No strings, arrays or conditions required but they can be helpful.
The goal of qualitative research is to find the user’s story. Through the process of coding, we uncover the common themes from the data points. These also inform what research needs to be done next. It also helps in process improvements. Are you getting good samples? Do you have the right sample set? Are the participants diverse enough or too similar? Does the current information align with previous research results?
Research coding is the act of segmenting, categorizing and aligning information collected during interviews, observation, articles or any other data collected. You do it over multiple passes, building deep connections across your data. You create codes from feelings, beliefs, drivers, blockers, and sequence of tasks or events as codes; whatever seems relevant. There’s no right or wrong way to label your data with codes, but experience helps. You determine what is important, and making the call becomes more accurate over time. The clearer you make your codes the more thorough and objective your findings become.
Documenting the methodology helps consumers of the research follow the logic of the process because let’s face it, you will be challenged, questioned and doubted. Stakeholders aren’t trying to be mean; it’s just the nature of the game. It might also have to do with research often being viewed as a blocker to fast iterations of design and development. It could also be that the investigation’s role isn’t understood in the development process and that all research has its limitations and flaws, but it’s better than nothing. Well, almost nothing, bad research can send a product down the wrong path that it might not recover from. Wait… how did I get down this rabbit trail; let’s get back to coding.
Documented methodologies set the context and objectives of the research. A consumer of the research should be able to see the evolution of multiple passes of the data and theories derived from it. To accomplish this create a system of visual cues to illustrate patterns in the data.
At a high level, the process of coding includes:
• Reading Text
• Tag portions of text with a word or phrase [codes]
• Categorize the myriad of codes
• Organize categories into Basic, Organizing and Global themes
• Develop theories [cause > effect]
• Repeat the process a few times, continually modifying the theories
• Make and define connections [converge and diverge]
• Is there a hierarchy of categories?
• Is one group is more important than others?
• Illustrate the results
Don’t lose your way by veering away from the data before you. Let it speak to your unbiased mind. But as a researcher, you get to decide what is relevant and what isn’t. Look for the patterns, and surprises as a guide.