Responsible Conduct of STEM Research

Why manage data?

• to preserve the integrity of the research
• to allow reuse by others
• to reduce risk of data loss

Why make data discoverable?

• to enable work to be reproduced
• to establish credibility and to hold trust
• to enable faster progress in research, within or across disciplines
• to meet requirements of funders, journal publishers, etc.

Why reuse data?

• to verify research claims
• to permit new discoveries from exisitng data
• to foster integration of data sets for new analysis
• to reduce duplication of effort

A successful Data Management Plan (DMP) answers these questions:

Creating your data

• What types of data will be produced for your project?
• What identifiers will you use for your data?
• How will you document your data?
• How much data will the project produce?
• How often will the data change or be updated, and will versions need to be tracked?

Organizing your data

• What file formats will be produced for your project and what kinds of data management risks do they present?
• How will you organize your files into directories and what naming conventions will you apply to both?
• Have you included project and data documentation?

Managing your data

• Who is responsible for managing and controlling the data?
• For what or whom are the data intended?
• How long must the data be retained?
• How secure are the data? Do you have a procedure for backing up the data?

Sharing your data

• Does project funding require your data to be shared or publicly accessible?
• When and where do you intend to publish or distribute your data?
• How should your data be cited?
• Are there issues with privacy or intellectual property?

Ownership of data

• Who owns the data? (typically, the institution)
• What may participants do (or not do) with project data? (See UF Intellectual Property Policy, p. 10)

When should you NOT share data?

• To protect ongoing research from premature disclosure
• To protect priority in publication
• To protect private information about research subjects, trade secrets, or classified government information

When may you exclude data?

• Human/technical error vs. outliers
• Cleaning/editing vs. falsifying
• Note excluded data in discussion

Supervisor's responsibilities

• Review data-keeping regularly
• Watch for inconsistent notation, misinterpreted instructions, malfunctioning equipment

Why keep a lab notebook?

• To have a complete record of the work you have done and to maintain your rights to your findings.  Keep all your procedures, data, and comments in one place.
• To prove that you did the work, and on what date -- especially if your work is novel, article-worthy, or potentially patentable.
• To leave a trail for someone who may be interested in completing your work.
• To serve multiple audiences: yourself, your colleagues, your funding sources.

Best practices:

• Follow the cultures and practices of your lab.
• Keep a separate notebook for each project.
• Maintain a Table of Contents (TOC) in each notebook.
• Record procedures, data, and ideas immediately (each day).
• Maintaining a laboratory notebook "includes the basics, but also some tricks" such as recommended pens.
• Check Good Laboratory Practice (GLP) from agencies or disciplines such as the World Health Organization, the Environmental Protection Agency, the Food & Drug Administration.
• Lab Notebook Tips from a Patent Litigator offer suggestions if your work is potentially patentable.
• Examples of Electronic Laboratory Notebooks from Univ of Utah

from NYU Health Sciences Libraries, "a mini series showing a data management horror story."

Examples of data sharing guidelines:

Looking for other researchers' data?  Try these data repositories and registries: