USA 3 August 2014. The BLS survey of business establishments for July showed a one-month increase of 209,000 jobs. This increase meets the test of statistical significance at the 90 percent confidence interval and points to improved conditions in the labor market.

At the same time, the household survey showed an “increase” in the jobless rate from 6.1 percent to 6.2 percent but this increase is not statistically significant. Even so, many commentators mistakenly reported that “the unemployment rate ticked upward.” With a standard error of 0.2 percent points, the July jobless rate as estimated by the BLS and as compared to the June rate would have to rise to 6.3 percent or fall to 5.9 percent to be statistically significant. 

As well, the one-month increase of 131,000 in the number of persons employed is not statistically significant. Neither is the 197,000 increase in the number of persons classified as unemployed. The household survey, in other words, indicates no change in labor market conditions.

Only one conclusion can be drawn from these two surveys: the information is inconclusive. On the one hand, the economy seems to be improving. On the other hand, it is not.

Perhaps as a result of this inconclusiveness, the Associated Press reached for another story. AP reported that “U.S. employers have gone on a hiring spree this year, but America’s 25-to-34-year-olds are hardly among the biggest beneficiaries,” asserting that the employment-to-population rate fell from 76.0 percent in February to 75.6 percent in July. True enough, but this rate was higher in July 2014 than in any month between March 2009 and December 2013. 

Other evidence from the household survey seems to confirm the AP’s assertion. The number of employed persons in that age group slipped by 38,000 between June and July but this difference clearly is not statistically significant. On the other hand, employment among persons in those ages climbed by 701,000 since July 2013.

The problem is not that the number of 25-34 year olds with jobs has dropped since December 2007 as AP suggests with this language: “but without enough jobs, they’re more likely to live in their parents’ basements.” The BLS household survey indicates a 252,000 increase in the number of jobholders since the start of the Great Recession, an estimate that even though it may not be statistically significant demonstrates that employment opportunities for 25-34 year olds indeed have returned to levels just prior to the onset of the Great Recession.

Not only did the AP’s reach with 25-34 year olds exceed its grasp, it missed a much important labor market development regarding married persons.

Among married men, spouse present, there are 1,882,000 fewer employed in July 2014 than in December 2007. For married women, spouse present, the numbers also are grim: 967,000 fewer in July than at the start of the Great Recession. Married men and women are the heart of family and community life. No other demographic contributes more to consumer spending and national economic activity. Today they make up only 53.9 percent of all employed persons. In July 1980 they accounted for 62.8 percent of everyone employed.

Putting aside the question as to what accounts for this enormous decline, we wonder why the AP is preoccupied with 25 to 34 year olds, why they erroneously reported that this group is not benefitting from improved economic conditions since the end of the Great Recession, and why they missed the much more important labor market developments of married persons. Does this kind of superficial and sometimes error-ridden reporting have anything to do with the relentless cable news cycle that demands instant commentary on the latest economic estimates with little or no regard for statistical significance, over emphasis on month-to-month change at the expense of long-run developments, and a blind eye to what is really happening?

It’s time for AP and others in the media to pay more attention to what’s happening in the family room and kitchen and less on who’s taking up residence in the basement. 

Edward J. O’Boyle is Senior Research Associate with Mayo Research Institute.