Three years ago, the National Endowment for the Arts released a now-famous study, “Reading at Risk,” which said that “literary reading” among adults in America had declined steeply. I thought the study’s results were widely overblown in media reports and editorial columns because, if one read the fine print, “literary reading” did not include such things as histories, biographies, nonfiction books in general — fewer than half of Americans over 18 read novels, plays, poetry or short stories.

In other words, fiction reading had declined, and although that is indeed a worrisome trend, I suspect the study wouldn’t have gotten one-tenth the attention it did if that had been the headline (and if editorial columnists read past the study’s opening paragraph or its stats). Actually, sad but true: America has never loved fiction much; until World War II, we were primarily a nation of Bible and almanac readers. Fiction isn’t “useful.” It’s not “self-improving.”

But now comes a different NEA report on reading, this one on students, and it may deserve some alarm bells. The NEA decided to address the weaknesses of “Reading at Risk.” So, first, this is not a single study; it’s a compilation of a dozen studies. Second, it takes in all reading and corelates it to academic performance and even employer tests of basic writing skills.

The conclusion: Reading scores among elementary students have been improving, but scores for middle and high school students are declining — and these are tied to time spent “reading for fun,” any kind of reading for fun. Proficiency in reading skills is also linked to eventual income.

And this time, The New York Times article cites a couple of academics who still have reservations about the conclusions.