Metadata Cheat Sheet

Sitting down to write a complete metadata record can be overwhelming. While the feature-rich metadata editor integrated into the Research Workspace is a helpful tool, the sheer number of arcanely-named fields can make writing metadata a daunting task. This page is designed to provide a manageable amount of information to get you started writing metadata, including recommendations on how to work metadata into your project as you go and the minimum amount of information needed for your record.

For more detailed information, please consult our full Metadata Best Practices page.

And if you read through the information below and still find yourself needing help, please email us at metadata@axiomdatascience.com.

Make Metadata Part of the Process

When it comes to metadata, the single most important best practice is to not leave metadata to the last minute. Documenting how data were collected, processed, stored, and analyzed after the fact is both tedious and difficult. Instead, we recommend making metadata creation part of your research process. It’s much easier to document as you go, creating an “in progress” metadata record and describing each stage of your project as it is completed.

The table below breaks down some common fields of a metadata record based on which (very general) stage of your project they can be completed during. Consider setting aside small chunks of time as your project progresses to work on metadata. Writing a little at a time is much easier than trying to do it all at once.

Project Onset	Data Collection	Project Wrap-up
Title Abstract Purpose Contacts Category and Form Keywords Taxonomic Information	Spatial Bounds Time Period(s) Data Table Attributes(e.g., column headers)	Lineage Statement Data Consistency Report Process Steps Completeness Report Status and Maintenance Constraints Metadata Info

Minimum Required Content

The table below includes a list of the most important fields in most metadata records, along with brief descriptions of what should be included in that field. The name of each field is linked to the relevant section of our full Metadata Best Practices, which includes a more in-depth guidance and examples.

Note

Filling out only the fields below does not necessarily mean that your metadata record is complete. Each dataset is unique and will have unique metadata needs. The table below is intended as a guide for getting started. Depending on the nature of your dataset, your funding requirements, and how your data will be archived, more fields may be required.

Section	Field	Minimum Required Content
Resource Overview	Title	Include location, data type, time period, and name of author, program, or institution. Do not use special characters. Keep title to fewer than 85 characters.
Resource Overview	Abstract	Describe the dataset rather than describing your project as a whole. Expand on the elements of your title with one sentence for each: location, data type, and time period. Include names of all data files and describe any relationships (e.g., database tables saved as separate files).
Resource Overview	Purpose	Include 1-2 sentences about the larger signifigance of your project and why it was conducted.
Resource Overview	Contacts	Provide contact info for the people who managed the project, collected the data, and/or generated the dataset. Use roles (e.g., Point of Contact, Principal Investigator, Originator) to indicate each person’s relationship to the project. Check the Workspace’s contact search for existing contact information.
Keywords	Keywords	Try to include at least 3 keywords from NASA’s Global Change Master Directory (GCDM) under Thematic Keywords. Add more specific keywords like geographic locations or species under Arbitrary Keywords.
Taxonomic Information	Taxonomic Information	Include one entry from the Integrated Taxonomic Information System (ITIS) for each taxa studied. Only include Observation Processes if they are particularly relevant to your project (e.g., results based on correctly identifying speices of aquatic instect).
Spatial and Temporal Extent	Spatial Bounds	Provide bounding box coordinates that encompass the geographic area of the project. Don’t worry about include granular geospatial data (e.g., survey tracks, buoy locations) in your metadata (those should be included in the dataset itself).
Spatial and Temporal Extent	Time Period(s)	Provide relevant date ranges at the most appropriate scale for your project.
Resource Content	Data Table Attributes	Provide the names, definitions, and units of the attributes of any data in tabular format (e.g., column headers in a CSV file). When possible, use the attribute tool in the Workspace’s metadata editor.
Methods	Lineage Statement	Provide a summary of the methods employed in your project (especially established methods in your field that can be cited). Include citations to standard operating procedures, field manuals, or other reference publications whenever possible.
Methods	Process Steps	Include step-by-step instructions on how your raw data were prepared for analysis. Include citations (or links) to any specific tools used. For open source projects, include a link to the software repository if possible.
Methods	Data Quality Reports	Describe anything you did to ensure the completeness and accuracy of your dataset (e.g., instrument calibrations, automated procedures, and manual/visual tests for outliers). Describe any data that were excluded from your dataset and note why.
Status and Distribution	Constraints	If your funding source or program has standard constraints language, include it here.