The Data Desert
RSSA potential source of school library statistics has dried up
Renee Olson, Editor-in-Chief -- School Library Journal, 05/01/1999
The 10-year-long National Library Power Program has now been prodded, poked, and probed to reveal its secrets for success. With all that scrutiny, you'd expect some statistics to turn up -- hard numbers that might have proved that good library media centers improve learning. But the official evaluation, while valuable, was surprisingly barren of statistics.
The fact that the DeWitt Wallace-Reader's Digest Fund, the program's sponsor, chose not to do a statistical study is disappointing due to the size of Library Power -- it involved some 700 schools in 19 states and nearly 400,000 students. A study based on it would have made Keith Lance's well-known Colorado study, which looked at 221 schools in one state, seem as small as an undernourished marmot.
But that wasn't part of the plan. When the Fund organized Library Power's evaluation in 1993, it held meetings where "everything you can imagine was discussed," says the Fund's evaluation officer, Adam Stoll. But six years ago, he says, the push for quantifiable results wasn't as central to education as it is now.
That makes no sense. There's a long history of statistical research on the impact of school libraries. In the early 1960s, Mary Gaver researched school libraries' impact on reading for the U.S. Department of Education. And in 1993, Lance published The Impact of School Library Media Centers on Academic Achievement (Hi-Willow), aka the Colorado study.
I had expected Stoll to pin the lack of statistics on the fact that comparing reading ability across states is nearly impossible, since there is no one test that all American children must take. For a multi-state study, Library Power would have had to install itself in schools whose students are tested by the National Assessment of Educational Progress (NAEP). This test, which involves some 100,000 students from 2,000 schools, provides the only national sample we have of reading ability.
This data void gives researchers little with which to work. "The lack of a common [nationwide] test, or at least comparable tests, is a big stumbling block," says Lance. "You can't mush test scores together."
Still, the Fund could at least have studied Library Power and academic achievement within selected states, using NAEP or statewide reading tests. That brings us back to Colorado, where Lance is now at work updating -- and expanding -- his 1993 study. "We're using many more questions on what library media specialists do and how these activities impact student achievement," explains Lance. The study's working title reflects that: "How School Librarians Help Kids Achieve Standards." He expects to release it about a year from now.
Until then, you can enlighten your colleagues and administrators by sharing the findings of the evaluation that Drs. Dianne McAfee Hopkins and Douglas L. Zweizig did for the Fund. In "Power to the Media Center (and to the People)", you'll find an essential seven-point checklist for successful school libraries.
But that's not enough when it comes to visionless administrators and taxpayers who are roused only when statistics are paraded in front of them. While the Fund clearly dropped the ball here, future large-scale studies on libraries and learning will be thwarted until our country legislates a national reading test.
Unless that barrier is removed, the hope of firm statistical support for improving school libraries is little more than a mirage.
Renée Olson
Editor-in-Chief
rolson@slj.cahners.com
