Charmed by data

29 Maggio 2011

Recently I have found on National Geographic magazine this chart which includes data that I find fascinating.  It is a very good visual rendering of  an analysis performed by the  Pew Research Center on data concerning the inter-racial marriages  celebrated in the United States in 2008.

The part of the picture which I find intriguing is represented by the two relatively larger rings in the row just below the largest one on the top.  They represent respectively the marriages between whites and asians, and whites and blacks.  What is striking in both cases is the disproportion between the two possible combinations of the two sexes. The percentage of asian females marrying white males in the first case, and the percentage of black males marrying white females in the second case, both represent approximately three-fourths of the total number of marriages.

What explains this unbalanced way in which the sexes combine across races?  National Geographic itself seems unwilling to advance hypotheses.  Is it genetics, the distribution of wealth, power, sexual preference, culture, some combination of those, or something else? The mind starts producing hypotheses almpost spontaneously.

This example shows what characterizes in my eyes good use of data.  This is an elaboration on (I imagine)  a widely available public data set which produces a finding only after it has been examined with an interesting question in mind: in inter-racial marriages, the two sexes combine evenly? It is interesting data because it raises many more curiosities than it satisfies, while presenting facts that are not well known.  The data does not represent opinions, but facts: in this case a behaviour.  The visuals in this case  seem to me to provide an excellent example of how to summarize many facts in an evocative way, and certainly contribute to making dry information into news.