When Does a Research Question Actually Require More Than One Discipline?

Does Your Research Question Really Need More Than One Discipline?

Some research topics seem interdisciplinary almost automatically. Artificial intelligence in education appears to involve computing and education. Climate adaptation may involve environmental science, economics, public policy, and sociology. Digital health can bring together medicine, information technology, psychology, and design.

But a topic touching several disciplines does not necessarily produce an interdisciplinary research question. The more useful question is whether the answer requires more than one discipline.

This distinction matters because interdisciplinarity introduces intellectual and practical demands. Researchers may need to reconcile different concepts, assumptions, terminology, methods, standards of evidence, and forms of explanation. Those demands can be worthwhile when the research problem requires them. Adding disciplines merely because they seem relevant, however, may make a study harder to design without making its answer appreciably better.

A Question Needs Multiple Disciplines When One Cannot Adequately Answer It

Interdisciplinarity Should Follow the Problem, Not Precede It

Start with what the question demands

The National Academies defines interdisciplinary research as research that integrates information, data, techniques, tools, perspectives, concepts, or theories from two or more disciplines or bodies of specialized knowledge. Importantly, the purpose of that integration is to advance understanding or address problems whose solutions extend beyond the scope of a single discipline or research practice.

That last part provides a useful starting point. Interdisciplinarity is not established merely by counting disciplines represented in a proposal, author list, literature review, or set of keywords. The intellectual justification comes from the problem itself.

A researcher therefore should not begin with, “How can I add another discipline to this study?” A better starting question is, “What must be understood to answer this research question adequately?” Only then should you ask which bodies of disciplinary knowledge are needed.

A complex topic is not automatically an interdisciplinary question

Complexity often motivates interdisciplinary research, and major accounts of interdisciplinarity explicitly connect it with questions that exceed what a single discipline can adequately address. Yet complexity by itself is not a sufficient test.

Consider social media use among university students. The topic could interest psychologists, communication scholars, educators, sociologists, computer scientists, and public-health researchers. That does not mean every study of social media use requires all, or even several, of those disciplines.

If your question asks whether frequency of social media use predicts academic procrastination among university students, an appropriately designed study might be conducted within one disciplinary tradition. If instead you ask how platform recommendation mechanisms, students' self-regulation, peer norms, and institutional learning practices interact to shape procrastination, the explanatory burden changes. Different components of the phenomenon may now require bodies of knowledge that developed in substantially different disciplinary traditions.

The topic did not become more interdisciplinary. The question changed.

Look for explanatory dependence, not merely relevance

Many disciplines can be relevant to a problem without being necessary to a particular research question. Relevance is therefore a weak criterion.

Suppose you are studying students' adoption of generative AI. Computer science is relevant because generative AI is a computational technology. Law may be relevant because data protection and intellectual-property issues exist. Psychology may be relevant because students form attitudes and behavioral intentions. Education is relevant because the technology is being used for learning.

None of this demonstrates that your particular research question requires all four disciplines.

A stronger test is explanatory dependence: does answering one essential part of the question depend on knowledge that another required part cannot provide?

This distinction can prevent a common form of disciplinary accumulation in which researchers keep adding perspectives because each seems interesting. Interesting is not the same as necessary.

Ask whether the question contains different kinds of explanation

One sign that a question may genuinely require multiple disciplines is that it asks about relationships among phenomena operating through different mechanisms or levels of analysis.

For example, imagine a question about why students continue using an AI tutoring system despite sometimes receiving inaccurate explanations. One part of the answer might concern properties of the technological system. Another might concern how learners judge credibility, regulate their learning, or develop trust. If the question explicitly seeks to explain the interaction between those mechanisms, treating either the technological or learning-related component as a black box may substantially weaken the explanation.

Similar situations arise when a question connects biological and social mechanisms, individual behavior and institutional structures, environmental processes and economic decisions, or technological design and human practice.

This does not mean that different levels of analysis automatically correspond neatly to different disciplines. Disciplinary boundaries overlap, and individual disciplines themselves can contain substantial theoretical and methodological diversity. The relevant question remains whether the necessary explanatory resources are available within one coherent disciplinary approach.

Ask whether one discipline can absorb the supposedly external contribution

Disciplinary boundaries are not impermeable. Methods, concepts, and tools routinely move between fields. The National Academies notes that techniques borrowed from another field may eventually become so thoroughly assimilated that researchers no longer regard their use as interdisciplinary.

This creates an important complication. Using machine learning does not automatically make an education study interdisciplinary with computer science. Using geographic information systems does not necessarily turn a public-health project into geography research. Conducting statistical modelling does not mean statistics has become a second disciplinary foundation of every study that estimates a regression model.

The question is whether you are simply using a tool that is already part of your field's research repertoire or whether the study requires substantive knowledge from another disciplinary tradition to formulate, interpret, or integrate the answer.

