Studies in the “science of team science” are now identifying specific practices that encourage more effective and productive collaborative teams. LTER synthesis working groups benefit from NCEAS’ and the LTER Network’s experience with working group dynamics, as well as the literature of team science and open science for synthesis.
Building your team
Participation in a synthesis group can be a career altering opportunity. Everyone wants a team with good chemistry, where you know people will pull their weight and respect each other’s ideas. So it can be tempting to choose only people you know personally. We urge you to reach beyond your comfort zone to engage individuals with different skills and approaches, from different backgrounds. Draw on your extended network, keep a special eye out at conferences, and build partnerships with undergraduate or minority-serving institutions.
Try to be deliberate about adding knowledge and skills to complement those already on the team. Adding another community ecologist or biogeochemist (for example) may not add as much to the team’s capabilities as a biostatistician, a modeler, or a hydrologist.
Identifying appropriate data
It’s easy to make assumptions about what data might or should be available when you are in the throes of a good idea. By thinking carefully about what data will be needed early in the process, researchers can save themselves a lot of time and frustration down the road.
Eleven Quick Tips for Finding Research Data (Gregory et al. 2018) has a plethora of ideas for where and how to look, but the authors’ first tip is this: “think about the data you need and why you need them.” Remember to consider what kinds of data may be needed to normalize your primary data and be sure they are available.
Derived data products developed under this funding opportunity must be made publicly available, so be sure that contributors understand the requirement when they agree to participate — and remember to keep track of the sources, access dates, and any constraints associated with each data set.
The very quality that adds value to a synthesis group can also be its downfall. By including multiple perspectives on a problem, researchers increase the chances of arriving at a new solution. But it can take a surprising amount of time to identify and resolve conflicting understandings. Participants often reach a plateau (or even a valley) where they feel they are simply not comprehending one another — or the insights they are reaching are completely derivative.
Persist! That’s what’s known as the groan zone (a concept drawn from Sam Kaner’s book, Facilitator’s Guide to Participatory Decision-Making) and it is where the real intellectual work of synthesis happens.
Not all researchers are born facilitators, but it is a skill that can be learned. Kaner’s book is one good resource, as are these shorter articles and reports:
- How to Spark Joy in the Groan Zone. C. Kappel. NCEAS web site. 2019
- Cutting through the Complexity: A Roadmap for Effective Collaboration. 2018. Stanford Social Innovation Review.
- Collaboration and Productivity in Scientific Synthesis. S. Hampton and J. Parker. 2011. BioScience. doi: 10.1525/bio.2011.61.11.9
- Collaborations: Recipe for a Team. 2015. Nature. doi:10.1038/nj7559-245a
- Practices for Facilitating Interdisciplinary Synthetic Research: the National Socio-Environmental Synthesis Center (SESYNC). M.A. Palmer, J. G. Kramer, J. Boyd, D. Hawthorne. 2016. Current Opinion in Environmental Sustainability. doi: 10.1016/j.cosust.2016.01.002
- Ten Simple Rules to Enable Multi-site Collaborations through Data Sharing. M.R. Boland, K.J. Karczewski, N.P. Tatonett. 2017. PLoS Computational Biology. doi: 10.1371/journal.pcbi.1005278
- Enhancing the Effectiveness of Team Science. N.J. Cooke, M.L. Hilton, eds. 2015. National Academies Press.
Candidly discussing expectations for authorship early in the collaboration process can avoid conflict, hard feelings, and wasted energy later in the process. When individuals from diverse fields, sectors, and backgrounds come together, it is especially helpful to have a transparent discussion of what the group understands as an author-level contribution. Clarity regarding expectations for credit is especially important for creating and maintaining inclusivity. For an anecdotal case of how and why groups need to discuss authorship, see Equity in Author Order: A Feminist Laboratory’s Approach.
