Background
The Emory University Libraries, Emory Center for Digital Scholarship and other Emory units have developed numerous digital collections over the past decade. Accompanying metadata originates via multiple business units, authoring tools and schemas, and is delivered to varied destination platforms. Seeking a more uniform metadata strategy, the Metadata Working Group initiated a project in 2014 to define a set of core, discovery-focused, schema-agnostic metadata elements supporting local content types.
Process
In collaboration with Emory's Office of Business Practice Improvement, quantitative and qualitative techniques were utilized to mitigate these complex organizational factors. A key research deliverable emerged from benchmarking: a structured comparison of over 30 element sets, recording for each standard its descriptive element names, their required-ness, and general semantic concepts.
Additional structured data collection methodologies included a diagnostic task activity, in which participants with varying metadata expertise created (simple) Dublin Core records for selected digital content. A survey of stakeholders provided greater context for local practices. Multiple public-facing discovery system interfaces were inventoried to log search, browse, filter, and sort options, and available web analytics were reviewed for user activity patterns correlating to these options.
Thematic analysis was performed on all benchmarking, system profiles, and web analytics data to map the results to a common set of conceptual themes, facilitating quantification and analysis. A weighted scoring model enabled the ranking of elements’ themes: the highest scoring concepts then explicated as an initial set of core elements, mapped to relevant standards and schemas.