PU565 Unit 9 Mixed Methods Matrix Example

Three Named Points Where the Two Strands Meet: A Unit 9 Mixed Methods Matrix Locating Integration in a Composite County Hearing Study at Design, at Instrument Building and at Interpretation, With a Fourth Point Rejected

[Student Name]

Purdue University Global

PU565: Qualitative and Quantitative Methods

Unit 9 Mixed Methods Matrix

[Instructor Name]

[Date]

The composite county study is invented; the integration framework and the research cited are real.

The Framework Being Applied

Integration in mixed methods work can occur at the level of design, at the level of methods, and at the level of interpretation and reporting, and the useful question is which of these a given study actually does rather than whether it calls itself mixed (Fetters et al., 2013). The matrix below answers that for one study.

Point One: Integration at Design

1. What it is: an exploratory sequential design, qualitative first, quantitative second, with the second phase's content dependent on the first.

2. What happens: the interview analysis determines whether a survey is worth fielding and what it has to ask. If the interviews had produced no candidate explanation beyond cost, the survey would have been a one-page cost question rather than a six-section instrument.

3. Document and date: the paradigm memo of month two, which set the sequence and named the condition that would have reversed it.

4. How to check it: the survey instrument should contain items that could not have been written in month one. Four items in Section C meet that test.

Point Two: Integration at the Methods Level, Through the Instrument

1. What it is: building instrument items from qualitative findings, which the framework calls connecting.

2. What happens: half the interview sample, six people out of twelve, had no clear idea what the new retail category covers, so four items were written to estimate how widely that holds. Separately, the interviews showed price and deferral doing different work, so the standard single barrier item was split into two response options.

3. Document and date: the instrument design of month six, where every item carries its source label.

4. How to check it: the item sources are printed in the instrument. Where the source says interviews, the wording should be traceable to a transcript, and in three of the four cases a participant's phrasing survives in the item.

The test of integration here is whether the instrument contains anything the literature alone would not have produced. Four items do, and the rest of the survey would have existed anyway.

Point Three: Integration at Interpretation

1. What it is: bringing the two sets of results together in one interpretation, which the framework calls merging, with explicit attention to where they disagree.

2. What happens: the report will place the survey's prevalence estimates beside the interview themes in one table, theme by theme, with three possible outcomes named in advance. Confirmation, where the survey finds the theme common. Expansion, where the survey finds it rare but the interviews explain why it matters for those it affects. And discordance, where the two point different ways, which is reported as discordance rather than reconciled.

3. Document and date: the methods paper of month nine, with the joint display specified before the survey is fielded.

4. How to check it: the joint display exists and contains at least one discordant row, or the report states that none occurred.

The Fourth Point, Rejected

The writer had planned to claim integration at sampling, on the grounds that the interview sampling dimensions reappear as survey stratification variables. On examination this is not integration but consistency: using the same severity bands and household categories in both strands makes the results comparable, which is good practice and involves no transfer of findings from one strand to the other. Claiming it would inflate the number of integration points without adding anything, and the matrix says so rather than quietly counting it.

What Each Strand Is Not Allowed to Do

1. The interviews may not be used to estimate how common anything is, including the information finding they produced.

2. The survey may not be used to explain why respondents answer as they do, including on the items the interviews generated.

3. Neither strand may be weighted against the other in the abstract: the abstract reports both, in the order they were conducted.

How the Joint Display Will Be Built

The display has one row per interview theme and four columns: the theme with the count of transcripts behind it, the survey item that bears on it, the survey result with its confidence interval, and the writer's reading of the pair as confirmation, expansion or discordance. Themes are entered in the order the thematic analysis produced them rather than in order of survey support, so the display cannot be reordered to flatter the quantitative strand (Braun & Clarke, 2006). Any theme with no survey item facing it appears with an empty column and a one-line explanation, which is the most useful cell in the table because it shows what the survey failed to carry over. The display is drafted before the survey is fielded and filled in afterward, and the draft version is kept so that the two can be compared.

Where This Design Is Weakest

The two strands share one blind spot, since both the interview recruitment routes and the mailed survey under-reach the most isolated residents. Integration cannot repair that, because integration combines strands rather than extending their reach, and the purposeful routes chosen for the interviews were never going to reach that group either (Palinkas et al., 2015), and a study whose two methods fail in the same direction should say so in the limitations rather than treating mixing as coverage. The honest version of this matrix therefore has three integration points and one shared weakness, and the weakness is the more important of the two for anybody deciding what to do with the findings.

References

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. https://doi.org/10.1191/1478088706qp063oa

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

Palinkas, L. A., Horwitz, S. M., Green, C. A., Wisdom, J. P., Duan, N., & Hoagwood, K. (2015). Purposeful sampling for qualitative data collection and analysis in mixed method implementation research. Administration and Policy in Mental Health and Mental Health Services Research, 42(5), 533-544. https://doi.org/10.1007/s10488-013-0528-y