Is there a positive relationship between public library services and early reading success?

Storytimes, lap-sit programs, and other services for young children area major part of most public libraries’ missions. And, according to children’s librarians, these services play a significant role in preparing children for success as readers. But is that really true? Of course it’s true, most readers would respond. Just look at the well-used collections of books, audiobooks, and videos—all chosen to encourage the love of reading—that libraries provide. Or check out the abundance of kids’ programs, ranging from homework help centers to summer reading campaigns, offered at most libraries. But anecdotal evidence aside, can we actually associate these services with measurable, long-term, positive differences in the lives of children? Do we have real evidence about the impact of our work that we can share with parents, community groups, elected officials, and library funders? Based on a first-time look at new state data from across the country, the answer is definitely yes.

Illustration by James Steinberg

Looking at the data

State-by-state data released by the National Center for Education Statistics (NCES) in November 2007 provides evidence of a strong, positive link between the amount of children’s materials circulated by public libraries and fourth-grade reading scores on the same agency’s National Assessment of Educational Progress (NAEP). Since 1990, NCES has been collaborating with the 50 states and the District of Columbia to compile basic statistics about public libraries, including the circulation of children’s materials and attendance at children’s programs. (Beginning this year, national public library statistics will be compiled and released by the Institute of Museum and Library Services.) Since the 1970s, NCES has also conducted the NAEP, the only nationally representative and continuing assessment of what America’s students know and can do in various subject areas and at selected grade levels, including reading. Comparison of the reading scores with children’s circulation statistics shows a strong, positive link, while comparison of the same reading scores with attendance at children’s programs demonstrates a positive, if somewhat weaker, link. These findings support the position that, the greater the amount of circulated materials and the greater the attendance at library programs, the more likely kids will do well in reading. The correlation between reading scores and circulation of children’s materials is an impressive .514 (out of a maximum of 1.000) as well as being highly statistically significant at p < .01. What does this mean? It means that there is less than 1 chance in 100 that this relationship occurred by chance. The correlation between reading scores and attendance at children’s programs is a more moderate, but respectable .288. For you statisticians out there, this falls into the lower, but still significant, level of p < .05. In other words, there are fewer than five chances out of 100 that this was an accidental result.

Books out, learning in

Another way to get a handle on how much children’s circulation and reading scores line up is to look at which states fall in the higher and lower halves of all states on both statistics simultaneously. Colorado, Minnesota, and Ohio—the three states whose programs we highlight in the accompanying sidebar—ranked in the top quartiles (the top 25 percent) in both reading scores and children’s circulation per capita. Of states ranking in the top half of all states on reading scores, more than four-fifths (82 percent) ranked in the top half on circulation of children’s materials per capita (see Chart 1). Conversely, four out of five states (83 percent) in the bottom half on reading scores also rank in the bottom half on children’s circulation (see Chart 2).

The attendance factor

Similar to circulation of children’s materials, the relationship between children’s program attendance and test scores can be better understood by looking at both variables simultaneously. Of states ranking in the top half on reading scores, seven out of 10 (70 percent) ranked in the top half on attendance at children’s programs per capita (see Chart 3). By contrast, seven out of 10 states (71 percent) in the bottom half on reading scores also rank in the bottom half on children’s program attendance (see Chart 4).

The role of adult educational attainment

p>If children have well-educated adults in their lives, do public libraries impact their test scores? After all, better-educated adults—parents, other family members, and friends—probably exert many positive influences on children’s test scores. For starters, they encourage their children to borrow books and other materials from the public library. Plus, they tend to have more reading materials in their homes. Their children also have ready examples of readers, and these parents may take a greater interest in their children’s education. Taking these factors into consideration, is there any evidence that the link between children’s services in public libraries and reading scores is more than coincidental? In addition to the kind of correlation analysis previously discussed, we also employed a statistical procedure called partial correlation, which weighs the relationship between one variable and another (such as the number of years of education that adults have completed and children’s test scores) while “controlling for”—or removing—the impact of a third variable (in our case, the amount of children’s materials circulated). When adult educational attainment alone is correlated with reading scores, the correlation is .576, definitely a strong relationship in this context. That strong relationship is also highly statistically significant at the .01 level. But when the effect of children’s circulation per capita is removed, the correlation between the percentage of adults age 25 and up who graduated from high school and reading scores drops to a more moderate .376. In other words, when the impact of children’s circulation is removed from the relationship between adult educational attainment and children’s reading scores, the strength of that relationship is reduced by more than a third. This finding suggests that the impact of parents borrowing children’s books from public libraries accounts for a substantial portion of the impact of parents’ education on their children’s reading scores. Certainly, the factors affecting children’s reading scores are many and complex. But by using readily available data about reading scores, children’s services in public libraries, and adult educational attainment, this analysis supports the widespread belief that the efforts of public libraries to promote early literacy pays off in terms of higher reading scores during elementary school. There is a positive and statistically significant relationship between children’s services in public libraries and early reading success at school. In fact, there is even the probability of even stronger statistical evidence in the future, but first we need more ambitious research on the relationship between children’s services and early reading success. In the meantime, the evidence at hand suggests that the services of children’s librarians make a real difference in children’s lives.

---

Keith Curry Lance recently retired as the longtime director of the Library Research Service at the Colorado State Library. Robbie Bravman Marks is president of Marks Information, a consulting firm based in Denver, CO.