NEW YORK, NY Dec. 5 (DPI) – More press outlets are trying to craft news stories solely from data and statistics, one of the odder media trends of our information-soaked age. And while reporters gaze more at the numbers and divine their meaning, readers too are instantly – and sometimes savagely – scrutinizing their conclusions.

A recent example: A young blogger for nytimes.com asserted that the divorce rate in America has been falling for 20 years, and any notions that the divorce rate remains high are, as the headline said, a myth. The author, a contributor to the Upshot column, declares that the data housed in an accompanying chart attest to it:

http://www.nytimes.com/2014/12/02/upshot/the-divorce-surge-is-over-but-the-myth-lives-on.html

But among the more than 500 reader comments, there was plenty of skepticism, including some remarks that were outright dismissive:

Are we all looking at the same chart? I see very little to indicate anything newsworthy. But most importantly — marriage rates are down significantly.  So we have traded “once was married” for “never been married”. That doesn’t seem to be mentioned here.

It has been a long time since I have been involved in the social sciences, but the chart seems to indicate that the study is not longitudinal enough. The chart indicates that all marriages reached the 15% divorce rate before the eighth year of marriage, with those married in the 2000s reaching it in eight years. Simple subtraction would mean that those marriages of shorter duration (2006-2014) have not yet reached the eight year threshold.

Comparing divorce rates of couples married 10 years to couples married 30 years is bad science. You need to wait until those married in the year 2000 reach 30 years of marriage, then see how many are divorced. As they say, there are lies, damn lies and statistics.

The author, Claire Cain Miller, may be right, but the divorce numbers haven’t budged much, either, leaving many to question whether the trend was even worth noting.  Meanwhile, the number of marriages has fallen much more sharply across the economic spectrum – which in many ways is a much more significant story.

An academic, one of three sources quoted in the report, weighed in with a subsequent blog post suggesting that the data may be wrong:

www.nytimes.com/2014/12/04/upshot/how-we-know-the-divorce-rate-is-falling.html

http://www.huffingtonpost.com/robert-hughes/is-the-us-divorce-rate-go_b_4908201.html