The CRediT framework has become the default for many journals and institutions in the past few years. This is a great starting point for determining authorship, though the literature is split into two camps that cite opposing failures with the framework. Some, such as Cooke et al., feel that authorship should expand to include a broader segment of those involved in producing work. Others feel that authorship expansion is out of hand. For a great overview of perceptions, Smith et al., survey many researchers and outline examples of how authorship views vary between people. Also, data-intensive research efforts such as synthesis may demand consideration of a wider variety of roles, as noted in this opinion piece from the Earth Science Information Partners (ESIP) Research artifact Citation Cluster.
The exact boundaries of what constitutes authorship will vary by the group and the type of paper, but they should be discussed as a group, early in the process, to minimize misunderstandings. The simplified authorship, data, and acknowledgements template linked here provides a solid starting point for discussion.
Working group meetings can be exciting, with in-depth discussion and rapid progress toward shared goals, but they will be most productive if the group has maintained momentum between meetings. Everyone doesn’t need to be involved in every decision, but working group leaders should establish a mechanism for regular group interaction between meetings. The best mechanism for your group is the one they will use. Some approaches that we have seen work well are:
- An email list. (LNO can set up a shared list with immediate or digest options.)
- Instant messaging on Slack and a running task board on Trello. Both have free plans for small groups and the LTER Network has a branded Slack account.
- A google website, with space for a shared calendar, uploaded documents, and group discussion. Simple to set up and maintain, this can also serve as a public presence for the group.
- For those who are reasonably fluent in it, GitHub has great versioning for code, markdown for private/public documents, shared wikis, and assignable task lists.
Collaboration during Covid-19
The Covid-19 pandemic has introduced new challenges to collaboration. “Maintaining momentum for collaborative working groups in a post-pandemic world,” authored by a number of researchers who represent synthesis groups, is an excellent resource specifically tailored to synthetic science.
Remote collaboration also introduces new opportunities for synthesis groups. Leveraging the liabilities of virtual collaboration, co-written by members of the LTER Network Office and NCEAS, suggests ways groups can capitalize on the remote environment.
There are many different resources that outline strategies for remote work in teams, but this resource is particularly well thought through.
Computing and analysis
NCEAS staff have deep experience in supporting and facilitating data-intensive collaborations. Shared server spaces, archived e-mail lists, wikis, Slack channels, Trello boards, and videoconferencing can all help keep collaborations moving between in-person meetings. Spend a little time as your team is assembling to discuss how your group likes to work and what tools they already know.
Using resources such as GitHub, R, Python and other scripting languages, savvy investigators can develop reproducible workflows that encourage effective documentation and allow users to quickly adapt analyses or rerun them with new or expanded data. The NCEAS/LTER computing staff is eager to help you find the technological solutions to your data and collaboration challenges. At least one member of each working group will be eligible to participate (at no cost) in week long training on Reproducible Research Techniques for Synthesis.
For a brief primer on the how and why groups should participate in open science and reproducible research, see The tao of open science for ecology and The History and Future of Data Citation in Practice.
Additional resources are also available through the NCEAS/DataONE Learning Hub, The Environmental Data Initiative‘s resource pages, Data Carpentry, and many other entities in the Earth and environmental sciences space.
Misunderstandings over authorship of large group products can be painful and time consuming. One of your first tasks as a working group should be to decide on a framework for determining authorship on publications. Is providing data a sufficient contribution to warrant authorship? Do all authors need to have read and commented on the manuscript? How and when should a potential author opt out?
When you do publish, please remember to acknowledge support provided through the LTER Network Office, grant DEB 1545288 and to alert the LTER Network Office, the NCEAS Communications Office, and the National Science Foundation Office of Public Affairs of your publication on acceptance.
The LTER Network Office and NCEAS also provide communications training and advice to broaden the impact and social utility of the synthesis research, and leverage connections through other partnerships such as the Science for Nature and People Partnership (SNAPP), the Science Policy Exchange, and COMPASS to engage audiences such as NGOs and public agencies in using synthesis results.