(Source: Google Images, May 4, 2020)
Key Takeaways From:
Starks & Trinidad. “Choose your Method: A Comparison of Phenomenology, Discourse Analysis, and Grounded Theory”
At the end of my Research Methods Class, I appreciate the comparative study that Starks and Trinidad on prostate cancer screening, where they use three research methods: phenomenology, discourse analysis, and grounded theory. They provided a useful hourglass metaphor (Figure 1) illustrating the similarities and differences among the three research methods. It appears that these research methods are different in the beginning and end but converge in the analytic and post-analytic stages, especially in the sharing methodologies for decontextualizing and recontextualizing data.
The researchers decided to use a comparative approach since they were undecided in terms of which research design to use. It is refreshing to read that other researchers struggle with this intellectual decision. I also struggled with my research design. Earlier in the course, I had decided on grounded theory, then decided against it since the coding aspect appeared overwhelming. Then I entertained the idea of a case study and phenomenology, but obtaining IRB approval during remote learning may cause further delays in the completion of my research proposal. So, I have decided on autoethnography, which is similar to phenomenology, in that both incorporate elements of a narrative in its research design.
Another exciting aspect of Starks and Trinidad’s research study is that they have three different audiences with three different goals: the novice researchers, researchers who are familiar with one research design but another, and teachers of research methods classes. In retrospect, I believe that they accomplished their goal of providing a framework for researchers and teachers in the three different research designs. In terms of their secondary purpose of deciding which research method to use during prostate cancer screening, I would recommend the Discourse Analysis: How the Discourses of Medicine and Public Health Construct Doctor-Patient Roles and Identities. By providing novice doctors a script when discussing prostate cancer screening with patients, the conversation will be more precise. The discourse analysis approach will produce more beneficence than the other two research methods. In a way, Starks and Trinidad had two purposes for their research study. Therefore, it is crucial to keep in mind the purpose of the research when deciding on a research design.
In my research proposal, I like the idea of multiple audiences and goals. My first audience is graduate and undergraduate students and the impact on online grammar checkers on their writing. The second audience is high-school English teachers or content-area teachers and writing instructors who are interested in helping students revise and edit their writing. My third audience is a decision-maker who may need the data to purchase the online grammar checkers for the entire institution.
In terms of Sampling, the researchers contend that it is not necessary to have a large sample of quality data. The average sample size is 1 to 10 participants. I had initially assumed that a larger sample size would produce quality results; however, it does not hold.
A key difference in the design is within the Analytic Methods, specifically under the Coding section, where phenomenologists are writing stories of people’s experiences, and discourse analysts are coding people’s language for patterns, themes, and roles. In contrast, phenomenologists use “a constant, comparative coding process” through three stages: 1.) open coding (“examining, comparing, conceptualizing, and categorizing data”); axial coding (“reassembling data into groupings based on relationships and patterns within and among the categories identified in the data”); and 3.) selective coding (“identifying and describing the central phenomenon”). Hence, a novice researcher may need guidance with the coding section of a phenomenological study.
Although the researcher may use self-reflection in the analytic stage, they must refrain from inserting their preconceived notions. In phenomenology and discourse analysis, researchers must “bracket” themselves where they set aside their assumptions so as not to bias the research study. In terms of products of a research study, there are surveys, explanatory theories, and stories.
I want to end with the most helpful takeaway, the Methodology (Formulating a Research Question) to help me formulatemy research questions:
- What is your experience with online grammar checkers (Phenomenology)?
- What discourses are used, and how do they shape identities, relationships, and activities? (Do the copy on Grammarly such as Awesome! motivate writers?) (Discourse Analysis)
- How does the social process occur in a particular environment? (Grounded Theory)
The last question did not necessarily fit with the purpose of my research proposal, so I can rule out grounded theory (which I did). I would say that this exercise is helping in deciding the research design. Overall, this study is valuable to researchers and had wished that I had read this earlier in the semester.