Quality Analysis of Metadata Records in the NSDL Metadata Repository

Kent State University, Kent OH

Investigators

Abstract

The National Science Digital Library (NSDL; http://nsdl.org) is a growing digital library of heterogeneous resource collections, organized in support of science, technology, engineering, and mathematics education at all levels. A key component of the NSDL is its Metadata Repository. The metadata in the NSDL Metadata Repository enables users to discover, explore, and retrieve resources and to assess whether the resources are relevant or suitable for particular educational objectives. The NSDL Metadata Repository takes a new approach to metadata that supports the reuse of metadata, encourages open sources, and simplifies the process of metadata creation. The NSDL Metadata Repository is composed of metadata records created according to heterogeneous metadata standards and submitted by different voluntarily contributing units. For the NSDL to maintain a central role in science education, it is imperative that metadata in the NSDL Metadata Repository be of the highest possible quality. This Targeted Research project is performing a quality assessment and analysis of metadata records in the NSDL Metadata Repository. The project's objectives are (1) to assess the metadata quality of the NSDL Metadata Repository through a systematic quality analysis of individual records and record sets, as well as the quality of selected metadata records from contributing collections; (2) to classify the quality problems evident in the collection, providing evidence of their origins and recommending quality assurance solutions; (3) to explicate the relationships between metadata quality and the design principles that underlie the metadata services of the NSDL Metadata Repository; and (4) to provide an empirical basis for an NSDL metadata quality assurance program. This study is the first to address issues of quality assessment, quality assurance, and quality control using the NSDL Metadata Repository. The results of the research will be useful to the NSDL Core Integration team and to developers of NSDL collections and services. Quality issues need to be assessed and addressed at the current early stage of the NSDL Metadata Repository so that quality assurance systems can be developed and implemented before the collection reaches a point at which rectification of quality problems becomes difficult or impossible.