did the best overall assessment of this situation. He has obvious training in logic, something too few people have nowadays and it’s essential to science and statistics (and thinking). When can we go back to church? Our state government is taking a “minimax” approach to this whole thing and locking us up indefinitely, but we must obey God rather than men.
Time to go back to church. What is it with these city-state dictators? File NYC under “not redeemed. “