Employment Stats: Cherry Picking Season

Did you know that it is cherry picking season when it comes to employment statistics? It appears that some media reports are choosing statistics from to tell a better story rather than telling the correct story.

In a report that is quickly gathering steam in the Bay Area after originating in the Contra Costa Times, there was terrific news regarding the job growth in and around San Francisco. The headline said it all: South Bay and San Francisco metro areas are nation’s strongest job markets.

The data to support the headline is given as:

The Bay Area boasts the two strongest job markets in the country, a new report by federal labor officials shows. The nation’s strongest employment market is the San Francisco-San Mateo-Marin region, which powered to a 4.4 percent increase in total payroll jobs during the 12 months that ended in July, the U.S. Bureau of Labor Statistics reports. The South Bay was the second-strongest job market in the United States. Santa Clara County posted a 3.5 percent increase in total jobs over the one-year period…The third-strongest job market in the country was the Houston-Sugarland region in Texas, which was up 3.2 percent.

What if I told you that according to another report looking at the same time period, “The largest over-the-year percentage gain in employment was reported in Lafayette, La. (+10.3 percent)…” or “The largest over-the-year employment increase occurred in New York-Northern New Jersey-Long Island, N.Y.-N.J.- Pa. (+90,400)…”

The BLS

While employment statistics are not the easiest to read and interpret, I decided to visit the Bureau of Labor Statistics (BLS) website. Luckily for many of us, the BLS places their data into nice little boxes for reporting comparable numbers. By this I mean that the BLS divides informational releases into several sub-sections that pull out the data by qualification. For instance, think of data for the city you live in versus your region. These numbers will be quite different based on the population they sample.

It did not take more than a few moments to find what appears to be the source report for the article here.

Areas v. Divisions

The release from the BLS breaks down unemployment and employment statistics into two categories: Metropolitan Areas and Metropolitan Divisions. According to the Office of Management and Budget (OMB) Metropolitan Areas are defined as:

Metropolitan and micropolitan areas consist of one or more counties. OMB defined a conceptually similar set of areas in New England using cities and towns as the geographic building blocks, referred to as New England City and Town Areas (NECTAs). The LAUS program uses these NECTAs as an alternative to the county-based metropolitan and micropolitan areas in the six New England states. There are 372 metropolitan areas and 585 micropolitan areas in the United States. In addition, there are 8 metropolitan areas and 5 micropolitan areas in Puerto Rico.

The OMB then describes the Metropolitan Divisions as:

Eleven of the most populous metropolitan areas in the United States are composed of 34 metropolitan divisions, which are essentially separately identifiable employment centers within a metropolitan area.

So a Metropolitan area would therefore be a larger sample than the division that it contains. If anyone disagrees with this assessment please let me know in the comments section.

The Numbers

According to the BLS, employment is broken down by overall increase or percent increase. If you want to know the overall employment increase for a Metropolitan Division (Nonfarm Employment) you are presented with:

The largest over-the-year increases in the metropolitan divisions occurred in New York-White Plains-Wayne, N.Y.-N.J. (+78,600), and Los Angeles-Long Beach-Glendale, Calif. (+58,900).

However, you can find the media reported 4.4 percent increase for the San Francisco-San Mateo-Marin region under the Metropolitan Division (Nonfarm Employment) section:

The largest over-the-year percentage increase in employment among the metropolitan divisions was reported in San Francisco-San Mateo-Redwood City, Calif. (+4.4 percent), followed by Peabody, Mass., and Seattle-Bellevue-Everett, Wash. (+3.1 percent each).

Note that this is a percentage and includes Peabody and Seattle-Bellevue-Everett, neither of which was mentioned in the article. If you were to look for the “Santa Clara County” and “Houston-Sugarland region” numbers from the article you would have to check the Metropolitan Area (Nonfarm Employment) section which does not mention them until they are qualified as having had “average employment levels above 750,000 in 2011”:

The largest over-the-year percentage gain in employment was reported in Lafayette, La. (+10.3 percent), followed by Columbus, Ind. (+8.3 percent), and Texarkana, Texas-Texarkana, Ark. (+7.0 percent)…35 of the 37 metropolitan areas with annual average employment levels above 750,000 in 2011. The largest over-the-year percentage increases in employment in these large metropolitan areas were posted in San Francisco- Oakland-Fremont, Calif., and San Jose-Sunnyvale-Santa Clara, Calif. (+3.5 percent each), and Houston-Sugar Land-Baytown, Texas (+3.2 percent).

Note that this section gives San Francisco a less breathtaking number, actually making it equal to San Jose-Sunnyvale-Santa Clara.

However, if you want to view “increases” the picture changes again:

The largest over-the-year employment increase occurred in New York-Northern New Jersey-Long Island, N.Y.-N.J.- Pa. (+90,400), followed by Los Angeles-Long Beach-Santa Ana, Calif. (+86,300), and Houston-Sugar Land-Baytown, Texas (+83,700). The largest over-the-year percentage gain in employment was reported in Lafayette, La. (+10.3 percent), followed by Columbus, Ind. (+8.3 percent), and Texarkana, Texas-Texarkana, Ark. (+7.0 percent).

So how is the media supposed to interpret all of this information?

While explanations are not always best friends with tightening column inches, an explanation of where these numbers are derived from or a qualifier of each number is warranted. Following either of these suggestions, comparable numbers should be presented and not picked to fit a headline. After all, the lead could have been: Lafayette leads US employment gains.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: