When I showed the pie graphs from Paul Solman, demonstrating the distribution of wealth in the United States compared to wealth distribution in Sweden, to a co worker, she dismissed them immediately. "Oh," she said, "That's all just numbers. You can bend and twist them any way you want."
I was astonished by her response, and assumed she simply did not want to deal with the implications of what those numbers implied: Namely, that capitalism, at least American capitalism is broken.
But today, in the New York Times, a shining example of the necessity to look closely at how numbers are gathered is essential. Emily Owens, a professor of Policy Analysis at Cornell, takes us through, step by step deconstruction of what those numbers actually mean. Such an exercise is the negative pole to the positive of the Rush Limbaughs and Bill O'Reilly's of the world, who find a number which supports what they already believe and run with it.
The numbers in question concern the murder rate in the United States before and after the initiation of Prohibition after the passage of the 18th Amendment . Census mortality statistics showed the national homicide rate jumped 40% and after it was repealed in 1933, murder rates plummeted. So, QED, criminalizing alcohol caused a huge increase in murder and, by extension, major crime.
It turns out, there are a few problems with this conclusion. In 1900 the Census counted deaths in only Indiana, Michigan and in the New England states. After 1910, states like Nevada and Texas were counted and those states had then and continue to have higher murder rates, before and after Prohibition. You simply started to count the murders in the murder states.
Second, it turns out Prohibition did not happen all at once: Prohibition began at the state level 70 years prior to the 18th amendment. By the time the 18th amendment took effect, 32 states were already dry. And when Prohibitions was ended by the 21st amendment, many states remained dry. Comparing murder rates in individual states from the dry to the wet years shows a 26% difference, not 40%.
But, you may say a 26% difference is still large.
So we get to thirdly: Other variables changed over the 14 years of Prohibition. Urbanization, population shifts of displaced Southern blacks into urban Northern ghettos created an explosive mix, as did the stock market crash of 1929. Immigration was also in the mix, as immigrants from Europe and China flooded into the inner cities.
Then there is the effect of murder committed by inebriated people. During Prohibition, there may or may not have been less drunkenness, but likely there was not more--and this would have contributed to fewer murders rather than more. This is a number which is difficult to tease out. Different analyses place the effect of Prohibition in a given state as anything from a 5% increase in murder to a 13% decrease.
The public perception through TV dramas and movies and newspaper headlines is that Prohibition resulted in a rampage of crime and murder, but the effects of fewer drunken murderers, the effects of urbanization, new racial mixing, immigration, financial stress caused by the stock market crash and the Depression, all likely had as much effect, if not more, on murder rates during Prohibition, which may have risen, fallen, or stayed unchanged during the 14 years of Prohibition.
The same sort of analysis can be done for "Death Rates" from various diseases which are always quote in papers in the medical literature and popular press. Knowing how death certificate data, from which these numbers are generated, I always dismiss all death rate data as bogus.
Ever since academic and local hospitals stopped doing autopsies, and more people started dying at home, and cause of death was "determined" by local physicians, death rate numbers have been sheer fantasy. Nobody knows any more what diseases are killing people in American communities.
It's simply "Garbage in, garbage out."
No comments:
Post a Comment