Research

Introduction to research data management

Planning your data management

You need to write a data management plan for your project. If you are applying to a funding council or charity for a grant, they may ask you to supply one as part of your application, otherwise you need to write one when your project starts.

A data management plan explains how you will gather, organise, store and document your data during your project, and how you will ensure those who need to can access your data once your project has ended. There are templates to help you write your plan:

These templates, including those from funders, are also available within DMPonline, a web-based tool you can use to write your plan.

Gathering data

If your research involves a study with human participants, you need to obtain informed consent from them when they agree to take part. As well as the terms of the study itself, your consent form should explicitly mention the following:

  • the data you will collect;
  • how you will use them;
  • which data you will retain after the conclusion of the study;
  • how they will be shared.

You must not share personal data openly that you have collected; you should ask for consent to do one or both of the following:

  • share the personal data only with other researchers, or those who have signed a non-disclosure agreement;
  • openly share an anonymised version of the data.

We provide guidance and advice on informed consent and restricting access to data:

Re-using third party data

Before you collect new research data, you should check whether you could derive your intended research results from existing data. There are archives all over the world that provide access to data that other researchers have shared; some are dedicated to particular types of data and others are more general.

If you do find data you would like to use, this is fine so long as you follow the terms of the associated licence, and give due credit to the originator. The most common way of doing this is to cite the dataset you used in any resulting publications.

Working with data

Storing your data safely

You should store your data on the University X Drive unless you have a compelling reason otherwise. Computing Services back up the X Drive daily and mirror it across multiple sites to prevent loss. You can restrict access to your data to certain individuals within the University, and share files with external collaborators through a secure web interface.

Storing and transferring data securely

If you are working with sensitive data, you need to take extra precautions to ensure only those with permission can access them. Sensitive data include data that can be linked to individuals, commercially sensitive data, data shared in confidence, and data relating to vulnerable groups.

You can achieve a certain degree of security by keeping data in restricted folders on the X Drive, and not copying data to your desktop, laptop or removable media. If you need more security, or more flexibility, than this, you should consider using encryption. You can encrypt files and folders on your local and network drives, or you can encrypt entire disks or devices. You can also encrypt email attachments, whole emails, and files transmitted over the Web. Encryption depends on using a strong password; it is vital that you keep this password safe.

Organising your data

You should organise your data files into a sensible hierarchy of folders:

  • At the top level, use folders that reflect your protocols for the data; for example, who can access them, or whether the files should be read-only or periodically deleted.
  • Below that, use folders that reflect how you would search for data; for example, by task, run number or data type.

You should name your files according to an agreed and documented convention. It could include codes that refer to the date of creation, run or sample number, data type, person responsible, or version. If you include a date, use the ISO 8601 format: 2009-12-31.

Documenting your data

You need to document your data properly, otherwise neither you nor anyone else will be able to understand them in future.

You should start by finding out what information you will be expected to provide at the end of the research process:

  • Your research community may have minimum reporting standards or other metadata standards
  • If you know to what archive you will submit your data, find out what information they will collect from you. For example, the University of Bath Research Data Archive has a minimum metadata standard.

You should then consider what you would want to know, if the data had been collected by someone else and they were explaining them to you.

This information will be most useful if you record it in a structured, standard way such as in a DDI file. If there are no suitable standards for your data, you can write it down in a file called 'README'; be sure to structure this file with appropriate headings to make it easier to navigate.

Archiving and sharing data

Writing a data access statement

You must include in your research publications a data access statement that informs readers how to access the underlying data. You can fulfil this requirement by adding a statement similar to the following to your acknowledgements section:

"All data supporting this paper are openly available from the University of Bath data archive at http://doi.org/10.15125/12345."

It is not sufficient to instruct the reader to contact you for access. Your data should be available from an archive, which will be able to mediate access to the data for you. You should also try to get a DOI or another robust identifier for your dataset; this helps others find your data, and helps you to track the impact of your data.

Getting a DOI for your dataset

There are two main ways of getting a DOI for your dataset, to use in a data access statement.

  1. Submit your dataset to an archive that specialises in your kind of data; for example, if you are funded by the ESRC you should offer your data to the UK Data Service in the first instance.
  2. Upload your dataset to the University of Bath Research Data Archive.

Either way, you need to register your dataset with us, then fill out a record for the Research Data Archive. If you intend to upload your dataset to our archive, we can predict the DOI it will have for you. Otherwise, we will link our record to the archive that holds the dataset; you will need to contact that other archive to find out what the DOI will be.

Complying with requirements

You may find the following resources useful when checking which data management requirements apply to you: