Subject analysis theories and their application to geographic subject metadata for electronic theses and dissertations in South African University libraries

Abstract

The study investigated the way in which theories of subject analysis facilitate the creation of effective geographic subject metadata for ETDs. The subject analysis practices of metadata creators, who contribute geographic subject metadata to ETD repositories in South African university libraries, and the underlying theoretical aspects were explored. Subject metadata is highly significant in facilitating the discovery of the subject content of ETDs. This study focused on geographic subject metadata as a type of metadata that helps to search for information related to a specific geographic locality. Geographic subject metadata facilitates the discovery of information resources and, in this way, it is vital in allowing a different form of subject access – i.e. apart from the common, topical approach. This type of metadata is useful in bringing together all information on a particular locality. A mixed methods research methodology was employed in the study. The study used the explanatory sequential mixed method research design, with three data collection methods. A survey questionnaire and interviews were used to investigate the experiences and perspectives of the subject metadata creators in South African university libraries. Content analysis was conducted to examine the ETD metadata records for the extent and nature of the use of geographic subject metadata. For the first, quantitative part of the study, a survey questionnaire was used, in order to include more respondents in the collection of primary data. A qualitative study was conducted by means of interviews, with a sample of metadata creators, so as to allow deep probing into the approaches followed by the metadata creators during subject analysis for the purpose of subject metadata creation. Quantitative/qualitative content analysis was employed to study the ETD records. The Statistical Package for the Social Sciences (SPSS) was used for the quantitative data analysis. Basic descriptive statistics was used to summarise and present this data meaningfully, so as to make the underlying information easily comprehensible. NVivo10 software assisted in organising and exploiting the qualitative research data. Thematic analysis facilitated detailed explanations of the data obtained from the participants’ responses to subject analysis approaches, metadata creation procedures and other general contextual influences. For the content analysis phase, Excel spreadsheets were used to perform analysis functions; to manipulate the extracted ETD records; and to perform the data analysis. The findings of the quantitative survey, qualitative interviews and the content analysis revealed a variation in the existing metadata practices for theses and dissertations and the approaches followed in subject analysis and ETD geographic subject metadata creation. Both the quantitative and the qualitative findings signified inconsistent practices and varied perceptions of the inadequacies of existing ETD geographic subject metadata. The records analysis confirmed the inconsistencies in the current approaches to ETD geographic subject metadata creation. The qualitative interview findings provided in-depth explanations that demonstrated that subject analysis theory is generally considered as important to improve efficiency. However, the level of knowledge of the theoretical basis was generally found to vary, while not always being explicit, which suggested that subject metadata creators do not apply the theory adequately. The findings revealed a variation in the existing metadata practices for theses and dissertations and the approaches followed in subject analysis and ETD geographic subject metadata creation and the level of knowledge on the theoretical impact. Increasing knowledge and awareness of the importance and impact of theory will help to improve ETD subject metadata creation and the discoverability of ETD content. Based on these research findings, recommendations were made for increasing knowledge and awareness of the importance and impact of theory, which will help to fill the existing knowledge gaps impeding on optimal ETD subject metadata creation and the discoverability of the ETD content. Appropriate metadata analysis practices – based on appropriate theoretical principles – will help to improve the discoverability of ETD content to the broadest user population possible. Additionally, the research findings provided information on the support policies and guidelines that support collaboration among different role players and the coexistence of geographic subject metadata created from hybrid sources. The theoretical approaches in the literature and the practical implications revealed in the research findings are recommended for wider implementation in related contexts, in order to guide subject analysis and the process of creating ETD geographic subject metadata.