On-line serials collection analysis

A design for an on-line serials decision-making and collection analysis system is proposed. It is composed of four basic components: citation data, conventional serial records data, utility/cost ratio compilation and journal ranking techniques, and user interface software. The system would have the ability to respond specifically to user interest profiles and to integrate locally generated data. It is postulated that such a system is capable of satisfactorily resolving the major criticisms of the use of citation data for selection purposes: that libraries are diverse in their interests and that no aggregate list can be more than generally relevant; that the inclusion of cost data is essential and that citation ranking without regard to cost can be misleading; and that other relevant data should be considered as well.