Data Management Planning

A data management plan (DPM) describes how data will be collected and documented, as well as how it will be analyzed, stored, and disseminated, and possibly how it may be preserved and shared with others. Creating a DMP is considered a best practice and can help you organize your research process, fulfill funding requirements, and respond to administrative needs, such as ethics approval.

In an effort to assist with this task, UNB Libraries recommends that researchers take advantage of the Portage Network's DMP Assistant as well as their Training Resources page. The DMP Assistant guides researchers through building a data management plan using a template that asks questions about data collection, documents and metadata, storage and preservation, sharing and reuse, responsibility and resources, and ethics and legal compliance. The training resources available include DMP Exemplars as well as other materials related to developing individual capacity and institutional strategies. 

Another great resource for planning is the University of Toronto Libraries' DPM Checklist, which can be used in conjunction with the DMP Assistant.

Documentation and Best Practices

Data documentation, or metadata, helps you understand your data and will be useful for your future research, publishing, and presenting. Furthermore, by describing details of data collection, processing, naming conventions, file formats, etc., it also helps other researchers find, use, and properly cite your data. Planning your data documentation and organizational structure prior to collection can save you a lot of time and stress when you are going through your files mid-way and post-collection.

Documenting Your Work

At the study or project level:

• Why was the data created? Provide a context.
• Who was involved (creators, collaborators, funders, others)?
• When was the data gathered or analyzed?
• How was the data gathered and/or generated (including software)? What was the methodology? What instruments or measurements were used?

At the file or database level:

• How were the files named? Are there any abbreviations that need explaining?
• What file formats are used?
• What information is contained within the file? What is the relationship between different files?

At the variable level:

• How are the variables named? What do they mean? What units of measure are used? Is there any other information needed to understand your data?

For more information, see the University of Guelph's Documenting Your Work page.

File Formats

Ideally you will choose file formats that have the following characteristics

• non-proprietary
• in common usage
• standard representation (ASCII, Unicode)

Examples

• Open document format (ODF) - not Word
• ASCII - not Excel
• TIFF or JPEG2000 - not GIF or JPG

For more information, see MIT Libraries' File Format for Long-term Access page.

Naming data files

Choose a filename system that is clear, meaningful, and consistent, and relevant to the content

• Use capitals or underscores rather than spaces or periods
• Use date formatting allows files to display chronologically (e.g., YYYY-MM-DD
• Include version numbers when applicable (use leading zeros for ease of sorting)
• Avoid using special characters, such as () [] {}, <>, !, etc.

For more information, see Stanford University Libraries' Best Practices for Filing Naming page.

Document adapted from the following: MIT Libraries - Data Management, Portage Network - How to Manage Your Data, University of Guelph - Library Data Management Planning, University of Toronto Libraries - Data Management Plans, Western Libraries - Research Data Management.