DBA Dissertation Methodology: Choosing Quantitative, Qualitative, or Mixed Methods

Choosing a DBA dissertation methodology is a decision most candidates make once, in Chapter 3, and then live with for the rest of the program. A DBA committee is not grading how sophisticated the proposed design sounds. It is testing whether the specific quantitative, qualitative, or mixed-methods approach a candidate proposes can actually answer the specific research question already on the table. Get that fit wrong and the mismatch rarely surfaces at the proposal review. It surfaces months later, at defense, when a committee member asks why a regression model was used to answer a question that was never really about correlation to begin with.

The instinct many candidates have is to treat mixed methods as the safe default, on the theory that combining approaches covers more ground and hedges against picking the wrong single method. Data on actual DBA dissertations shows close to the opposite pattern: a genuine, well-integrated mixed-methods case study design is one of the rarest choices in the population, not the insurance policy candidates assume it is. What decides the right methodology is not which one sounds most rigorous on paper. It is which one matches the research question as written, the applied orientation a DBA specifically rewards, and what a program’s own handbook will and will not approve.

This article covers how DBA candidates actually choose between quantitative, qualitative, and mixed-methods designs, where individual programs’ stated preferences genuinely diverge, and the specific mistakes that show up most often once a methodology chapter reaches committee review. Still scoping a topic before methodology is even the next decision? DBA dissertation topics covers choosing one a committee will approve. Methodology already chosen and the analysis itself is the next hurdle? Jump straight to matching support to this stage.

1. The Short Answer

A DBA dissertation methodology is the research design, quantitative, qualitative, or mixed methods, that a candidate proposes in Chapter 3 and a committee approves before data collection begins. Quantitative designs test a relationship that can be measured numerically, most often through correlational, survey, or regression approaches. Qualitative designs understand a phenomenon in the words of the people living it, and in DBA research specifically, a case study of a single organization or a small, purposeful sample of them is by far the most common form that takes. Mixed methods designs combine both when neither alone can answer the full research question, and despite years of methodology textbooks recommending it as an option, it remains the rarest of the three choices actually completed. Still deciding between a DBA and a PhD before any of this is relevant? DBA vs PhD dissertation covers that earlier decision, and DBA dissertation proposal covers getting chapters 1 through 3, methodology included, through committee review.

2. What Actually Decides the Choice

Three things determine which methodology actually fits a given DBA project, and none of them is which approach a candidate happens to find more comfortable.

The first is the research question as literally worded. A question asking whether, and by how much, one variable predicts another is a quantitative question, regardless of what design a candidate might personally prefer. A question asking how or why a specific group of people experienced or made sense of something is a qualitative question. Questions that genuinely require both, one part measuring an outcome and a separate part explaining why that outcome occurred, are the only questions that justify the added time cost of a true mixed-methods design.

The second is what a DBA specifically rewards, as distinct from a PhD. DBA vs PhD dissertation covers that distinction in full, but the short version is that DBA research is applied and practitioner-oriented by design, which is a large part of why case study, grounded in a specific, accessible organization, shows up so disproportionately often among the qualitative designs DBA candidates actually complete.

The third is access. A working professional pursuing a DBA typically has direct access to one organization, their own employer, or a small network of comparable ones, and comparatively little access to the large, randomly sampled populations a strong quantitative design usually needs. That access constraint alone rules out some designs before methodological preference ever enters the decision.

Weighing those three factors together, more than any abstract judgment about which paradigm is more rigorous, is what actually lands most candidates on their final choice.

3. Quantitative Designs in DBA Research

A quantitative design is the right choice when a candidate has, or can realistically obtain, numeric data at a sample size large enough to detect the effect being studied, and the research question is genuinely about magnitude or relationship rather than meaning.

The designs that show up most often in DBA quantitative chapters are correlational and survey-based, testing the relationship between variables like leadership style and employee retention, or a process change and a measurable performance outcome, typically using instruments already validated in the existing literature rather than built from scratch. Regression, in its simple, multiple, or logistic forms, is the workhorse analysis behind most of them. Quasi-experimental designs, comparing an outcome before and after an intervention without a fully randomized control group, are the next most common, useful when a candidate can study a real organizational change as it happens but cannot ethically or practically randomize who experiences it.

