NSF Final Report #0085839, Collaborative Project 2000-2002 1 DeFelice NSF Grant #0085839, Collaborative Project: To Gather, Document, Filter and Assess the Broad and Deep Collection of the Digital Library for Earth System Education. Final Report, 2000-2002 Barbara J. DeFelice 1. Major research and educational activities The purpose of this collection assessment project is to develop and provide information about the nature and use of the DLESE collections to library builders and collection developers. This work is intended to serve as a model for collections evaluation and assessment for NSDL as a whole, and for other NSDL collections. Collections assessment is a systematic comparison of the actual collection with the desired or needed collection, as expressed through collection scope statements, collection development projects, user feedback, and user needs as expressed through searches in the collection. The goal is to provide information needed to support development of a highly relevant and useful digital library collection as a teaching and learning resource for a diverse user population. A collection geared to user’s needs is one of the many factors that contribute to a high quality, effective digital library, I devoted the first year and a half of the project to developing the prerequisites for a collection assessment for DLESE, a brand new digital library with just an early development test collection and a small group of early adopters as users and contributors. Although this kind of systematic assessment has been done for print collections, there were no models for applying this to a digital education library. I concentrated on educating the developing DLESE and NSDL communities about the concept and value of having an overview of the scope and balance of a community built collection of diverse objects. I introduced the idea of a collections assessment as a way to obtain this overview to better guide collection development at a time when calls for contributors to the collection were starting to be sent out. I discussed how this overview could be provided through a formal collections assessment procedure. I emphasized that it was necessary to have some way to describe the relative depth of the collection by different parameters to develop a collection that was not lacking in key subject areas, in material for all learning resource types, and in material that meets all grade level needs. It was also necessary to compare that collections depth to the desired collection as expressed by the collections scope statement, by searches and by input from users of DLESE and NSDL through meetings and forums. At the same time, I worked with the technical, metadata and library experts who were developing DLESE to create the technical structure needed to do this assessment. I determined what kinds of data and metadata record structures were needed to support this work. The DLESE metadata framework and first version of the search engine was just being developed, so what was possible kept changing. A result of this work were recommendations to the DLESE Program Center (DPC) technical staff about which metadata fields needed to be indexed so the data would be available for use in collections assessment. The DPC technical staff did this where they agreed that the information is important, where there is already metadata, and of course where the data structure allows