Study Links Drop in Test Scores to a Decline in Time Spent Reading

Interesting negative correlataion between reading and test scores. The "reading at-risk" crisis mentioned here is a good way to frame what we've been seeing for years: Declining test scores by the middle-school years. -Angela

November 19, 2007
Study Links Drop in Test Scores to a Decline in Time Spent Reading

By MOTOKO RICH
Harry Potter, James Patterson and Oprah Winfrey’s book club aside, Americans — particularly young Americans — appear to be reading less for fun, and as that happens, their reading test scores are declining. At the same time, performance in other academic disciplines like math and science is dipping for students whose access to books is limited, and employers are rating workers deficient in basic writing skills.

That is the message of a new report being released today by the National Endowment for the Arts, based on an analysis of data from about two dozen studies from the federal Education and Labor Departments and the Census Bureau as well as other academic, foundation and business surveys. After its 2004 report, “Reading at Risk,” which found that fewer than half of Americans over 18 read novels, short stories, plays or poetry, the endowment sought to collect more comprehensive data to build a picture of the role of all reading, including nonfiction.

In his preface to the new 99-page report Dana Gioia, chairman of the endowment, described the data as “simple, consistent and alarming.”

Among the findings is that although reading scores among elementary school students have been improving, scores are flat among middle school students and slightly declining among high school seniors. These trends are concurrent with a falloff in daily pleasure reading among young people as they progress from elementary to high school, a drop that appears to continue once they enter college. The data also showed that students who read for fun nearly every day performed better on reading tests than those who reported reading never or hardly at all.

The study also examined results from reading tests administered to adults and found a similar trend: The percentage of adults who are proficient in reading prose has fallen at the same time that the proportion of people who read regularly for pleasure has declined.

Three years ago “Reading at Risk,” which was based on a study by the Census Bureau in 2002, provoked a debate among academics, publishers and others, some of whom argued that the report defined reading too narrowly by focusing on fiction, poetry and drama. Others argued that there had not been as much of a decline in reading as the report suggested.

This time the endowment did not limit its analysis to so-called literary reading. It selected studies that asked questions about “reading for fun” or “time spent reading for pleasure,” saying that this could refer to a range of reading materials.

“It’s no longer reasonable to debate whether the problem exists,” said Sunil Iyengar, director of research and analysis for the endowment. “Let’s not nitpick or wrangle over to what extent is reading in decline.”

In an interview Mr. Gioia said that the statistics could not explain why reading had declined, but he pointed to several commonly accepted culprits, including the proliferation of digital diversions on the Internet and other gadgets, and the failure of schools and colleges to develop a culture of daily reading habits. In addition, Mr. Gioia said, “we live in a society where the media does not recognize, celebrate or discuss reading, literature and authors.”

In seeking to detail the consequences of a decline in reading, the study showed that reading appeared to correlate with other academic achievement. In examining the average 2005 math scores of 12th graders who lived in homes with fewer than 10 books, an analysis of federal Education Department statistics found that those students scored much lower than those who lived in homes with more than 100 books. Although some of those results could be attributed to income gaps, Mr. Iyengar noted that students who lived in homes with more than 100 books but whose parents only completed high school scored higher on math tests than those students whose parents held college degrees (and were therefore likely to earn higher incomes) but who lived in homes with fewer than 10 books.

The new report also looked at data from the workplace, including a survey that showed nearly three-quarters of employers who were polled rated “reading comprehension” as “very important” for workers with two-year college degrees, and nearly 90 percent of employers said so for graduates of four-year colleges. Better reading skills were also correlated with higher income.

In an analysis of Education Department statistics looking at eight weekly income brackets, the data showed that 7 percent of full-time workers who scored at levels deemed “below basic” on reading tests earned $850 to $1,149 a week, the fourth-highest income bracket, while 20 percent of workers who had scored at reading levels deemed “proficient” earned such wages.

The new report is likely to provoke as much debate as the previous one. Stephen Krashen, a professor emeritus of education at the University of Southern California, said that based on his analysis of other data, reading was not on the decline. He added that the endowment appeared to be exaggerating the decline in reading scores and said that according to federal education statistics, the bulk of decreases in 12th-grade reading scores had occurred in the early 1990s, and that compared with 1994 average reading scores in 2005 were only one point lower.

Timothy Shanahan, past president of the International Reading Association and a professor of urban education and reading at the University of Illinois at Chicago, suggested that the endowment’s report was not nuanced enough. “I don’t disagree with the N.E.A.’s notion that reading is important, but I’m not as quick to discount the reading that I think young people are really doing,” he said, referring to reading on the Internet. He added, “I don’t think the solutions are as simple as a report like this might be encouraging folks to think they might be.”


Copyright 2007 The New York Times Company