The step candidates skip most often, and the one committees ask about first, is a formal a priori power analysis establishing the minimum sample size the design needs before data collection starts, typically run in G*Power. A quantitative chapter built on a sample too small to detect a meaningful effect is a design flaw a statistician can identify in minutes, and a committee just as fast. Editverse’s Dissertation Statistical Analysis service prices a straightforward two-to-three-question quantitative chapter, t-tests, ANOVA, correlation, or chi-square, at its $600 entry tier; a design needing multiple regression, ANCOVA, or mediation analysis runs $900 to $1,200 depending on how many research questions the design has to answer.

4. Qualitative Designs in DBA Research

Case study is not just a common qualitative choice in DBA research. In the population that has actually been measured, it is the dominant one. Bell and Warren’s 2023 analysis of DBA theses awarded across UK institutions between 2012 and 2021, drawn from 1,130 theses tracked centrally through the British Library’s EThOS platform, found that approximately 30 percent adopted a case study design, more than any other single approach measured. An earlier analysis by Costley and Armsby, cited within Bell and Warren’s own review, found case study the most popular design across professional doctorate research generally, a broader population than DBA dissertations specifically, followed by action research, ethnography, and survey designs, in that order.

The reason case study fits DBA research so well is structural, not stylistic. A candidate studying their own organization, or a small set of comparable ones they can actually access, is already working within the boundaries a case study design is built around: a bounded system, studied in depth, using multiple sources of evidence rather than one. Robert Yin’s Case Study Research and Applications, the field’s standard reference, frames a strong case study around a clear unit of analysis, explicit propositions the evidence is tested against, and triangulation across several data sources, interviews, documents, and archival records, rather than any one of them alone. A DBA case study that skips the propositions and the triangulation, collecting interview data and reporting it as narrative with no explicit analytic framework, is the version committees push back on hardest, not because case study itself is weak, but because that version of it is undisciplined.

Action research, common enough in DBA programs to appear on more than one approved-methods list, differs from case study in a specific way worth naming: it requires the candidate to actually intervene in the organization being studied, implementing and then evaluating a change, rather than studying an existing situation from the outside. That makes it a strong fit for a candidate positioned to drive an actual initiative at their employer, and a poor fit for one who has only observational access.

5. Mixed Methods, Properly Scoped

Creswell and Plano Clark’s Designing and Conducting Mixed Methods Research, the field’s standard methodology text, defines three core mixed-methods designs, and a DBA candidate proposing mixed methods should be able to name which one, and why, rather than describing the project as using “some qualitative and some quantitative” data.

Bell and Warren’s same UK dataset puts a specific number on how rare the combination actually is: only just over 2 percent of the 1,130 DBA theses in their 2012-2021 sample adopted a mixed methods case study design specifically, the narrower combination of mixed methods and case study together, not mixed methods as a category on its own. That is a specific figure about one specific design combination in one national dataset, not a claim that mixed methods generally is rare among DBA dissertations everywhere. It is, however, a useful corrective to the assumption that combining methods is the easy or safe option: it is the design candidates complete least often, and the UK data is consistent with why.

The reason is integration, not data collection. Running a survey and also conducting a handful of interviews is not, on its own, a mixed-methods design. It is two separate, unintegrated studies bound into one document. A genuine mixed-methods design has an explicit point, named in the methodology chapter, where the two data types are actually brought together to answer something neither could answer alone, and that integration step is exactly what a committee scrutinizes hardest in a mixed-methods defense. Editverse’s Dissertation Statistical Analysis service prices a properly integrated mixed-methods design, structural equation modeling combined with qualitative coding, or a full convergent or sequential integration, at its Complex tier, starting at $1,500.

6. Where DBA Programs Disagree

General methodology advice, including everything above this point in the article, assumes a candidate is choosing freely among quantitative, qualitative, and mixed methods. Individual DBA programs do not always leave that choice fully open, and the two handbooks below show how differently “the program allows it” can actually work in practice.

Trident University’s DBA Handbook is the more prescriptive of the two. Its list of approved research designs is entirely qualitative and applied, case studies, action research, program evaluation, process improvement, field studies, survey studies, and business process improvement, with no traditional quantitative hypothesis-testing design included. On mixed methods specifically, the handbook states: “Mixed methods designs are acceptable. Due to the time required for creation and validation of quantitative instruments, it is highly recommended that quantitative components are used as triangulation of data through secondary data sources, such as government database records.” In practice, that steers candidates toward using existing government or organizational data to support a qualitative core, rather than collecting new quantitative data primary to the study. The same handbook sets a floor of “at least three different data sources with a minimum total sample size of 15 participants” for the methodology section, a threshold characteristic of a qualitative design, not a quantitative one.