Check whether the disciplines must actually interact

There is also a difference between several disciplines contributing separately and disciplinary insights being integrated.

The OECD distinguishes multidisciplinary research, in which disciplines may work in parallel while retaining their own approaches, from interdisciplinary research that crosses disciplinary boundaries to create integrated knowledge toward a common goal. The National Academies makes a similar distinction, describing multidisciplinary work as involving separate disciplinary contributions.

Imagine a project investigating an educational technology intervention. An education researcher evaluates learning outcomes, a computer scientist reports system performance, and a psychologist measures learner motivation. If each produces a separate analysis and the three findings simply appear beside one another, the project may be multidisciplinary rather than genuinely interdisciplinary.

If the research question instead requires explaining how system behavior changes learners' motivational responses and how those responses subsequently influence learning, the contributions have to meet somewhere. That need for integration is much stronger evidence that the question itself is interdisciplinary.

This distinction becomes particularly important when deciding whether an idea is genuinely integrated rather than merely multidisciplinary.

Use the removal test

One practical way to evaluate necessity is to remove each proposed disciplinary contribution mentally and inspect what happens to the question.

Necessity does not mean equal disciplinary weight

An interdisciplinary question does not require every discipline to contribute equally. One discipline may provide the central conceptual framework while another contributes a necessary explanatory mechanism, specialized method, or body of evidence.

This matters because researchers sometimes equate interdisciplinarity with symmetry. They may try to give every participating discipline equal theoretical space even when the question does not warrant it.

A study can remain interdisciplinary while having a clear intellectual center. The harder issue is determining which discipline should provide the primary conceptual framework without reducing the others to decorative citations.

Check whether the disciplines conceptualize the problem differently

Sometimes the need for interdisciplinarity becomes visible only after reading across fields. Two disciplines may study what appears to be the same phenomenon while defining it differently, explaining it through different mechanisms, or treating different outcomes as important.

Those differences can be productive. They can also expose incompatibilities.

Before combining frameworks, determine whether the fields are genuinely offering complementary accounts or whether they are addressing the same research problem differently. If their basic assumptions conflict, combining terminology from both will not resolve the disagreement. In some cases, those different disciplinary assumptions can make a question internally inconsistent.

Sometimes the apparent interdisciplinary need disappears after a better literature search

A question can appear to require a second discipline because the first literature you encounter does not contain the concepts or evidence you need. Before concluding that you have discovered an interdisciplinary gap, search more broadly.

Another field may already have developed the explanation you are seeking. Conversely, the supposedly missing idea may already exist within your own discipline under different terminology.

This is particularly important because disciplines often develop distinct vocabularies for related phenomena. Searching only your familiar terms may therefore produce a misleading picture of what is known. When terminology differs substantially, you may need to search for the problem across different disciplinary vocabularies before deciding what kind of integration the question requires.

Testing Whether an AI-in-Education Question Is Genuinely Interdisciplinary

That label is premature. Consider how different versions of the question change what the study requires.

Now apply the removal test to the third question. If the technological component is removed, the study may describe learner susceptibility but cannot adequately explain how characteristics of generated responses contribute to that susceptibility. If the learning component is removed, the study may characterize system outputs but cannot explain why students interpret those outputs as credible.

The interdisciplinary case therefore comes not from the presence of “AI” and “students” in the same sentence. It comes from the explanatory relationship the question asks the researcher to investigate.

Even then, the project should remain bounded. Once several disciplines become relevant, it is tempting to add ethics, institutional policy, socioeconomic inequality, interface design, teacher practice, and every other legitimate dimension of the problem. A defensible interdisciplinary question still needs limits, which is why researchers should consider how to prevent an interdisciplinary question from becoming impossibly broad.

Common Ways Researchers Mistake Topic Breadth for Interdisciplinarity

Decide What the Question Needs Before Deciding That It Is Interdisciplinary

When developing a research question, resist the urge to establish interdisciplinarity as a goal in itself. Begin by defining the phenomenon and the explanatory claim you want the study to make. Then identify what knowledge would be necessary to support that claim.

This approach also changes how you think about collaborators. Once the question has revealed which forms of expertise are genuinely necessary, you can judge whether you possess sufficient competence yourself or whether the study requires collaborators with deeper disciplinary expertise. That decision should ideally occur before theoretical and methodological choices become difficult to revise, especially when considering whether to build an interdisciplinary study before relevant collaborators are involved.

Most importantly, do not penalize a good research question for being disciplinary. Single-discipline research remains appropriate for many scientific questions, and NIH guidance on team science similarly recognizes that traditional single-investigator approaches are well suited to many scientific endeavors. Interdisciplinarity becomes valuable when it responds to an intellectual need rather than an aesthetic preference for broader-looking research.

Before Calling Your Research Question Interdisciplinary, Check This

Questions Researchers Ask About When Interdisciplinarity Is Necessary

Use Multiple Disciplines Because the Answer Requires Them

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.