Documenting your data is a way of making them more explicable to others and to yourself long after you would likely otherwise have forgotten the specifics of what each dataset contains. Doing so will allow others to interact with your data in many ways, whether to understand or verify your findings, review your publication, replicate your results, design a similar study, or archive your data for access and re-use.
How should I document my data?
According to The University of Edinburgh data documentation can take place in:
- Laboratory notebooks and experimental protocols
- Questionnaires, codebooks, data dictionaries
- Software syntax and output files
- Information about equipment settings and instrument calibration
- Database schema
- Methodology reports
- Provenance information about sources of derived data
The documentation needs to take place at the project level, the file or database level, and the variable or item level. The project level should document the goals of the study, how it contributes new knowledge to the field, the research questions, methodology, materials, and anything else that a user would need to understand in order to make best use of the data for their purposes. The file or dataset level should make clear how all the files relate to each other, their format, and should account for all the files in the project. Finally, at the smallest level a variable should be named, but also have its meaning and how it was operationalized explained.