If that’s not you, persevere and you’ll find enough substance to be rewarded. When the material gets too weighty, it’s worth slowing down or even rereading. We live in this kind of world now. Wheelan’s book deserves to be a candidate for the reference work to go along with your pocket Bill of Rights and online dictionary.
Who knows? Statistical sophistication could get a foothold in political campaigns. Journalists could stop asking candidates who the prime minister of Japan is or which newspapers they read and instead ask whether an increase in Warren Buffett’s income changes the mean or the median of U.S. income or whose data the candidates rely on for their position on education and what the strengths and weaknesses of the data are.
Wheelan is good at running through the statistics basics and interrupts his explanation with useful caution. “Statistics cannot be any smarter than the people who use them. And in some cases they can make smart people do dumb things” is his way of summing up an explanation on the shortcomings that helped cause the financial crisis.
With technology helping gather, store and exploit ever more amounts of data, statistics is posing uncomfortable questions. Correlation isn’t causation, but is it possible that correlation is good enough? Scientists may not be able to draw the line between cause and effect, but the strong relationship backed by more data than anybody ever dreamed possible may be too much to ignore.
The Republicans, of course, face a correlation temptation — or is it closer to a superstition? — as the 2016 election approaches. Former Florida Gov. Jeb Bush would help test whether the Bush name is still a variable that signals success. Then again, even a victory would still be only one more data point in the sample, vulnerable to future data that could change the relationship.
Statistics are like that.
Randolph Walerius is an analyst for the CQ Roll Call Washington Securities Briefing.