When Should Two Research Questions Be Combined Into One Study?

Do These Two Questions Really Belong in the Same Study?

You begin with one research question. While planning the study, another question emerges. It concerns the same general topic, perhaps the same participants, and you could collect the additional data without much difficulty. Should you combine both questions into one study?

Sometimes that is exactly the right design. A second question may explain, extend, or qualify the first, allowing the study to tell a more complete scientific story. In other cases, the apparent connection is mostly logistical: the questions happen to involve the same participants, dataset, instrument, or data-collection period.

The key issue is therefore not whether two questions can be investigated together. It is whether answering them together produces a coherent study.

Combine Questions When They Form One Coherent Investigation

What Makes Two Research Questions Part of the Same Study?

There is no universal numerical rule specifying how many research questions a study may contain. Research questions vary substantially across disciplines and methodologies. What matters more is alignment among the problem being investigated, the questions being asked, the evidence required to answer them, and the design used to obtain that evidence.

A useful starting point is to ask whether the two questions represent different parts of one intellectual problem or two intellectual problems that happen to be nearby.

Start With the Research Problem, Not the Dataset

Research questions are not independent items that researchers simply collect under a broad topic. They guide consequential design decisions, including what evidence is needed, who or what should be studied, which variables or phenomena require attention, and how the resulting evidence will be interpreted.

Suppose you are studying students' use of generative AI for academic writing. These two questions might plausibly belong together:

• How frequently do students use generative AI for academic writing?
• What factors are associated with differences in their use?

The second question extends the first. Both contribute to understanding the same phenomenon in the same population, and a compatible design could reasonably address both.

Now consider adding: How do faculty members redesign assessment tasks in response to generative AI? The topic is still generative AI in education, but the population, phenomenon, evidence, and likely analytical approach have changed. Thematic similarity alone does not necessarily make the questions one study.

Look for a Shared Explanatory Purpose

Two questions are especially compatible when one naturally extends the other. A study might first establish whether a phenomenon exists and then examine factors associated with it. It might estimate an effect and then investigate whether that effect differs across theoretically meaningful conditions. A qualitative study might ask how participants experience a phenomenon and what processes appear to shape those experiences.

In such cases, the questions perform different functions while contributing to one overall purpose.

This distinction matters because researchers can easily mistake topical proximity for conceptual coherence. Two questions can both concern online learning, artificial intelligence, teacher development, cancer treatment, or climate adaptation and still represent separate investigations.

Check Whether the Questions Require Compatible Study Designs

Conceptual connection is necessary, but methodological compatibility also matters. Each research question implies requirements concerning participants or cases, measurements, data, timing, comparison conditions, and analysis. A clear research question therefore helps determine the study design rather than merely describing what will eventually be analyzed.

Ask what would be required if each question were studied properly on its own. Then compare those requirements.

Compatibility does not require identical methods. Mixed-methods research, for example, may intentionally use different forms of evidence because they contribute to an integrated purpose. The question is whether the methodological differences are justified by the study's logic rather than accumulated because several interesting questions happened to arise.

Determine Whether One Question Is Primary

Multiple-question studies often benefit from a clear hierarchy. One question may define the study's central purpose while another extends, explains, or qualifies it. Methodological literature commonly distinguishes primary from secondary questions for precisely this reason: additional questions can increase design and analytical complexity and should not compromise the primary question.

This hierarchy can influence sample-size planning, outcome selection, analysis, interpretation, and reporting. In confirmatory quantitative studies, particularly those involving multiple outcomes or comparisons, adding questions may also increase the number of statistical tests and require appropriate planning for multiplicity.

A secondary question is not necessarily less interesting. It is secondary because of its role in the architecture of that particular study.

Ask Whether Each Question Can Be Answered Properly

A study does not become stronger merely by answering more questions. Each question creates obligations. You need adequate evidence, an appropriate analytical strategy, sufficient methodological justification, and enough space to interpret the findings responsibly.

The FINER criteria provide a useful general check: a research question should be feasible, interesting, novel, ethical, and relevant. When considering multiple questions, feasibility becomes particularly important because complexity can accumulate quickly through additional measures, recruitment requirements, analyses, expertise, and researcher time.

This means a pair of individually good questions can still make a poor combined study. The problem is not the quality of either question. It is the cost of answering both rigorously within one design.

Shared Participants Do Not Automatically Mean One Study

Using the same people for two questions can reduce recruitment and data-collection burden, but participants are a source of evidence rather than the intellectual boundary of a study. The same participant pool can support questions with very different purposes.

For example, university instructors might provide data about their acceptance of generative AI and, in the same survey, about occupational burnout. If the project has no conceptual reason for connecting those phenomena, collecting both sets of responses at once does not create a coherent study. Questions that merely share the same participants may still belong to different research projects.

Shared Data Do Not Automatically Mean One Research Project Either

The same principle applies to datasets. Large surveys, longitudinal datasets, institutional databases, registries, and administrative records can support many legitimate research questions. Researchers should not infer that every question answerable from one dataset belongs in one study.

A dataset is a resource. A study is an organized inquiry around a research problem. Consequently, several questions can use the same dataset while representing different research projects.

Consider Whether the Second Question Changes the Study's Center of Gravity

A useful diagnostic is to imagine removing the second question. Does the remaining study still have essentially the same purpose? Then reverse the exercise. If removing the first question leaves another complete study with its own rationale, design logic, analysis, and conclusions, you may be looking at two studies rather than one multi-question study.

This is not an infallible test, but it exposes an important distinction. A genuinely secondary question usually depends intellectually on the larger investigation. A competing primary question often has enough independence to demand its own justification.

If the added question begins to redirect the literature review, require new theoretical arguments, introduce another population or outcome structure, and occupy a substantial portion of the analysis, consider whether the secondary question is distracting from the primary question.

Testing Whether Two Questions Form One Investigation

Now imagine the researcher proposes a third question: How do university administrators develop institutional policies governing generative AI? It remains broadly related to AI in higher education, but it introduces a different population, phenomenon, evidence base, and likely methodology. Including it would require more than simply adding another variable. It begins to look like another study.

The distinction is important: one topic can contain many studies. A broad thematic umbrella should not be confused with a single research design.

Why Research Questions Get Combined for the Wrong Reasons

How to Decide Whether to Keep the Questions Together

When deciding whether two questions belong in one study, resist starting with practical convenience. Instead, work from intellectual coherence toward methodological feasibility.

You should also consider the scale of the work. A thesis or dissertation may legitimately contain several related questions because the purpose and expected scope of those projects differ from a small empirical paper. Even then, the questions need an organizing logic. A collection of loosely connected analyses does not become coherent merely because it appears between the same covers.

Likewise, a sustained set of related questions may eventually be better understood as a research program rather than a single study. Recognizing that boundary can be useful. You do not have to answer every important question at once.

Before Combining Two Research Questions, Check Their Alignment

Questions About Combining Research Questions in One Study

One Study Should Answer One Coherent Set of Questions

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.