PHLT 8060 Week 8 Three-Cut Scope Revision Example

Scope Revision: Three Cuts That Turn a Research-Team Claim About Rural Hepatitis B Birth Doses Into One a Single Doctoral Candidate Can Finish With Public Records

Student Name

Walden University

PHLT 8060: Dissertation Seminar I

Instructor Name

Month Day, Year

Scope Revision: Three Cuts That Turn a Research-Team Claim About Rural Hepatitis B Birth Doses Into One a Single Doctoral Candidate Can Finish With Public Records

The Starting Claim

After the scoping search confirmed that the core comparison is open, the problem statement read: "The 2021 change requiring rural hospitals to purchase newborn hepatitis B vaccine caused birth dose coverage to fall at the nine birthing hospitals in six rural counties, because pharmacy budgets led hospitals to stock less vaccine and staff to skip the dose, and interviews with nurses and pharmacists will show how." Every part of that sentence promises more than one candidate could deliver with the data available. This memo shows three cuts, each set against the sentence it changed, and explains what each gives up.

Cut 1: The Causal Verb

Before: "The 2021 change ... caused birth dose coverage to fall."

After: "Birth dose coverage ... differed between births in the 30 months before and the 30 months after the 2021 change."

A causal claim would require ruling out every other explanation for a change in coverage over the same period: the pandemic's effect on hospital staffing, a new statewide newborn order set promoted in 2022, changes in registry reporting software and shifts in the families giving birth at rural hospitals. With nine hospitals and no comparison group of hospitals unaffected by the change, the design cannot exclude these alternatives. Shadish et al. (2002) describe history, the occurrence of other events between a pretest and posttest, as the central threat to simple before-and-after comparisons, and every one of these alternatives is a history threat. The revised verb asks whether coverage differed, a question the data can answer, and leaves the explanation to the discussion chapter, where possible causes can be weighed honestly.

What is given up: the dissertation will not be able to say that the purchasing change reduced coverage. It will be able to say whether coverage changed around that time and by how much.

Cut 2: Private and Out-of-System Births

Before: "at the birthing facilities in six rural counties," implicitly including every birth in the region.

After: "among births at the hospitals that send records to the state registry every month and whose birth certificates can be matched to those records."

A tenth facility, a small privately owned hospital, delivers about 120 infants a year and reports to the registry only intermittently, and births at a freestanding birthing center are not consistently linked. Including them would mean either imputing missing doses or treating unrecorded doses as missed, both of which could distort coverage. Excluding them narrows the population slightly and states plainly who is counted. Creswell and Creswell (2018) advise that a study's population be defined by the data that can actually identify its members, and the revision applies that rule.

What is given up: coverage at the excluded facilities remains unknown, and the problem statement will say so.

Cut 3: The Interview Component

Before: "and interviews with nurses and pharmacists will show how."

After: removed from the dissertation; noted as a future study.

Interviews would address the most interesting question, why coverage differs, but they would add a second research tradition, a second approval process and travel to nine hospitals across a large rural region, roughly doubling the work. They would also be retrospective, asking staff to recall practices from up to five years earlier, which weakens their value as evidence about the past period. Booth et al. (2016) caution that a project trying to answer several questions at once often answers none of them well. Removing the interviews keeps the dissertation to one question answered with one kind of data.

What is given up: the dissertation will describe whether and where coverage changed, not why. The mechanism question becomes the first recommendation for future research.

The Revised Statement

After the three cuts, the problem now concerns well newborns of 2,000 grams or more delivered across five calendar years at the nine hospitals with complete registry reporting, and asks whether the share whose first hepatitis B vaccination appears in the registry within a day of birth was different in the two and a half years after the state stopped supplying newborn vaccine than in the two and a half years before. That version can be completed by one candidate using linked registry and birth records under an existing data agreement, within two years.

Checking the Revision Against the Original Concern

A scope revision can cut away the reason a problem mattered. This one keeps it: the concern was that rural infants may be leaving hospitals unprotected, and the revised statement still measures exactly that, at the hospitals where it occurs, across the period when the concern sharpened. What it drops is the explanation of why, which the original concern assumed rather than demonstrated.

References

Booth, W. C., Colomb, G. G., Williams, J. M., Bizup, J., & FitzGerald, W. T. (2016). The craft of research (4th ed.). University of Chicago Press.

Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE.

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.