Metadata Quality in Institutional Repositories May be Improved by Addressing Staffing Issues
A Review of:
Moulaison Sandy, H., & Dykas, F. (2016). High-quality metadata and repository staffing: Perceptions of United States–based OpenDOAR participants. Cataloging & Classification Quarterly, 54(2), 101-116. http://dx.doi.org/10.1080/01639374.2015.1116480
Information Specialist, Cochrane Airways Group
St George’s, University of London
Tooting, London, United Kingdom
Received: 31 May 2016 Accepted: 15 July 2016
2016 Stovold. This is an Open Access article distributed under the terms of the Creative Commons‐Attribution‐Noncommercial‐Share Alike License 4.0 International (http://creativecommons.org/licenses/by-nc-sa/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one.
Objective – To investigate the quality of institutional
repository metadata, metadata practices, and identify barriers to quality.
Design – Survey questionnaire.
Setting – The OpenDOAR online registry of worldwide repositories.
Subjects – A random sample of 50 from 358 administrators of institutional repositories in the United States of America listed in the OpenDOAR registry.
Methods – The authors surveyed a random sample of administrators of American institutional repositories included in the OpenDOAR registry. The survey was distributed electronically. Recipients were asked to forward the email if they felt someone else was better suited to respond. There were questions about the demographics of the repository, the metadata creation environment, metadata quality, standards and practices, and obstacles to quality. Results were analyzed in Excel, and qualitative responses were coded by two researchers together.
Main results – There was a 42% (n=21) response rate to the section on metadata quality, a 40% (n=20) response rate to the metadata creation section, and 40% (n=20) to the section on obstacles to quality. The majority of respondents rated their metadata quality as average (65%, n=13) or above average (30%, n=5). No one rated the quality as high or poor, while 10% (n=2) rated the quality as below average. The survey found that the majority of descriptive metadata was created by professional (84%, n=16) or paraprofessional (53%, n=10) library staff. Professional staff were commonly involved in creating administrative metadata, reviewing the metadata, and selecting standards and documentation. Department heads and advisory committees were also involved in standards and documentation selection. The majority of repositories used locally established standards (61%, n=11). When asked about obstacles to metadata quality, the majority identified time and staff hours (85%, n=17) as a barrier, as well as repository software (60%, n=12). When the responses to questions about obstacles to quality were tabulated with the responses to quality rating, time limitations and staff hours came out as the top or joint-top answer, regardless of the quality rating. Finally, the authors present a sample of responses to the question on how metadata could be improved and these offer some solutions to staffing issues, the application of standards, and the repository system in use.
Conclusion – The authors conclude that staffing, standards, and systems are all concerns in providing quality metadata. However, they suggest that standards and software issues could be overcome if adequate numbers of qualified staff are in place.
In the first part of the article the authors reviewed the available literature to define what is meant by quality metadata and why it is important. They identified interoperability as being particularly significant, and discussed how the use of standards and best practices can facilitate interoperability, and therefore improve metadata quality. The literature review sets the scene very well for their investigation into repository metadata quality.
To evaluate the author’s survey, Boynton and Greenhalgh’s critical appraisal checklist for a questionnaire study (2004) was used as a guide. A survey questionnaire was an appropriate tool to address the stated aims of the research; however, the authors do not describe how they developed the survey, whether it was based on previously published or validated measures, or whether they piloted the survey first. The questionnaire is not included in the article or as an appendix so it is not possible to assess the suitability of the format or instructions, nor is it possible for readers to replicate the research in their own settings. The methods section is brief and does not give any detail about coding methods for the free-text questions.
The chosen sample repositories were all based in the United States of America, representing 14% of member American repositories and 1.7% of the total members. The authors do not explain their rationale for selecting this sample, nor discuss how the geographic limitation and small sample size may affect the applicability of their results. However, the analysis was appropriate, with results presented as both absolute numbers and percentages. The free text answers were interesting and added context to the quantitative data. Clear themes emerged around consistency of applying standards, and available time to do such work.
Despite some shortcomings in the reporting of their methods, the authors have provided a good overview of the issues relating to metadata quality. Their literature review picked up on staffing issues and this is reflected in the results of their survey. The authors found that even when professional staff are involved in selecting standards, internationally recognized standards are not always used in repository settings. There is a role for librarians to advocate for the use of recognized metadata standards where appropriate, and a challenge to develop metadata creation and curation skills so there are adequately skilled staff available to do this work. Cox, Verbaan, and Sen (2012) have conducted a useful audit of required competencies for librarians involved in research data management.
The authors sought to investigate the perceptions of metadata quality, while recognizing that the idea of ‘quality’ is subjective. Future research could be aimed at trying to quantify the notion of quality, for example by conducting an audit of adherence to recognized metadata standards in repositories.
Boynton, P. M., & Greenhalgh, T. (2004). Hands-on guide to questionnaire research: Selecting, designing, and developing your questionnaire. BMJ, 328, 1312-1315. http://dx.doi.org/10.1136/bmj.328.7451.1312
Cox, A., Verbaan, E., & Sen, B. (2012). Upskilling liaison librarians for research data management. Ariadne, 70. Retrieved from http://www.ariadne.ac.uk/issue70/cox-et-al