Predicting Book Circulation by Subject in a University Library |
| |
| Abstract: | The purpose of this paper was to identify and measure variables in the academic environment which best predict (estimate) circulation of monographs by academic subjects. Fall 1974 circulation at the University of Southwestern Louisiana was the dependent variable. Academic department was the unit of analysis. Subjects were defined as sixty-three academic departments including one miscellaneous category. The subject of each circulated book was categorized in one of the sixty-three. Twelve independent variables were identified: the hard/soft, pure/applied, and life/nonlife characteristics; credit hours, lower-level course enrollments, upper-level course enrollments, master's-level enrollments, Ph.D. -level enrollments, shelflist, Ph.D. program, master's program, and bachelor's program. Three linear regression hypotheses were formulated: a full twelve-variable hypothesis, an eleven-variable hypothesis excluding shelflist, and a three-variable subject-characteristic hypothesis. For the full 12-variable hypothesis, only shelflist, master's-level enrollments, hard/soft, and upper-level enrollments, in that order, were significant, accounting for 76% of the variance. This was called the shelflist model. For the eleven-variable hypothesis without shelflist, only credit hours, hard/soft, and upper-level enrollments, in that order, were significant, accounting for 60% of the variance. This was called the nonshelflist model. For the three-variable subject-characteristic hypothesis, onyl pure/applied and hard/soft, in that oder, were significant, accounting for 26% of the variance. Regression equations for estimating circulation were formulated. Estimates are given using the shelflist model. |
| |
| Keywords: | |
|
|