Westcliff University’s DBA Handbook takes a different kind of control entirely. It does not specify which methodology type is preferred or excluded. What it does specify is process: during the BUS 900 course, the candidate submits chapters 1 through 3, the full proposal including methodology, for the dissertation chair to review before the wider committee ever sees it, and the chair separately reviews the IRB application before it is submitted. The constraint at Westcliff is not which method a candidate picks. It is that the chosen method has to clear one specific person, the chair, at one defined checkpoint, before it goes any further.

The practical implication is the same regardless of which program a candidate is in: read the actual handbook before assuming any general methodology advice, including the advice in this article, applies without modification. A Trident candidate proposing a pure quantitative hypothesis-testing design is arguing against the handbook itself, not just against convention. Franklin University’s DBA research methods guide is a useful neutral starting point for candidates whose own program has not published anything this specific, though it deliberately stops short of recommending one method over another, leaving that judgment to the candidate’s own committee.

7. The Rigor Standard Doesn’t Change

Whichever methodology a candidate ultimately proposes, the committee’s underlying standard does not move. A quantitative chapter still needs a defensible sampling rationale and a power analysis justifying the sample size used. A qualitative chapter still needs a defensible case-selection rationale and an explicit account of how trustworthiness, the qualitative equivalent of validity, was established. A mixed-methods chapter needs both of those, plus the integration argument tying them together.

That standard does not relax once the methodology chapter is approved, either. It resurfaces in full at the final oral defense, where a committee that signed off on a methodology in writing is entitled to ask, live and without advance notice, why that specific test was chosen over a comparable one, or how a specific coding scheme was actually applied to the raw data. DBA dissertation defense covers what that live questioning actually looks like, program by program. A methodology chapter that reads as defensible on paper but that the candidate cannot explain out loud, in their own words, under a direct follow-up question, is one of the more common ways a defense goes worse than the written manuscript would predict.

8. Common Methodology Mistakes

Four mistakes account for most of the methodology-chapter pushback DBA candidates receive, across every program and every method type.

The first is a research question and a methodology that do not actually match, most often a question phrased with “how” or “why” paired with a quantitative design built to answer “whether” or “how much.” DBA dissertation proposal covers this specific mismatch as one of the more common reasons a full proposal gets sent back at that earlier checkpoint, and it is far cheaper to catch there than after data collection.

The second is a quantitative sample sized by convenience rather than by a power analysis, then defended after the fact with a claim of significance the sample was never large enough to reliably detect.

The third is what amounts to a mixed-methods design in name only: a survey bolted onto a handful of interviews with no named design type and no explicit integration step, which reads to a committee as unplanned rather than as a deliberate convergent, explanatory sequential, or exploratory sequential choice.

The fourth is a broader pattern worth candidates knowing about regardless of which methodology they choose: statistical reporting errors are common even in published, peer-reviewed research, not just in dissertation drafts. A widely cited PLOS ONE analysis of 697 articles across six major psychology journals found that 63 percent contained at least one p-value inconsistent with its own reported test statistic, and 20 percent contained an error serious enough to change whether a result was reported as statistically significant. That figure comes from psychology journals, not DBA dissertations specifically, but the lesson generalizes directly: a result that looks clean in a table is not automatically a result that was calculated and reported correctly, which is exactly the kind of error a second, independent check catches before a committee does.

9. Matching Support to This Stage

A candidate who has settled on a methodology and now needs it executed, the actual data cleaning, testing, and analysis, rather than the chapter that proposes it, has a different problem than a candidate still choosing between designs, and Editverse prices the two differently.

Still choosing a methodology rather than needing one executed? The Silver tier, $1,250, covers the chapters 1-3 proposal stage, methodology chapter included, without the analysis-execution component Gold and Platinum add; see DBA dissertation proposal for how that earlier checkpoint works. Every service on this page lives in Editverse’s Dissertation Services category alongside standalone support for candidates who need one specific piece rather than a full package.

10. Frequently Asked Questions

11. The Bottom Line

A DBA dissertation methodology is judged on fit, not sophistication: does the chosen design, quantitative, qualitative, or mixed, actually match the research question, the applied nature of DBA research, and what the candidate’s own program will approve. Case study dominates the qualitative designs DBA candidates complete for structural reasons, not fashion. Mixed methods, done properly, is the rarest choice precisely because doing it properly is the hardest version of the assignment. Whichever design a candidate lands on, the committee’s rigor standard is the same one that resurfaces at defense, live and unscripted.