Jeremy Piven has a great blog today on why the latest Department of Labor unemployment data is misleading us – Simpson’s Paradox:

Simpson’s paradox (or the Yule-Simpson effect) is an apparent paradox in which the successes of groups seem reversed when the groups are combined. This result is often encountered in social and medical science statistics, and occurs when frequency data are hastily given causal interpretation; the paradox disappears when causal relations are derived systematically, through formal analysis.

The paradox example in the DOL data (which Jeremy clearly explains) shows that even though the total unemployment rate is lower than it was in 1983, the rate for some very key groups (including “college grads”) is in fact worse.

This is a great example of why we need to be very careful accepting high level metrics for any critical data, and why segmenting the workforce is critical – if we don’t, we risk losing vital information in the averages.  Segmentation is critical to all workforce planning and analysis…are you doing a good job of it, or losing important information to Simpson’s Paradox?