Reference no: EM134021972
Discussion 1:
Keri replied to week 8: I have completed the task of uploading all of my documents into Dedoose, with a final count of 21. They are not organized in any particular fashion at this point, only in the order they were added, as I am still reviewing tutorials for the program to better understand how to use the tool. That said, I am using an Excel sheet to organize a coding form. I have created document descriptors in the form and in Dedoose, such as participant age, study location, study date, and interventions performed. These descriptors have not been applied to specific articles at this time. I developed codes and have started excerpting, which has helped me begin to recognize trends for analysis. However, the more I research, I believe that I need to create code trees with parent and child codes to help categorize data. My research advisor is meeting with me this weekend to review my process and project status, which will help to guide my next steps.
Silvia notes that providing details and being transparent about the procedures used are important to study reviewers (Silvia, 2018). While following the IMRAD structure, the methods section should explain the "how" and "why" and demonstrate its connection to previous research to strengthen the study's validity (Silvia, 2018). Following Silvia's recommendation to choose a target journal first, I am adopting The Journal of Rural Health's submission guidelines. Currently, the PRISMA flow diagram is complete, and I have outlined the structure of the Methods, including study eligibility criteria, databases, search strategies/process/methods, and definition of outcomes, in an effort to maintain clarity and provide relevance to the readers. Considerations related to Quantitative Research Techniques may also support methodological transparency and rigor.
Q: Since you have already begun excerpting and identifying trends, what criteria will you use to decide whether a code should remain independent or become part of a larger parent-child coding hierarchy?
Discussion 2:
Christopher replied to week 8: By the end of week 8, I have made substantial progress on data preparation and cleaning in anticipation of future analyses. I formally closed my survey, aggregated the complete and in-progress spreadsheets into one full 182-participant document, imported it into SPSS, selected cases for consent, age = 18+, English language, and independent completion ability, and finished the survey, all coded as 1 ("yes" or "complete"), resulting in a cleaned sample of 82. An additional respondent was removed for straightlining responses across both the MFQ and SWAI measures. The total number of cleaned sample participants is 81, comprising 55 clinical supervisors, 26 supervisees, 38 matched dyadic participants (19 linked supervisor-supervisee pairs), 36 unmatched supervisors, and 7 unmatched supervisees. Additionally, I have recoded the standard Qualtrics export of 1-6 scores on MFQ items to match the validated 0-5 scale outlined in Graham et al. (2011) and presented to respondents in my survey, to allow for comparison of sample means across studies that also use the MFQ. I still need to compute the MFQ and SWAI subscales, composites, and overall scores, and prepare the data to compute dyadic congruence scores for week 9 analyses.
The IMRAD journal writing structure, Introduction, Methods, Results, and Discussion, is a tried and true outline for many journals and is essential to keep in mind when conducting and writing social science research (Silvia, 2018). The methods section is particularly primed for emulation from the target journal, as indicated by Silvia (2018), and was something I incorporated early in the research process, as Dr. J suggested I find a sample article from my target journal, The Clinical Supervisor, and found in the work of Cook and Zhu (2026). This process allowed for an essential plug-and-play approach, as I laid out the participant statistics sections in fill-in-the-blank format to be completed once my data collection closed. Data preparation was mostly straightforward: After removing respondents who did not consent or did not complete both measures, I was left with 82 respondents. My inclusion rules (completing both measures, response times greater than or equal to 3 minutes, and passing the evaluation for straightlining responses and two attention checks) resulted in only 1 response being removed, leaving the final total at 81. One additional decision I made was how to treat a supervisor with two linked dyadic supervisee responses. I determined that the first response, completed earliest and in the capitalized format, was the included and linked response, while the later supervisee with the same code but in lowercase would be treated as an unlinked supervisee. This decision will be included in the methods section regarding inclusion/exclusion. The use of Survey Sampling principles and Statistics methodologies can help support accurate interpretation of participant data and analytical decisions.
Q: As you move into calculating dyadic congruence scores, what potential challenges do you anticipate when interpreting differences between matched and unmatched supervisor-supervisee participants?