Research Data Management: Overview

1. always keep original data
2. back up regularly (automate this if at all possible)
3. document your data thoroughly (metadata, data dictionary)
4. name and organize files according to a schema
5. use version control
6. secure the data appropriately
7. cite any secondary data you use
8. consider your long-term plan
   1. What will you keep, for how long, where, and who will pay for it?
   2. What kinds of reuse or sharing will be allowed? In what timeframe?

Research data is loosely defined as information collected, observed, or created for purposes of analysis to produce original research.

This includes observational variables like rainfall, wind speed, water quality, or survey data; simulated data from earthquake models; experimental data from lab instruments; and derived or compiled data for text mining or testing algorithms. Research data can take almost any digital file format (video, text, photographs, numbers), so managing it effectively can be a challenge.
 

If you've found this page, it's likely that you manage some form of research data, either your own, your lab's, or your advisor's. Many researchers are not taught data management skills in their graduate courses. This guide and accompanying workshops aim to help fill this gap with the expertise that librarians and data specialists can offer.

Good data management:

• ensures integrity of data
• ensures that data is findable and usable when grad students leave projects over the years
• makes the data of a project readily understandable to people outside the project
• enables the sharing of data within and across disciplines
• makes it easier to archive and preserve data in the long term
• encourages data citation to increase the impact of the research