Monday, December 03, 2018

The Demographic Representativeness of the 116th Congress

This table* about the demographics of the 116th Congress is fun to analyze:


To see what it tells us about how representative the new Congress is, let's compare it to the population proportions of various demographic groups in it.  The last column gives the overall totals, and the third and fifth give the totals by the two parties.

Starting from the total percentages column, it's clear that among the larger groups women are quite seriously underrepresented, that blacks are represented in proportion to their population percentage (though these data don't let us see if this is true separately for black women and black men), and that Hispanics, as a group,  are also quite seriously underrepresented**.

The more fascinating columns to study are of course the party percentage columns.  Those reveal that the Republican Party really is the party of white men. Only seven percent of Republican Congress critters are women,  non-Hispanic whites are ninety-five percent of the total,  and the figures for all racial or ethnic minorities are as tiny as fly specks.  That the black representation matches the black population percentages overall is because the Republican under-representation is compensated for by relative numerical over-representation in the Democratic Party.

May I use this opportunity to, once again, complain about the diversity concept.  If you look at the rows in that table they show diversity, right?  There are wimminz there and all sorts of other demographic groups are represented.  So all is good.

But many of the table percentages are not the same as the population percentages of various groups***.   The system is clearly not representative of the country.  The diversity concept does not reflect that. 

In a sense "diversity" is not about fairness in the same manner as fair representation is.  Why it's so popular might be because it can be used by both sides for their own purposes.  Those who don't really want to see fair representation can add a couple of tokens to various committees, and, presto, diversity is achieved and complaining voices are silenced.  Those who fight for the rights of a numerically very small marginalized group may be able to get it over-represented by using the diversity argument.


*  My apologies for not crediting the creator of the table.  I copied it in this form from somewhere online.  If you know the name of the creator(s) please let me know and I will add the acknowledgement.

**  Once again, this table doesn't show Latinos and Latinas separately.

Note, also, that the Hispanic category is the one where the comparison of Congress percentages to population percentages are less useful.  This is because roughly one third of all individuals in this category were first generation immigrants in 2013. 

It takes time to become integrated into a culture, including seeking a political career.  A better comparison for the overall statistical representativeness of this group might consist of the population percentage of Hispanics who are at least in the second generation in the US (though that would be a very rough guide, too).

*** Depending on the context in which we examine diversity, the correct comparison might not be to population percentages but to, say, the percentages of people in various demographic groups who have been trained for certain jobs and so on.  Still, in all those cases the goal is to see if the system works fairly.