As more artifacts are created in a digital format, there is a need to have metadata associated with the artifacts to increase the chance for resource discovery by others. This is especially the case with non-textual artifacts. Once these artifacts have descriptive metadata associated with them, they have the potential to become learning objects which can be used by others in their own teaching and research. This study explored the design and development of a performance support tool to create descriptive metadata by users that are most familiar with the non-textual learning objects, yet may not have an understanding of the various metadata schemas and standards required by other institutional/knowledge repositories and search engines. In order to create such a tool, certain features need to be included in order for users to create appropriate metadata. The tool needs to have Unicode character support in order metadata entry, display and searching. Research found that characteristics such as controlled vocabularies, tooltips, validation rules, and having a relevant image on the same screen as the metadata form help users to create appropriate and accurate metadata; yet, no existing tool was found that contained all of these features to assist faculty in describing their non-textual learning objects. These characteristics were operationalized in the design and development of the performance support tool. Findings from the evaluation of the tool indicate that the owner of the learning objects was able to create a customized, non-standard metadata form that users were then able to use to create appropriate and accurate descriptive metadata. / Ph. D.