Best Practices for Collaborating on Research


Across the disciplines, collaboration and interdisciplinary work is growing. As teamwork in scholarship increases, it’s important to establish good practices for collaboration. Laying clear ground rules (Howard Gadlin and Kevin Jessar call these a “prenuptial agreement for scientists”) and having an open discussion about expectations helps the collaboration run smoothly. This is equally true if you are collaborating with different labs at other universities, working with private industry, or keeping your mentor up to date on your own research. Sharon Ann Holgate has a few tips for successful collaboration that we’ve expanded on:


  1. Address mutual expectations. Every member of a team may have different expectations about how each person will contribute and how they’ll be credited. By discussing these expectations openly, it’s easier for each team member to contribute to the project effectively. Gadlin and Jessar recommend keeping a written record because it’s “most important that collaborators commence their project by anticipating, discussing, and resolving possible areas of disagreement.”

  2. Clearly divide and establish who’s responsible for each task. Along the same lines as addressing expectations, a clear division of labor makes each team member’s role in the project clear. This will facilitate conversations about authorship.

  3. Determine authorship. In a collaboration, it may appear that each person has a clear role. This assumption, however, can lead to confusion and disagreement over first authorship. Agree on authorship at the beginning of the project. If the project changes or takes a new direction, authorship may change, too. As soon as this happens, all parties re-negotiate authorship (with the consensus preferably in writing). When it comes time to publish the research, disagreements over authorship are far less likely.

  4. Communicate frequently. Keep open lines of communication with the team. If you don’t have a clear timeline or clear research goals, it becomes easy to fall out of touch. If you haven’t heard from a contributor in months, don’t assume that no news is good news, according to Joan Schwartz of the National Institutes of Health. Keep in touch about deadlines and the project’s overall status to move your research to completion.  

    The same advice holds true for working with your advisor. Keep your advisor up-to-date on your research. If you’re out of touch for months on end, there’s a chance that your advisor has different expectations about how much progress you’ve made or just what you’re working on.

  5. Take minutes of meetings and then distribute to everyone involved in the research. Send an email to everyone after phone conversations and face-to-face meetings. This provides documentation that can be referred to in later conversations. If you forget what was covered or if there’s a disagreement about what was decided in the meeting, the minutes help resolve those issues.

  6. Access to data. Not all parties may have access to all data (some projects pool all data among those involved; others share everything, reports Francis Macrina in Dynamic Issues in Scientific Integrity). A clear conversation at the beginning of the project is necessary to establish who will have access to what information. If the project’s direction changes or the project grows, revisit the question of who has access to which data.

  7. Discuss the expectations for the data with all researchers before research begins. How will the research be communicated (presentations, publications) and how soon will it be shared after the research is conducted?

    When research collaborations span academe and industry, there are different standards for sharing data, so this discussion is particularly important. An academic may be eager to publish findings for the tenure dossier, while the researcher in industry wants a patent to be filed on the idea first (if data is already published, it can’t be patented).

    Graduate students can run into a similar issue, not just with publications, but also with their dissertations. If you use data from industry in your dissertation or if you collaborate on collecting data with an industrial partner, you may be asked to embargo your dissertation with ProQuest (all UNL doctoral candidates are required to file their dissertations with ProQuest). This means that no one can access your dissertation digitally and you cannot publish the data in another forum.

  8. Collaboration on research also means a shared responsibility for integrity in research integrity. Verification of data across labs may not be feasible, especially when each lab has technical abilities to acquire part of the data that the other labs cannot. When there’s an error in the data or some portion of the data is compromised, it’s vital that failure to comply with research regulations be shared with all parties involved—researchers, institutions, and funding agencies (Macrina). For this to be possible, all researchers need to fully disclose funding sources at the beginning of the collaboration. If new funding lines open up or another institution becomes involved in the research, all involved in the research should be notified.

These guidelines for collaboration provide only the basics to help you get started. For further guidelines and specific questions you can use to guide your own collaboration, see the NIH Guidelines for the Conduct of Research .



Gadlin, Howard and Kevin Jessar. (2001) “Preempting Discord: Prenuptial Agreements for Scientists.” In: ORI: The Office of Research Integrity.

Holgate, Sharon A. (2012) “How to Collaborate.” In: Science Careers.

Macrina, Francis L. (1995). “Dynamic Issues in Scientific Integrity: Collaborative Research.” American Society for Microbiology.

Reis, Richard M.  (2001) “Avoiding Misconduct in Your Scientific Research.” The Chronicle of Higher Education.

Schwartz, Joan P. (2011). “Silence is not Golden: Making Collaborations Work.” The Office of Research Integrity