Tuesday, October 02, 2018

BLS: Don't Do What We Did

Wow, what a mess. In a 33-page Monthly Labor Review article, the Bureau of Labor Statistics explains its delay in releasing the data on what it calls "electronically-mediated work"—defined as "short jobs or tasks that workers find through websites or mobile apps that both connect them with customers and arrange payment for the tasks." Here is the key sentence:
"BLS should not again attempt to collect data about electronically mediated work using the four new questions fielded in the May 2017 CWS [Contingent Worker Survey]."
In other words, the first attempt made by the BLS to count the number of electronically-mediated workers was a failure.

Some background: In June, the BLS released its estimate of the gig workforce, measured by a special supplement to the 2017 Current Population Survey. The results were surprising, showing no growth or even decline in the gig workforce since 2005. But the survey did not include those who engaged in electronically-mediated work. These figures would be revealed, said the BLS, before September 30 in a Monthly Labor Review article. As promised, the article appeared on the BLS web site on Friday, September 28th. It is a sorry tale of earnest efforts to add four questions on electronically-mediated work to the Contingent Worker Supplement of the Current Population Survey and the flawed numbers that resulted from those efforts.

So flawed were the results, in fact, that the BLS was forced to recode the answers. There were too many false positives. To recode, BLS statisticians probed the survey's microdata files for clues as to whether respondents were or were not actually engaged in electronically-mediated work. The "yes" answers to the survey's four questions summed to 5 million respondents engaging in electronically-mediated work in the past week, or 3.3 percent of the nation's employed workers. But after probing the microdata and diligently recoding, BLS statisticians reduced the number to just 1.6 million, or 1.0 percent of the employed.

"If BLS were to collect data about electronically mediated work in the future," the article concludes, "questions would need to be substantially revised. It may simply be that the concepts are too complicated for four questions to properly identify all the information BLS was attempting to measure." Kudos to the BLS for being forthcoming about this failure, but we're back to square one in our understanding of the size and shape of the gig economy.

Source: Bureau of Labor Statistics, Monthly Labor Review, Electronically mediated work: new questions in the Contingent Worker Supplement

No comments: