Haves, Halves, and Have-Nots: School Libraries and Student Achievement in California Page: 180

                                
curriculum-integrated information literacy instruction, and total services, all at r =.19, p <.001.
Just one of the fourteen services did not remain significant in relationship to English Language
Arts CST scores when controlling for school and community variables in partial correlations.
Total services remained significant in all partial correlations, the strongest when controlling for
ethnicity at r =.23,p <.001.
Slightly stronger bivariate correlations between social studies CST scores and library
services were seen, the strongest being total services at r =.24, p <.001, and informally
instructing students in the use of resources and providing reference assistance to students and
teachers, both at r =.22, p <.001. All but two of sixteen services remained statistically
significant in partial correlations with each school and community variable as a control, and
library services remained significant in these partial correlations with r values ranging from .20
to .28,p <.001.
At the high school level, in 21 of 22 categories there was a statistically significant
relationship between library services and student achievement. The relationship between
English Language Arts CST scores and library services was very similar in strength to that of
U.S. History CST scores, which will be summarized here. The strongest bivariate correlations
included total services at r =.51, p <.001, providing teachers with information about new
resources, r = .49, p <.001, and informally instructed students in the use of resources, r = .47, p
<.001. Partial correlations between total services and U.S. History CST scores remained
significant when controlling for each school and community variable, ranging in strength from
r = .46, p < .001, controlling for average teacher salary and r = .53, p < .001 when controlling
for percentage of English learners.

180