Good information, but I disagree with Lott’s conclusion.
The “household survey” of 110,000 people is far more subjective than the, ADP business survey. The household survey attempts to estimate 2 variables. The number of people working, and the number of people “looking”.
Using a smaller sample, unverifiable responses, and estimating 2 variables makes the “household” survey inherently more subjective than the business survey which directly represents a sample of about 22 million workers.
To understand the current employment trends, I believe it would be useful to consider “only” the “total Jobs” estimate, and ignore the “unemployment %”. The direct link to ADP is: http://www.adpemploymentreport.com/