03 · What You Need to Know
A Research Design Can Exist Even When It Does Not Have One Neat Name
Every empirical study has a structure
Research design is not created by naming it.
The design exists in the decisions that organize the investigation: what or who is studied, how observations are generated, whether comparison groups or conditions exist, whether researchers manipulate an intervention, when measurements occur, how participants or cases are selected, and how those features permit the research question to be answered.
A study therefore has design characteristics even if the researcher never writes a sentence beginning “The research design was...”
This follows from what research design is actually supposed to accomplish. The design is the logic connecting the question to the evidence, not merely the terminology used to summarize that logic.
A design label is shorthand for that structure
Although a label does not create the design, a good label can make the design easier to understand.
Terms such as randomized controlled trial, prospective cohort study, case-control study, cross-sectional study, ethnography, or an established mixed methods design can convey substantial methodological information to readers familiar with the relevant tradition.
That is why recognized reporting guidance sometimes explicitly asks authors to identify the design.
For observational epidemiological research, STROBE recommends indicating the study's design with a commonly used term in the title or abstract and presenting key elements of the study design early in the article.
In such contexts, refusing to name a clearly recognizable design would not make the paper more methodologically sophisticated. It would simply make readers work harder.
But not every study maps perfectly onto one conventional label
Research designs can be complex, hybrid, adaptive, nested, multiphase, interdisciplinary, or shaped by methodological traditions whose categories do not align neatly with those used elsewhere.
A project might contain a quasi-experimental component, longitudinal follow-up, and qualitative implementation inquiry. A qualitative study may draw on a coherent methodological orientation without fitting a single design taxonomy familiar across all fields. Secondary analyses may use existing data whose original structure was created for a different purpose.
In such cases, one short label can hide as much as it reveals.
The solution is not necessarily to invent a new design name. It is to describe the study's structure accurately.
Do not invent a design label because you think every study needs one
This is an important practical rule.
If you cannot find a recognized methodological source using the design term you intend to use, pause before creating one.
Researchers sometimes combine familiar adjectives into labels that sound plausible: “integrated descriptive-developmental causal-comparative design,” for example. The individual words may have methodological meanings, but their combination does not automatically become an established design.
Inventing terminology can make a study harder to interpret because readers do not know what methodological commitments the label is supposed to imply.
Clear prose is usually safer:
“We collected questionnaire data from the same cohort at three points during the academic year to examine changes in AI-use practices and their associations with academic self-efficacy.”
A reader can understand important features of the design even before a conventional label is attached.
Sometimes the design name is expected or required
Whether you should explicitly name the design is partly determined by the reporting context.
Your university may require a “Research Design” subsection in a thesis proposal. A funding application may ask for the study design. A protocol template may require it. A journal may follow a reporting guideline that expects a commonly recognized design term.
STROBE provides a concrete example. Its checklist recommends identifying the design with a commonly used term and presenting key design elements, while also emphasizing that STROBE is guidance for reporting observational research rather than a prescription for designing studies.
That distinction matters. Reporting requirements tell you what information should be communicated. They do not necessarily imply that every research tradition shares the same design taxonomy.
Check the reporting guideline appropriate to your study
Before deciding that naming the design is optional, check whether an established reporting guideline applies.
Different study types have different reporting expectations. In fields where formal reporting guidance exists, design terminology may be expected in the title, abstract, methods, protocol, or other sections.
For observational research within STROBE's scope, for example, cohort, case-control, and cross-sectional studies are treated as distinct designs, and authors are asked to identify the design using commonly used terminology.
The safest practice is therefore contextual: check the target journal, reporting guideline, institutional requirements, disciplinary convention, and methodological literature relevant to your study.
Transparent description matters more than the label alone
Even when a design is explicitly named, the label is not enough.
Suppose a paper states:
“This was a cross-sectional study.”
Readers still need to know the setting, population, eligibility criteria, participant selection, measurement procedures, variables, potential sources of bias, study size, and analytic methods.
STROBE makes this explicit. In addition to asking authors to name the observational design, it asks for key design elements and detailed information about setting, participants, variables, measurement, bias, study size, and statistical methods.
A correct label therefore cannot rescue an incomplete methods section.
Naming the design
Provides concise methodological orientation using recognized terminology.
Reporting the design
Provides enough information about the study's actual structure and procedures for readers to understand, appraise, and interpret the evidence.
The second task is indispensable even when the first is easy.
An inaccurate label can be worse than no label
Design terminology creates expectations.
If you call a study randomized, readers expect random allocation. If you call it longitudinal, readers expect a temporal structure consistent with that term. If you call it ethnographic, readers may expect sustained engagement with culture, practice, context, and the methodological commitments associated with ethnographic inquiry.
Using a label that the study does not satisfy can therefore mislead readers about the evidence.
This is why choosing the appropriate level of specificity when naming a design is preferable to selecting the most impressive-sounding term available.
If no established label fits cleanly, describe the design first and use broader terminology where appropriate.
Not naming a design does not mean “anything goes”
There is an important difference between rejecting unnecessary taxonomy and failing to design the study.
You cannot replace a design statement with:
“The researchers used surveys and interviews.”
That identifies methods but leaves the study's structure unclear.
When were the survey and interviews conducted? Who participated in each? Did interview sampling depend on survey results? Were the components intended to answer different questions? Were their findings integrated? Was there a comparison group? Were observations repeated?
The distinction between research design and research methods still applies when the design is described rather than formally named.
Complex studies may be better described hierarchically
A complex project does not always need one all-encompassing label.
Suppose researchers evaluate an educational intervention using a quasi-experimental outcome component and qualitative case studies of implementation.
They might describe the overarching study as mixed methods, identify the quasi-experimental and qualitative components, and then explain how the components are integrated.
Alternatively, if the qualitative component is embedded within a predominantly quantitative evaluation, the reporting might emphasize the primary design and describe the embedded component separately.
This approach is especially useful when one study contains multiple research designs or design components.
Qualitative research makes the limits of universal design labels especially visible
Qualitative traditions differ in how they use terms such as methodology, approach, design, strategy, and method.
Some studies are explicitly grounded in recognized traditions such as ethnography, phenomenology, grounded theory, narrative inquiry, or case study research. When the methodological commitments of one of these traditions genuinely organize the investigation, naming it can be highly informative.
Other qualitative investigations may be appropriately described more generically, provided the researcher explains the research question, sampling or case selection, context, data generation, analytic approach, reflexivity where relevant, and the logic connecting evidence to interpretation.
Attaching a tradition-specific label solely because it seems more scholarly can be counterproductive if the study does not actually follow that tradition.
Secondary data research may require description more than a neat label
Studies using existing datasets create another interesting case.
The original data may have arisen from a cohort, survey, trial, administrative system, learning platform, registry, or another source. The secondary study may ask a question different from the one that motivated the original data collection.
Researchers should describe both the provenance and structure of the data and the design logic of the current analysis. Simply borrowing the original study's design label may be misleading if the new analysis uses only a subset of the data or addresses a fundamentally different question.
Again, transparent description becomes more useful than forcing one label to perform too many jobs.
A design label should reduce ambiguity
The easiest test is pragmatic.
Does naming the design help a knowledgeable reader understand the study more quickly and accurately?
If yes, use the label.
If the label is required by a reporting standard, journal, institution, or protocol, use it accurately.
If the label is ambiguous, disputed, or likely to misrepresent the study, explain what you did in clearer methodological language and define the term if you still need to use it.
The purpose of terminology is communication, not certification.