Week 11 Research Journal

 Carter et al: "The Use of Triangulation in Qualitative Research"

Triangulation: the use of multiple methods or data sources to understand what is being studied; tests validity through convergence

Method triangulation: multiple methods within the same study (field notes, observations, and interviews on same topic, for example)

Investigator triangulation: Multiple researchers for multiple observations and conclusions

Theory triangulation: Analyzing and interpreting data with different theories.

Data source triangulation: Collecting data from different types of people to gain multiple perspectives and validate data

A good deal of this article focused on comparing in-depth interviews and focus groups to demonstrate that while these seem like very different approaches, they can be used to triangulate data (along with things like observations) to create a successful study.

Methods are usually selected based on the best fit for the research question.

Pros and Cons of Combining IDI and FG:

• Increased participation
• Increased validity of findings (triangulation and member checking)
• How are IDI and FG data analyzed together?
• Weighting of data? 

 Maurya: "Rigor in Qualitative Research"

Member Checking (respondent validation): validate interpretations/findings by consulting with original study participants; to do this you must create your initial analysis of the data, share your findings (key themes, quotes, and conclusions) with the participants, gather their feedback, and revise your interpretations. Enhances trustworthiness by treating participants ethically.

Triangulation: examining a phenomenon from multiple perspectives. This enhances accuracy, provides broader perspectives, and increases credibility. 

Audit Trail: keeping track of all research activities and decisions for the study. All steps are documented, including designing the research, collecting data, and analyzing the data. This allows for transparency and accountability.

Peer Debriefing: engaging in discussions with a peer or expert throughout the research process to ensure ethical consistency, avoid bias, enhance trustworthiness, and generally examine your work every step of the way for any major inconsistencies or flaws.

Rich and Thick Description: presenting detailed, vivid, and nuanced data that captures the essence of the phenomenon being studied. Interview skills are critical, as well as transcription. This helps researchers situate the findings in the participants' contexts.

Researcher Reflexivity: being aware of your own beliefs, biases, and assumptions as a researcher. Self-interviews and reflexive journaling are two ways to engage in researcher reflexivity and allow for addressing potential biases and improving credibility.

Maximum Variation: deliberately selecting participants who differ in several dimensions relevant to the topic to uncover patterns and variations within a larger data spectrum. This allows for more diverse data, uncovering new patterns and improving credibility.

 Negative Case Analysis: seeking out data that contradicts patterns in your study; helps developed a more nuanced understanding of the phenomenon. This is accomplished by identifying contradictions, reevaluating themes, and incorporating alternative explanations.  

 Eslit: "Pedagogy in practice: a qualitative exploration of English language teaching (ELT) for graduate school using narrative inquiry, instructional material analysis, and observational inquiry"

The methods presented a good overview of the planned triangulation of the semi-structured interviews, the instructional materials, the classroom observations, and Orange visual mapping. A six-phase framework for interpreting the data was also discussed. It also clearly stated that member checking and peer debriefing would be utilized to enhance trustworthiness.

The data revealed 10 core themes, and in the results section, each theme was discussed through the triangulation of data. The researcher never identified a theme that had only one data source to back it up; there were always multiple groupings of data to verify that the theme was valid.

The "conflict of interest" statement and "generative AI statement" at the end of the article established the transparency of the author

Discussion Question #1: Our readings presented different types of triangulation which can be used in a study (method, investigator, theory, and data source), and each of those imply multiple entities (multiple methods, multiple investigators, multiple theories, multiple data sources). So can you have too much triangulation in a study? Is that even possible?

Discussion Question #2: Are there certain methods for ensuring rigor that align better with particular frameworks and/or methodologies?

Discussion Question #3: I feel like the Eslit article did a good job of telling us what they did to ensure rigor. However, I would have liked more info on this. If we use processes such as member checking or peer debriefing, should we include sections which describe these techniques (and any outcomes from them) in depth?