Research Data Management: Wā Whakamahere | Before

First, you will need to consider how you are going to acquire data, from where or whom, what kind of data you're going to be collecting, how much of it there will be, and what formats it will be in.

For example, let's say you are doing interviews which will be recorded. For this, you will likely need the recordings and they will be saved as audio files which could be .mp3 .mp4 .wav .wma etc. You may then decide that you'll need transcripts of the interviews so you can analyse the interviews and grab quotes or compare them for themes. Will these transcripts be .pdf .doc .rtf or some other format?

As another example, let's say you are using geospatial data, which could be .sdc . lyr .shp .e00 etc., file types. You could then export some maps or images into .jpg, .png, or other image formats. Perhaps your data will just be quantitative, perhaps it will be qualitative. Maybe spreadsheets, graphs, photographs, etc.

While you may not have all these answers yet, fill in what you do know, and add to it as you go. A data management plan can change over time, but it is always a useful roadmap.

Metadata, which is descriptive information about your data, can be a powerful tool to help you. Metadata could be:

• File formats
• File and folder naming conventions
• Folders and data structure
• Dates and times

It also includes information that other researchers will need if the data is shared later, such as:

• Decisions on what was included
• What was shared
• What was gathered
• When and where the data was gathered
• By whom it was gathered
• What controlled vocabulary was used

Some subjects and research fields have their own internationally recognised or recommended metadata schemes:

Ethical consent, risk management, and access and sharing are considerations throughout your research, and working through these issues at the start will help. There are a number of things related to data and ethics you should consider:

• Are there limitations on how, where or when you can process the data?
• Have you thought about the access, collection, storage and analysis of the data in relation to Data Sovereignty?
• Have you got processes in place to minimse impact if something goes wrong? How granular is the data and can it be used to identify people or places that participants would rather remain unidentified?
• Is anonymisation an option?
• Will you be wanting to make your data Open?

There might not always be a perfect solution when issues arise. In these situations, it is important to find a balance between the communities, people, or regions the data comes from - and those who will use it. If in any doubt at all, contact the university's Ethics Committees.

Research Ethics applications, and data management plans, are not static. They can be amended and adapted. But for Ethics applications, you will need to submit a supplementary application for ethical approval.