Content area
The purpose was to identify organizational variables which could predict the adoption of technological innovations in college libraries and to determine which, if any, form a common pattern in the prediction of adoption of innovations. The hypothesis was that the adoption of automation as a technological innovation in a college library could be statistically predicted by using readily available measures of size and finances of the college library.
Eight library automation innovations were examined: (1) acquisitions systems, (2) circulation systems, (3) computer output microform catalogs (COMCAT), (4) micrcomputer use, (5) Online Computer Library Center (OCLC) or other bibliographic utility membership, (6) online searching, (7) online public access catalogs (OPAC), and (8) serials systems. Each innovation was separately examined to identify which of the following organizational size and finance variables would predict its adoption: (1) college student enrollment, (2) number of periodical subscriptions, (3) number of volumes in the library, (4) periodicals budget, (5) salary budget, (6) total library budget, and (7) total staff in the library.
Data were from statistics published in the American Library Directory and the Higher Education General Information Survey (HEGIS). The population was 1,000 four-year college libraries. The analysis was done using the BMDP stepwise discriminant analysis program with cross validation (BMDP7M).
Predictors were found for seven of the eight innovations. No predictor was found for the COMCAT innovation. Predictors found were: Periodical budget for acquisitions systems, total budget and number of book volumes for circulation systems, salary budget and number of periodical titles for microcomputer use, total staff for OCLC, total budget and salary budget for online searching, salary budget and number of periodical titles for OPACs, and number of book volumes for serials systems.
Percentage of correct prediction ranged from 59.9% to 95.2% in the derivation samples and from 64.6% to 92.4% in the cross validation samples. The percentage of variance accounted for ranged from 2.52% to 39.50%.
The conclusion was that size and finance variable could be used to predict adoption of technological innovations. No patterns were detected among the significant predictors for the seven innovations for which predictors were found.