Create/collect your dataset(s).
What to do:
- Create your data set using consistent and transparent naming conventions for both the data files and for the individual variables.
- As you capture/collect your data, keep written notes of everything that other researchers would need know to re-analyze the data at a later date.
Why do it:
- Maintaining control over the data collection process is vital to producing a clean, replicable dataset in the end.
- Describing the content of your datasets as you create them requires less time than going back and doing it retrospectively.
How to do it:
- Construct or re-use a template for data collection
- Develop variable labels that relate to the content. Here is some help with that.
- Create a separate document where you:
- describe the general content of your data files
- define the variable labels
- explain the differences between various iterations of your files
- document known issues
The Open Science Foundation offers a free platform for housing active research projects. Collaborate with your team members and store files in a shared digital space.
Things to consider:
- Are you creating or collecting data in formats suitable for sharing and long-term preservation? Best practice promotes the use of open file formats (for example, .csv, .doc, .pdf) for long term storage. Will you be able to save your active files in open formats when the time comes?
- What documentation and metadata should accompany the data? Consider what other researchers would need to replicate or reproduce your work.
- Is your data securely created, collected and stored? In terms of storage, consider using the 3-2-1 backup rule: 3 copies on 2 different media (e.g. hard drive vs cloud storage) with 1 backup copy offsite
