The March 5, 2010 New York Times headline read, “jobless rate holds steady, raising hopes of recovery”. Apparently it doesn’t take much to raise hope, at least among those doing the reporting. Why is two consecutive points with the same value (i.e. 9.7% in both January and February) a reason for hope? Would two consecutively made baskets in basketball constitute a scoring run? How about two winning hands at the blackjack table? Clearly not! Two points does not constitute a trend. Thus, the call for hope of a recovery is misleading; the result of a lack of understanding of how to read variation.
This is not the first instance of a misreading of data. In June 2009 the headline read, “Hints of hope in jobless data even as rate jumps to 9.4%”, and in August 2009 the message was “…the jobless rate unexpectedly fell to 9.4 percent, from 9.5 percent, the first decline since April 2008.” Such commentary is not reflective of any understanding, it merely reflects reaction to a difference. It is merely reporting the large or the small or the most recent—any uninformed person could do this! Although it may sell papers it doesn’t really help anybody.
Unfortunately such misuse and misinterpretation of data is epidemic! Given the widespread use of variance budget reports, it happens in business organizations on at least a monthly basis (see https://forprogressnotgrowth.com/2010/01/01/by-the-numbers ).
Those using data should learn how to understand the variation in the data they are using, and stop telling stories. Until there actually is a (meaningful) trend in the data, reflective of a predictable pattern or of a change in the system, people should limit their story telling to children at bedtime—at least this would serve some good.
Our world is one of systems within systems, where each responds to and produces variation. Thus, given the pervasiveness of variation, the need to learn how to understand it is paramount. Yet the lack of statistical thinking is epidemic.
We wouldn’t think it was okay to place someone in a responsible position that couldn’t read, yet we regularly place those ignorant of statistical thinking is such positions. Why is an ability to understand systems and the associated variation produced not seen equally as essential to competence as an ability to understand the written word?