Consensus coding is a method used in qualitative data analysis, applied after an initial codebook is generated. It involves multiple researchers independently coding the same transcripts and then comparing the results on a one-to-one basis. This method helps balance the inherent subjectivity of qualitative coding by reducing variability in how codes are applied or perceived, thus increasing the trustworthiness of the analysis.
A researcher might use consensus coding to ensure rigorous, systematic, and defensible coding definitions. It provides a transparent audit trail to explain coding decisions, contributing to overall quality assurance. However, it's worth noting that this method is more time-consuming due to the high level of rigor involved.
Please see below a step-by-step instructions on how to do consensus coding in Delve.
Consensus Coding - Researchers Code the Same Transcripts
Step 1: Create a project.
Create a new project by clicking the arrow next to the project name, and then clicking "New Project"
Step 2: Share the project
Share the project with the people you will be working with. See how.
Important: Make sure that everyone is using the same project. Do not make separate projects.
Step 3: Upload your transcripts to your project.
Upload transcripts to your project. See how.
Step 4: Create codes
Create codes and add code descriptions to make sure everyone knows the correct way to apply a code. The codes within the project are shared across everyone.
Step 5. Code the transcripts. Use "Coded by Me" to hide how everyone else coded.
Have each person code the same transcript. If you don't want to see how others coded, use the Coded By Me interface which will hide all codings except your own.
Step 6: Compare how you coded using "Coding Comparison"
Now that everyone has coded the same transcripts, you can go to Delve’s Coding Comparison page to compare how you coded.
Step 7: Use memos to discuss coding decisions
Discuss your coding decisions using memos. Click here to learn how.
Step 8: Summarize findings in the code description.
Write down your findings and thoughts in code descriptions. See how.
Qualitative analysis is an iterative process, and researchers will often repeat steps 4 - 8 until they have analyzed all their transcripts and aligned on their analysis.