"The number of adults who read at least one novel, play or poem within the past year fell from 50% in 2008 to 47% in 2012, according to a new survey of over 37,000 Americans, “A Decade of Arts Engagement” by the National Endowment for the Arts. (Thirty years ago, that number was at 56 %.)"
I usually use the Pew survey to answer this question. They have a very different result.
"As of January 2014, some 76% of American adults ages 18 and older said that they read at least one book in the past year. Almost seven in ten adults (69%) read a book in print in the past 12 months, while 28% read an e-book, and 14% listened to an audiobook."
This is a really big difference and I wonder why. Expect updates.
ETA: In the secondary coverage, the NEA surveys and Pew surveys do not provide what counts for "reading a book" etc. When you drill down, the NEA disallows any reading you do for work or school. They are _only_ counting reading they you did other than for work or school. Pew doesn't care. Therein lies the difference.
If you quote NEA survey data as a measure of _reading_ by Americans, you are wrong. The NEA survey data is a measure of _reading other than for work or school_ by Americans. The Pew data _is_ a measure of _reading_ by Americans.
Pew data is sample size about 1000. NEA data is sample size about 37K, if I understood that right, and is based on questions in the SPPA, which ETA got that wrong! "In 2002, the SPPA began to be administered as a supplement to the U.S. Census Bureau’s Current Population Survey (CPS), the nation’s premier source of labor statistics. "
Sample size info (the 37K estimate was right): http://arts.gov/artistic-fields/research-analysis/arts-data-profiles/arts-data-profile-5/about-2012-survey-public