PU565 Unit 10 Mixed Methods Methods Paper Example

Interviews First, Then a Mailed Survey: A Unit 10 Methods Paper Defending an Exploratory Sequential Design for a Composite County Hearing Study Against the Four Designs It Could Have Been

[Student Name]

Purdue University Global

PU565: Qualitative and Quantitative Methods

Unit 10 Methods Paper

[Instructor Name]

[Date]

Culbreth County and its study are composite; the designs compared, the instruments and the research cited are real.

The Question and Why It Needs Two Strands

Most older adults in the county with self-reported hearing difficulty use no hearing aid, and nobody can say why. The use figure is consistent with national work finding that a small minority of older adults with hearing loss have used a device (Chien & Lin, 2012). A single strand cannot answer both halves of what the county needs: a count of how common each reason is, and an account of what the reasons mean to the people giving them.

The Design

Phase one: 24 semi-structured interviews with county residents over 60 who report hearing difficulty and use no aid, purposefully sampled for maximum variation across severity, household composition and previous hearing care contact. Analysis by thematic analysis, two coders, codebook developed to stable definitions (Braun & Clarke, 2006). Phase two: a mailed survey of residents over 60, combining a published handicap screen used intact, items adapted from national wording, and four items written from phase one. Analysis by weighted proportions and two prespecified stratified comparisons. Integration at three points: design, where phase one decides whether and what phase two asks; instrument, where interview findings become items; and interpretation, in a joint display built before the survey is fielded (Fetters et al., 2013).

Alternative One: Survey Only

Cheaper, faster, and it is what the county asked for. Rejected because its answer choices would have been borrowed from published work instead of from the county's own residents, and the study's most useful finding would have been invisible: half of those interviewed had no working idea what the retail devices are. No item set written in advance would have contained that, because the literature frames the question as cost and stigma.

Alternative Two: Interviews Only

Defensible on its own terms and rejected for one reason. The county has to decide whether to spend money on information, on subsidy or on neither, and that decision needs a proportion. An interview study would have produced a strong account and no denominator, and the account would have been quoted as though it were prevalence, which is the misuse this paper is written to prevent. The survey alone would have measured the wrong things precisely. The interviews alone would have been quoted as though they were measurements.

Alternative Three: Explanatory Sequential, Survey First

The common design in this field: measure, then interview the people whose answers were surprising. It would have been right if the county had a candidate explanation to test. It had none, and a survey run first would have fixed the explanations into its response options before anybody knew what they were. The reversal is the whole argument of the paradigm memo written in month two.

Alternative Four: Convergent, Both at Once

Attractive on timeline, since nine months of interviewing and six of survey work would have overlapped. Rejected because the instrument depends on the interview findings, and a convergent design would have required writing the survey before phase one was analyzed, which is alternative one with extra steps. Convergent designs suit questions where both strands can be specified in advance, and this was not one.

Quality Criteria, Applied Separately to Each Strand

Phase one is judged on credibility and on an auditable trail: field notes after every interview, a codebook with dated changes, the three versions of one contested definition kept rather than replaced, and member checking with participants who agreed to review a summary. Phase two is judged on the screen's existing validation, response rate, and the honest statement that a pilot of 20 cannot establish reliability. The paper applies two sets of criteria rather than one, and says why, because a report that asks for reliability from interviews or credibility from a mailed survey will be criticized by whichever reviewer notices first.

Ethics

Review board approval for both phases. Interview participants receive 40 dollars and are told the interviewer is not from the health department or a device company. Accommodations for hearing difficulty are written into the protocol rather than left to the interviewer, including printed question cards and a break offered unasked at 25 minutes. No participant is identified by occupation in any quotation, which matters in a county where one retired teacher is identifiable from that detail alone.

Why the Screen Was Borrowed Intact

One instrument decision carries enough weight to be defended in the methods paper rather than left in the instrument design. The handicap screen is used word for word, including its older phrasing, because rewriting items from a validated screen produces an instrument with no validation behind it and breaks comparison with every published estimate that used it (Ventry & Weinstein, 1982). The cost is that two items read oddly beside the new ones written from the interviews, and the pilot tests exactly that. The alternative, a modernized wording, would have bought readability at the price of the only external benchmark the survey has, and for a study whose purpose includes telling a county how it compares, that trade is the wrong way round.

Limitations, Including the One Integration Cannot Fix

The interview sample cannot support any prevalence claim, and the survey cannot support any explanation. Self-reported handicap is not audiometric loss, so the denominator differs from clinical studies. Stated willingness to pay is not behavior. And both strands under-reach the most isolated residents, since every recruitment route requires leaving the house or reading mail, which means the study's two methods fail in the same direction rather than covering for each other. A third strand, through home-delivered meal routes or community health workers, is what would address it, and the paper recommends it as the next study rather than claiming the current design reaches those residents.

What the County Will Be Able to Do With This

Two things. It will have the proportion of residents over 60 who do not know what the over-the-counter category is, which is the information a public education decision needs. And it will have an account of deferral that separates price from timing, which matters because a subsidy aimed at people waiting for something other than money ends up in the hands of those already on the point of buying. The second finding came from interviews and could not have come from anywhere else, which is the practical defense of the design this paper has been arguing for.

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

Chien, W., & Lin, F. R. (2012). Prevalence of hearing aid use among older adults in the United States. Archives of Internal Medicine, 172(3), 292-293. https://doi.org/10.1001/archinternmed.2011.1408

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

Ventry, I. M., & Weinstein, B. E. (1982). The hearing handicap inventory for the elderly: A new tool. Ear and Hearing, 3(3), 128-134. https://doi.org/10.1097/00003446-198205000-00006