MRC01
Major Contributor
I agree some data sets are clean enough, but even clean data is incomplete which can create a misleading picture.I'm as cynical as anyone. All data is subject to all sorts of bias, political or otherwise, that doesn't meal all data is corrupted. I think that the data on wages and GDP over time is pretty clean.
One example of drawing incorrect conclusions from clean data, relates to common misunderstandings of variance and how cohorts are formed, best explained by Phil Birnbaum: http://blog.philbirnbaum.com/2014/09/income-inequality-and-fed-report.html
Another example is Picketty's famous book from a few years ago. He made some conclusions from data, and while the data he used was clean as far as it went, he failed to consider the complete picture which undermined his conclusions.
Another example is studies on minimum wage. UW and Berkeley both studied the impact of Seattle's minimum wage, and drew opposite conclusions by applying different methods and assumptions to the same data. In fact, the city of Seattle first asked UW to do the study and when they didn't like its conclusions, they asked Berkeley to publish another.
My point, that even clean data can be misleading and analysis complex and subjective, is nothing new. Bastiat phrased it as considering both what is seen, and what is unseen. Hayek in The Fatal Conceit said, "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design." All this doesn't mean we can't get the answers, but it means it's not as simple as posting data and discussion here in this forum. Which is why I said what I did back in post #402.