Lai Yee, the co-founder of Delve, recorded a seminar about Grounded Theory. Tonight I watched the 32 minute session titled “Intro to Grounded Theory Analysis with Delve.” In addition to creating software for qualitative data analysis, they create free resources including webinars, free courses, and infographics. Lai Yee noted that this is one approach and they focused on when to use grounded theory. The webinar took a “practical” approach. Grounded Theory was defined as “a qualitative research method that enables you to derive new theories based on the iterative collection and analysis of real world data.” In grounded theory, you first collect and analyze data, and then you derive theory. Grounded theory is a cyclical process where data collection and analysis happen cyclically, noted Lai Yee. Grounded theory is used when there is no existing theory to explain a phenomenon. The benefits of grounded theory are that the findings represent real world settings, are connected to the data, and allow for new discoveries. Limitations of grounded theory include difficulty in recruiting participants, time consuming to collect data, and challenges in analysis (since it occurs in a rolling basis). Lai Yee explained that data collection and analysis are cyclical and raw data such as transcripts is used to derive the theory. The data collection process in grounded theory is called theoretical sampling: you recruit, for example, some people, begin to transcribe, and then code. The first step in analysis is open coding to turn group excerpts into codes. You then continue doing data collection. You group codes together with axial coding and refine the codes into categories. Selective coding is used to take a core category that becomes the basis for your theory. Lai Yee shared an example project with the research question: how did high schoolers in New York state adapt to learning during the pandemic? Interviews are transcribed and brought into the Delve tool in this demo. Open coding is the process of breaking transcripts into snippets. Lai Yee starts by selecting snippets and coding them as uncategorized. Next, the constant comparative method groups codes. In open coding, snippets are organized by comparing them. In the example, Lai Yee grouped quotes/snippets into the code “Loneliness.” Lai Yee then described the use of analytical memos as a way to write notes and document reflections and analyses (explain contradictions and thoughts about your findings). Lai Yee defined theoretical saturation as the point at which additional data does not teach you more about your topic. Additional recruits may be needed. Recruiting parents of the high schoolers may be the next step. Delve has a participant information tab that I haven’t seen! Roles can be assigned: students or parents, in this example. Transcripts can then be assigned to students or parents. Again, Lai Yee categorized snipped by performing selective coding. Snippets can then be filtered by the code uncategorized and the role parent. Lai Yee organized and refined the codes by making comparisons and connections. Memos were also added to document ideas. Axial coding is then used to compare codes with codes and create categories or axes to connect them. Delve allowed Lai Yee to easily view the “code book” and decide what other connections can be made. Selective coding is when you compare categories with categories to create a core category. However, Lai Yee noted that not everyone thinks there should be a core category. After multiple rounds of data collection (theoretical sampling) and data analysis through selective, axial, and open coding, it is time to summarize to represent your central idea or theory! Codes and memos are turned into the narrative. You can export codes as a document to write your manuscript to “tell the story of your data.”
