This system will be undergoing maintenance Tuesday, May 5, 2015 from 10:00 AM to 11:00 AM CDT.

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

variation in staffing levels to produce statistically significant results when staffing levels were
part of the overall library variable. In Missouri, for example, Miller et al. (2003) found no
significant correlation between staffing levels and test scores in their regression analysis, but
found that a combination of all LMC services accounted for over ten % of the variance in test
scores. In any case, the overall library program variable in most of these studies accounted for
anywhere between 1 and 21 % of the variance in student test scores, with five of the seven
reporting changes in R2 of 7.3% or higher. As these analyses generally account for the major
school and community variables, they provide perhaps the strongest evidence of a link between
student achievement and library programs. See Table 15 for a summary of these multiple
regression analyses.

75

Achterman, Douglas L. Haves, Halves, and Have-Nots: School Libraries and Student Achievement in California. Denton, Texas. UNT Digital Library. http://digital.library.unt.edu/ark:/67531/metadc9800/. Accessed May 4, 2015.