RES 6013 Module 3 Research Design Justification Example

Why Interviews Come First: Selecting and Justifying an Exploratory Sequential Mixed Methods Design for a Study of First-Year Nurse Turnover in Small Rural Hospitals

Student Name

American College of Education

RES6013: Research Methods

Module 3 Assignment

Instructor Name

October 26, 2026

Introduction

The research questions from Module 2 ask how newly licensed nurses describe their first year in small rural hospitals, which of the reasons they give are associated with leaving and how the two sets of findings fit together. A design must be able to answer all three. This paper considers five candidates, explains why four fall short and justifies the exploratory sequential mixed methods design I have chosen, including how its phases connect, how quality will be protected in each and what the design will cost in time and effort.

What the Questions Require

The qualitative question asks how and why, and it concerns experiences that no existing instrument has been built to capture in this setting. The second, quantitative question asks how strongly candidate factors track departure across a large group of nurses. The third, integrative question asks how the two sets of results bear on each other. Any design that answers only one kind of question, or that treats the two as unrelated studies, will leave part of the problem untouched. Johnson et al. (2007), after analyzing definitions offered by leading methodologists, described mixed methods research as combining qualitative and quantitative approaches for breadth and depth of understanding, which is precisely what these questions demand.

Alternatives Considered

A correlational survey alone would test associations between turnover and factors such as shift, unit, pay and job satisfaction across many nurses. It fails here because the factors it can test are only those I choose in advance, and Module 2 established that the reasons for leaving small rural hospitals are unknown. A survey built from urban research could measure the wrong things with great precision.

A qualitative case study of the three hospitals would answer the how and why question in depth, drawing on interviews, documents and observation. It fails the second question, since 15 or 20 interviews cannot show which reasons are common across the region, and administrators deciding where to spend limited money need that.

Among mixed methods designs, Fetters et al. (2013) described three basic forms: convergent, in which both kinds of data are collected at once and compared; explanatory sequential, in which a quantitative phase is followed by a qualitative phase that explains its results; and exploratory sequential, in which a qualitative phase comes first and informs a quantitative phase. A convergent design would again require a survey before I know what to ask. An explanatory sequential design has the same problem in a different order, starting with a survey whose content is uncertain.

The Chosen Design

An exploratory sequential design fits because the problem's central gap is that no one knows what to measure. In Phase 1, I will interview 16 to 20 newly licensed nurses from the three hospitals, roughly half who stayed past one year and half who left, using a semi-structured protocol about their first year and, for those who left, their decision. The interviews will be analyzed with reflexive thematic analysis, which Braun and Clarke (2019) described as an approach in which themes are actively developed by the researcher rather than simply found in the data, a stance that suits an insider who must account for her own influence.

In Phase 2, the themes will be turned into survey items, combined with established scales where suitable ones exist, such as a validated measure of job satisfaction, and tested with a small pilot group for clarity. The final survey will be given to newly licensed nurses at small rural hospitals across south Georgia at about six months after hire, and their employment status will be recorded at 12 months. Logistic regression will test which factors are associated with departure.

Where the Phases Connect

Fetters et al. (2013) explained that integration can occur at the design, methods and interpretation levels. At the design level, the exploratory sequence itself connects the phases. At the methods level, integration occurs through what they called building: the qualitative results build the quantitative instrument. Each survey item will be traced back to the theme it came from in a table that the committee can review. At the interpretation level, the findings will be presented in a joint display placing each theme beside the survey result for the related items, so that confirmations, extensions and contradictions can be seen directly. This answers the integration research question.

Rigor in Each Phase

Each phase follows the quality standards of its own tradition. For Phase 1, credibility will be supported by member checking, in which participants review a summary of their interview, and by a second coder, a nursing faculty member from another college, reviewing a sample of transcripts. Dependability will be supported by an audit trail recording each analytic decision. Because I know many participants, I will keep a reflexive journal and recruit nurses who left through a neutral third party.

For Phase 2, content validity of new items will be supported by tracing each to interview data and by expert review from two nurse researchers. Internal consistency will be checked with Cronbach's alpha in the pilot and full samples. The sample size for the logistic model will be set using the events-per-variable guideline, which will determine how many factors the model can include.

Costs and Limitations

The design has real costs. It takes longer than a single-method study, because Phase 2 cannot begin until Phase 1 is analyzed and the survey is built and piloted; I estimate 18 months for data collection alone. It requires competence in two kinds of analysis. And the survey's validity depends on the quality of the interview phase: themes that are weak or shaped by my insider position would carry their bias into Phase 2. The design is also not causal. Even a strong association between, say, preceptor turnover and new nurse departure will not show that one caused the other, and the dissertation will use the language of association throughout.

Timeline

The phases fit a realistic schedule. Months 1 through 3 cover IRB approval and recruitment for Phase 1. Months 4 through 7 cover interviews, transcription and analysis, with member checks completed by month 8. Months 8 through 10 cover item writing, expert review and a pilot with about 30 nurses who are not in the main sample. The main survey opens in month 11 and stays open for each new hire's six-month point through month 18, with 12-month employment status collected through month 24. Analysis and integration fill the final months. The 12-month follow-up is the step that cannot be shortened.

Fit With the Paradigm

Module 1 adopted pragmatism, which lets the question choose the method. The exploratory sequential design is a direct expression of that choice: it uses an interpretive approach where understanding is needed and a measurement approach where breadth is needed. In the view of Johnson and Onwuegbuzie (2004), combining methods earns its keep when it follows the logic of the research questions rather than a preference for either tradition, and that standard is the one this design is meant to meet.

Conclusion

The exploratory sequential mixed methods design is the only one of the five candidates that can find out why new nurses leave small rural hospitals and then test how common those reasons are. Its phases connect through building and a joint display, each phase has its own safeguards for quality and its costs are acknowledged. Module 4 turns to sampling and data collection for both phases.

References

Braun, V., & Clarke, V. (2019). Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health, 11(4), 589-597. https://doi.org/10.1080/2159676X.2019.1628806

Fetters, M. D., Curry, L. A., & Creswell, J. W. (2013). Achieving integration in mixed methods designs: Principles and practices. Health Services Research, 48(6, Pt. 2), 2134-2156. https://doi.org/10.1111/1475-6773.12117

Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed methods research: A research paradigm whose time has come. Educational Researcher, 33(7), 14-26. https://doi.org/10.3102/0013189X033007014

Johnson, R. B., Onwuegbuzie, A. J., & Turner, L. A. (2007). Toward a definition of mixed methods research. Journal of Mixed Methods Research, 1(2), 112-133. https://doi.org/10.1177/1558689806298224