Strategies for teamwork

A growing number of projects involve multiple researchers working collaboratively. While teams offer higher productivity and a richer perspective, they also present a number of management challenges. NVivo has a number of tools to help you manage the practicalities of file distribution and collection: see Teamwork. But it also pays to consider strategies for dealing with the content of collaborative projects before you begin.

One challenge is balancing the need for all team members to code (and think) in a way that is consistent across the team, while allowing team members to bring their independent insights to the project.

Early in a project it is important to determine the approach your team will take to:

• Collecting and organizing data
• Creating and cataloging themes (the node structure)
• Coding the data

Whether you work with or using a project master file, you might want to consider the following:

• Appoint a team leader who will keep the team on track and make final coding decisions.
• Have regular team meetings to discuss interpretations, address issues and assign tasks—record the outcomes in a memo.
• Have each team member keep a memo to record their progress, including any hunches, suggestions or questions—you could also do this in a single ‘teamwork journal’. Memos
• Early on, have multiple team members code the same collection of files, then compare coding (using coding stripes or a coding comparison query)—this can help ensure a consistent approach. Coding comparison query
• To start with, make a node hierarchy for each team member.  After team discussion, you can refine, merge and reorganize. Reorder and organize nodes
• Aim for a clear node structure and use descriptions (in node properties) to make the purpose of a node clear for all team members.
• To help team members understand the meaning of nodes, create a codebook that lists the nodes and their descriptions. Export nodes
• As the project progresses, see which nodes have been created or modified and by which team member—do this in Code List View.
• While a common node structure is important for efficiency and reliability it should remain flexible so that new insights and exciting ideas are not lost.

Compare the coding of team members

If multiple researchers are coding the same data, you may be interested in the consistency of their coding. You can run a Coding Comparison query to check the 'inter-rater reliability' using the Kappa coefficient.

Where coding consistency is important, it may be useful to agree on a node structure early and have regular discussions about how the structure is evolving.