Thursday, 8 December 2011

Stats for Social Scientists


  What’s wrong with the infographic? The problem is a basic misunderstanding of combinatorial mathematics. For example, some of the men who are in jail might also be not interested in women. This can be described by a Venn diagram like the one below comparing whales, fish and scrimp where some of the characteristics of the species overlap. If one uses a valid mathematical interpretation of the situation then there are a lot more black males available for black women than three per hundred.

  The problem with this faulty analysis is that it gets into the literature (where are you peer review?), gets published as books and ends up as articles in those lofty arbiters of received wisdom the Economist and the NYT. Here is an example on fore mentioned topic taken from the Economist.  

AT SOME events on his book tour, black men have accused Ralph Richard Banks of advocating genocide. In fact, the Stanford professor of family law has merely written a book called “Is Marriage for White People? How the African-American Marriage Decline Affects Everyone”. But abuse is what you get for suggesting, as Mr Banks does, that black women—not only the “most unmarried” group in American society but also the one that least intermarries with other races—should look to white, Latino or Asian men as potential mates. After all, the alternative is often no marriage or relationship at all.

  Fortunately we have access to data which reputes the conclusions. In Economics, the American Census Bureau is an excellent source but as I stated in an earlier post, the annual Statistical Abstract of the United States which has been published since 1787 will no longer be available as of next year in print or online and this puts a big hole in easily accessible data. I discovered last week that the Census Bureau may not do its five year census of data used in the determination of the American GDP as well as cancelling the annual household census used for the calculation of household debt due to “budgetary problems”. Can you imagine having any sort of objective discussion of wealth inequality without this information? The conversation will be reduced to my ideological stance and anecdotal evidence versus yours which will certainly help the libertarian, Hyek loving, Koch brother financed right wing.


1 comment:

  1. Hm, now I am finally starting to figure out what your field might be. I never figure out how you get to travel so much and so on to this unlikely combination of places but perhaps you are running statistics on them???