Master's Thesis: User Understanding of Dublin Core Metadata; or, What Exactly Do You Mean by Coverage?
My master's thesis study explored what metadata users found useful in a digital image collection, specifically focusing on Dublin Core metadata. In more detailed terms, this study looked if there are elements that users don't find useful (either because they do not need the information or because the labels are confusing); if users feel useful information is missing; and if the labels and elements that are provided are arranged in a way that makes sense to users.
The results showed that simple Dublin Core metadata did provide sufficient information to users: there were no cases where participants expressed an interest in different information than what was provided. In some cases, they wanted more of specific types of information that were already provided, such as the subject of the image. ("More tags, please!") Problematically, the Dublin Core vocabulary was, in some cases, not user-friendly. Although the terms are accessible to users, they can be misunderstood. For example, 'Source' was reliably misconstrued to mean the entity that published the image online or the website an item came from. Finally, my results indicated that undergraduate users are not blank slates: they have an idea of how searching is supposed to work (based largely on their experience of Google and other search engines) and when presented with something they don't understand, users fall back on what they are used to.