The Potential of Conjoint Analysis for Measuring Value in Collection Development

The potential for using new conjoint analysis software in measuring value in the collection development process is examined. A small pilot study in a federal library is described that tested whether or not a conjoint analysis program could successfully predict what CD-ROM titles would be selected by a group of librarians. Results indicate the program can predict which titles are preferred by a specific group, based on criteria the group considers important for making selections. It is suggested, then, that new conjoint analysis programs may be useful in measuring what library customers value when selecting from among competing titles. Further analysis using customers and a larger, more random sample is recommended.