The Outsiders

Make no mistake. One of the unintended consequences of the “65 percent solution” is that it provides another basis for assaults on school libraries. On firstclasseducation.org—and, more significantly, in state legislatures across the nation—Overstock.com’s CEO Patrick M. Byrne and his supporters are lobbying that public schools be required to devote at least 65 percent of their overall spending for “in-classroom”—versus “out-of-classroom”—purposes. Guess what? Libraries are deemed an “out-of-classroom” expense. Which is why this “solution” would damage school libraries and lower students’ test scores.

This proposal from First Class Education (FCE), a national advocacy group, has two major problems. One is the ambivalence about whether librarians are instructional or support staff—FCE relies on the outmoded definitions of “in-classroom” and “out-of-classroom” from the U.S. Department of Education’s National Center for Education Statistics (NCES). The other difficulty is that there is at least as much evidence (indeed, more) for the positive impact on test scores of school libraries as there is for the value of instructional—or “in-classroom”—spending.

From FCE’s rhetoric, you would never guess that libraries are under attack. “We think libraries and librarians are important and have a direct impact on classroom instruction,” reads a message on the group’s Web site. “The NCES classifies librarians and libraries as outside the classroom and we would urge NCES to reconsider this classification. In states where First Class Education is pushing ballot measures and we have controlled the specific language, we have added libraries and librarians as an inside the classroom expense.”

So far so good. But FCE’s site goes on to state that “[i]n states where legislative enactment is being considered, First Class Education suggests keeping the NCES definition in tact [sic]—otherwise opening the definition for Legislative amendment risks watering down the classroom instruction definition to the point of being meaningless. If a school district has reached its maximum efficiency and is still below 65 percent, First Class Education would encourage that a waiver be granted to such a district, as our proposal provides, rather than cutting libraries and librarians.”

So what FCE is saying is that it’s OK for them to classify libraries and librarians as “in-classroom expenditures,” but nobody else—unless FCE deems that a district has “reached its maximum efficiency.” That term isn’t defined.

Inconsistencies at the federal level confuse things even more. NCES uses different definitions of instructional and support services (that is, in-class and out-of-class) in its Schools and Staff Survey (SASS) and Common Core of Data (CCD)—the most significant survey of schools nationwide—and other major data collections and the No Child Left Behind (NCLB) Act have their own conflicting notions of these basic terms.

In NCES’ defense, it’s not the source of the assumption that in-classroom expenditures are good and out-of-classroom expenditures are bad. In fact, NCES does not use the terms “in classroom” and “out of classroom.” FCE substituted them for NCES’ original language, “instructional” and “support services,” respectively.

NCES does not endorse the “65 percent” criterion. In fact, NCES challenges the idea that instruction happens only in the classroom. According to the 2003 edition of “Financial Accounting for Local and State School Systems,” instruction encompasses all “activities dealing directly with the interaction between teachers and students. Teaching may be provided for students in a school classroom [or] in another location….”

Yet the CCD identifies instructional staff as regular and part-time teachers, teachers’ aides, homebound teachers, hospital-based teachers, substitute teachers, teachers on sabbatical leave, classroom assistants, and even clerks and graders. But the CCD omits principals, full-time department heads, supervisors of instruction, teaching-school nurses, and librarians.

What’s the logic in acknowledging that instruction isn’t limited to the classroom, but excluding teaching nurses and librarians? In the case of librarians, many states are adopting the term “teacher-librarian” to recognize that their librarians are required to be certified as, or similarly to, teachers.

Nevertheless, the CCD classifies librarians as support staff along with administrators, counselors, speech pathologists, bookkeepers, bus drivers, food service staff, and others. Essential positions, yes, but not as directly involved with developing curriculum or delivering instruction as librarians, department heads, and supervisors of instruction. After all, quality instruction does not spring forth fully formed in the classroom.

Still on the fence? Consider what is included in NCES’ definition of “instruction”: school-sponsored programs such as cocurricular activities and athletic programs that supplement regular instruction and enhance student motivation, enjoyment, and skill improvement. The CCD provides examples of activities, events, and groups that fit that definition: band, chorus, and speech, but also senior proms and the Future Farmers of America. Nothing against the prom, but counting it as instruction adds insult to injury.

NCES may resist efforts to revise its CCD definitions, arguing (admittedly, this is a rumor) that such changes would disrupt its year-by-year trend data. But if the whole subcategory—“Instructional Staff Support Services”—is moved from “Support Services” to “Instruction,” data from previous years could be arrayed similarly. To refuse to do that would be a case of the tail wagging the dog.

