Understanding journal usage: A statistical analysis of citation and use

This study examined the relationship between print journal use, online journal use, and online journal discovery tools with local journal citations. Local use measures were collected from 1997 to 2004, and negative binomial regression models were designed to test the effect that local use, online availability, and access enhancements have on citation behaviors of academic research authors. Models are proposed and tested to determine whether multiple locally recorded usage measures can predict citations and if locally controlled access enhancements influence citation. The regression results indicated that print journal use was a significant predictor of local journal citations prior to the adoption of online journals. Publisher-provided and locally recorded online journal use measures were also significant predictors of local citations. Online availability of a journal was found to significantly increase local citations, and, for some disciplines, a new access tool like an OpenURL resolver significantly impacts citations and publisher-provided journal usage measures.