Journal Packages: Another Look at Predicting Use

Abstract

This study uses linear regression analysis to determine if interlibrary loan (ILL) requests and/or denials are predictors of subsequent usage of journals acquired in a package subscription. In 2016, Auburn University Libraries acquired the SAGE Premier package and gained access to 572 previously unsubscribed titles. The study used interlibrary loan request data and denial data from 2015 to determine if these two metrics were predictors of subsequent use for the unsubscribed titles from this package. A strong correlation was found between ILL requests and later usage but almost no correlation between denials and later usage.

Keywords:

• COUNTER
• denials
• ILL
• journal use
• prediction of use
• turnaways

Notes

¹ Drott and Griffith (1976) and McGrath (1976), in letters to the editor, examined the paper’s methodology and rationale. They questioned the original paper’s findings and identified several mathematical errors.

² Readers may be more familiar with the t-test, which is used to determine if there is a significant difference between the means of two groups. Linear regression uses F-tests, which determine if the variances of two groups are equal. For more information about F tests see Surbhi (2017).

³ The prediction equation includes the mathematical constant e because the data were transformed with the natural logarithm; e is the base of the natural logarithm. Most calculators include an e^x function; the formula = EXP can be used in Excel. For more information about using the e function in Excel, see Virginia Tech (2014).