Metadata should be as comprehensive as possible, using standards and conventions of a discipline, and should be machine-readable. Metadata should always accompany a dataset, wherever it is stored.
Documenting metadata will make it easier for you and others to identify and reuse data correctly in the future.
Many academic displines have formalized specific metadata standards. You can check them out on the Digital Curation Center website.
Data Documentation Initiative (DDI)
This initiative is an international standard for describing the data produced in the social, behavioral, economic, and health sciences. DDI standards help data being documented, discovered and interoperable. Specifications and tools are found on the DDI website.
" Documentation, metadata, citation" Mantra online tutorial.