After the CCD, the Schools and Staffing Survey (SASS) is probably NCES’ most ambitious ongoing survey of schools. The latest SASS report—“Characteristics of Schools, Districts, Teachers, Principals, and School Libraries in the United States, 2003-04”—endorses the concept of school librarians as teachers. The report describes their roles as “library and media curriculum development and instruction, library media resource person for classroom teaching staff, and informational specialist and program advisor for all library and media related classes and curriculum.” And in the SASS questionnaires, a library media specialist is defined as a “paid professional staff [member] who is certified by the state… in the library media specialty area.”

This acknowledgment of librarians as educators is repeated in the report when “library media specialist/librarian” is included in the listing of “school teachers by main activity”—along with “regular full- and part-time teachers, itinerant teachers, long-term substitutes, administrators, and other professional staff.”

The No Child Left Behind Act also challenges the narrow interpretation of instruction, stating that “[t]he term 'instructional staff’… includes principals, teachers, supervisors of instruction, librarians, library school media specialists [sic], teachers of academic subjects other than reading, and other individuals who have responsibility for assisting children to learn to read.” Clearly, NCLB’s definition is at odds with the CCD’s. Presumably, the former should prevail, since it represents the federal government’s current education policy.

In answering the question “Will the 65 percent solution improve student achievement?,” FCE points to a correlation between the percentage of classroom spending and student test scores. That finding, which is attributed to the Independence Institute, a public policy research organization based in Golden, CO, cites that “...the percentage of dollars reaching the classroom has five times greater correlation (49% correlation) [sic] with increased test scores than simply spending more money (10% correlation) [sic].”

Because school library expenditures are not itemized in the CCD survey, it is impossible to measure the impact of those expenditures when combined with instructional expenditures. But, drawing on the CCD and SASS data, it is possible to measure the relative impact of instructional and library expenditures per student on test scores. The table below presents the relationship between those expenditures and reading scores on the SAT and the National Assessment of Educational Progress (NAEP), better known as “The Nation’s Report Card,” administered to fourth and eighth graders every two years.

At each grade level, library expenditures per student correlate with test scores as strongly as, and usually more strongly than, instructional expenditures per student. In fact, average SAT scores of high school seniors decline as instructional expenditures per student rise. What this means is that library expenditures are a better predictor of how well students will do on tests than the amount spent on in-classroom instruction. Furthermore, these results suggest that there is a strong link between well-supported libraries and increased student achievement.

Another argument in favor of the “65 percent solution” is based on a comparison of classroom spending percentages for the highest- and lowest-scoring states. As the tests are not identified, that sort of comparison cannot be duplicated; but it is possible to compare the average test scores of the highest- and lowest-spending states based on instructional and library expenditures per student.

As seen in the table below, library expenditures per student “outperform” instructional expenditures per student in their impact on test scores. (The 10 percent decrease in average SAT scores from the five lowest- to highest-spending states on instruction confirms the negative correlation between instructional expenditures per student and SAT scores.)

These facts make as compelling an argument for higher library spending as for higher instructional spending. But for school libraries, at least, the evidence is not limited to a mere correlation reported by a single organization. Since 2000, seven teams of researchers in no less than 15 states and Ontario have produced a substantial body of evidence that shows the impact of school libraries on boosting students’ test scores. And 11 of those studies have controlled for factors such as community socioeconomic status, student demographics, and adult educational attainment, as well as school spending, teacher-pupil ratio, and teacher education and experience. The research findings are clear: the impact of school libraries and librarians on student learning is a consistently positive one, and it cannot be explained away by those other factors.

After hearing from school librarians and their advocates, perhaps NCES is coming around to this way of thinking. For the first time, in its July 2006 report, “Current Expenditures for Public Elementary and Secondary Education, School Year 2003–04,” a single figure was reported for each state under the designation “instruction and instruction-related.” If that change in nomenclature is signaling a shift in NCES’ problematic definitions, state legislators who are unsure how to vote on the “65 percent solution” should take note. On the other hand, while a single combined figure has been reported, the document’s glossary still defines “instruction” and “instruction-related” separately.

It’s time for our political leaders, state officials, education proponents, and the voting public to exercise their critical thinking skills—the sort that school librarians teach students every day. They might start by examining the often conflicting information presented by FCE and NCES and the many studies that confirm the value of school libraries. The “65 percent solution” should not be enacted by our state legislatures without taking steps to ensure that it does not harm school libraries—and, ultimately, student achievement.