8 Metadata
Data repositories such as Dryad and ICPSR are designed to permanently store the data that are used in research so it is available to future scholars. To use the data, it is important to have good metadata: metadata describe the data themselves, rather than why the study was conducted, hypotheses, results, or conclusions. Metadata include thorough description of the data file, the context in which the data were collected, the measurements that were made, the tools used to gather the data, and the quality of the data themselves.
We want data and metadata to adhere to the FAIR Principles for Scientific Data Management and Stewardship, meaning they should be:
- Findable: persistent identifiers & basic machine-actionable metadata
- Accessible: can be read by machines & humans
- Interoperable: use shared vocabularies / ontologies, machine-accessible (meta)data
- Re-Usable: descriptions allow others to reuse and link with other data sources
The metadata for the types of data used in our work should also (typically) include the 5 Classes of metadata descriptors from Michener et al. 1997
- Class I: alert potential secondary users to the existence of data sets that fall within specific temporal, spatial, and thematic domains.
- Class II: describe the research resulting in a data set.
- Class III: status of the data set and information on its accessibility.
- Class IV: all the attributes related to the structure of the data file.
- Class V: all other related information that might be necessary for secondary usage, publishing the data set, or auditing the data set.
8.1 What are Metdatadata?
E. Bruna Video for LAS6292: Why Metadata?
ICPSR: What is a Codebook?
8.2 Best practices for preparing metadata
“Best Practices in Creating Social Science Metadata.” p.32 in the ICPSR Guide to Social Science Data Preparation and Archiving: Best Practice Throughout the Data Life Cycle (6th Edition).
Michener, W.K., et al . 1997. Non-geospatial metadata for the ecological sciences. Ecological Applications 7: 330–342.
[read online]Pp 446-450 in Bernard, H.R. and Bernard, H.R., 2013. Social research methods: Qualitative and quantitative approaches. Sage.
DataONE Community Engagement & Outreach Working Group (2017) “Metadata Management”. Accessed through the Data Management Skillbuilding Hub on Aug 31, 2020
8.3 Discipline-specific metadata standards
It is worth looking in these catalogs to see if you can find metadata standards for your discipline. They will provide suggestions on not only what to include, but the standard vocabulary for your discipline.
UK Digital Curation Center Directory of metadata standards for different disciplines
Ecological data: Ecological Metadata Language
Museum Specimens: Darwin Core
Geography Markup Language (GML): Emphasis on geographic features (roads, highways, bridges)
Humanities: UF Digital Collections (UFDC) key metadata fields used for non-published items such as posters, archival materials, artists’ files, field notebooks, etc. Includes a link to a template you can download. See also the different Metadata Preparation Guides provided by the University of Florida Digital Collections (UFDC), the UF Humanities Archives Metadata Template, and the metadata required by the Qualitative Data Repository.
7, Linguistics: Lameta is a tool to help with organising collections of files, aimed primarily at collections made in the course of documenting language, music, and other cultural expressions.
8.4 Metadata templates
I have created metadata templates based on information from ICSPR (for social sciences) and Michener et al. 1997 (for biophysical sciences) that can be downloaded and edited; you can add more fields or delete any that are not relevant. Note that Table 1 in Michener et al. is much more comprehensive and provides additional guidance on how to make sure the metadata are useful. The templates are available in .txt, .Rmd, and .qmd format).
Metadata Template for Social Sciences based on ICSPR standards: .txt format or .Rmd format.
Metadata template for Biophysical Sciences based on Table 1 from Michener et al. 1997: .txt format or .Rmd format
Metadata for the Humanities or those working primarily with Qualitative Data: The metadata required often depend on the type of material with which you work (e.g., oral history, photos, digital, printed). If your data is in this domain, you can use this general template from the UF Humanities Archives: Template #2. You can also review the metadata required by the Qualitative Data Repository.
8.5 Tools for creating machine-readable metadata
I include these here in case you want to try using them. It’s not required, but it could definitely make your life easier if there is a standard tool for your discipline (e.g., MORPHO if you are working with ecological data).
giant list from the RDA of tools for creating standardized metadata for different disciplines
Coastal and Marine Science Data Repository - GRIIDC: Select the Dataset Information Form Guidance from the Training and User Guides page.
8.6 Metadata Dictionaries
Global Change Master Dictionary [Note: deprecated June 2020]
8.7 Organizations developing metadata standards and schema
The Research Data Alliance (RDA)“has the goal goal of building the social and technical infrastructure to enable open sharing and re-use of data.”
DDI Alliance: “Established in 2003, the Data Documentation Initiative Alliance (DDI Alliance) is an international collaboration dedicated to establishing metadata standards and semantic products for describing social science data, data covering human activity, and other data based on observational methods.”
The Dublin Core Metadata Initiative is “an organization supporting innovation in metadata design and best practices across the metadata ecology”.