03 · What You Need to Know
Interdisciplinary Integration Has to Reach Below the Vocabulary
Every research question carries assumptions
A research question does more than identify variables or phenomena. It also implies something about what exists, what relationships are plausible, what can be observed or inferred, and what kind of answer would count as satisfactory.
Researchers do not always state these assumptions because many are ordinary within their disciplinary traditions. They become more visible when another field approaches the problem differently.
Recent work on interdisciplinary research describes disciplines as operating with differing epistemic systems that may value different research aims, methodologies, theories, concepts, forms of evidence, and other epistemic goods. These differences can generate practical conflict when researchers attempt to integrate their perspectives.
The difficulty is therefore not simply that disciplines have different jargon. They can differ in how knowledge about the problem is constructed and evaluated.
Ontology can differ: what kind of thing is the phenomenon?
Ontological assumptions concern, in broad terms, what entities or phenomena researchers treat as real and how those phenomena are constituted.
Consider “student engagement.” One framework might conceptualize engagement as an individual-level construct with behavioral, emotional, and cognitive dimensions that can be measured for each student. Another research tradition might emphasize engagement as relational and situated, emerging through interactions among learners, teachers, tasks, technologies, and institutional environments.
These perspectives may be connectable, but they are not automatically equivalent.
A research question that simultaneously treats engagement as a stable individual attribute and as something wholly constituted through context needs to explain how those positions relate. Otherwise, the study may measure one version of the phenomenon while interpreting the findings as evidence about another.
Epistemological assumptions can differ: how can we know about it?
Disciplines and research traditions can also differ in what they consider persuasive knowledge.
One approach may seek measurable regularities and treat standardized observations as a strong basis for inference. Another may regard participants' interpretations and situated meanings as indispensable because the phenomenon cannot be adequately understood independently of context.
This does not reduce to a simplistic quantitative-versus-qualitative divide. Quantitative and qualitative traditions are themselves diverse, and both can operate under different epistemological positions.
The relevant question is what kind of knowledge claim the study intends to make and whether its methods can support that claim.
Methodological difference
Researchers use different techniques to gather or analyze evidence, potentially while sharing compatible assumptions about the problem.
Epistemological tension
The approaches differ more fundamentally in what knowledge about the problem means, how claims can be justified, or what evidence should count as persuasive.
Disciplines can disagree about what counts as an explanation
Suppose a researcher asks why students use generative AI during academic tasks.
A psychological explanation might focus on beliefs, motivation, cognition, or individual decision processes. A sociological explanation might emphasize norms, institutional arrangements, social position, or patterned opportunities. An information-systems account might explain adoption through relationships among perceptions of a technology and behavioral intentions.
These accounts can potentially complement one another. But they locate explanatory importance differently.
A conceptual framework becomes unstable if it casually moves between levels, for example treating an institutional condition as though it were an individual psychological trait or interpreting an individual association as sufficient evidence of a structural explanation.
Before integrating explanations, determine whether the disciplines are addressing the same problem differently and whether those differences are complementary, competing, or grounded in deeper assumptions.
Levels of analysis can create hidden contradictions
Interdisciplinary questions frequently connect individual, group, organizational, technological, and societal levels.
Cross-level research can be valuable. The difficulty arises when constructs migrate between levels without justification.
For example, institutional “readiness” is not automatically the aggregate of individual employees' readiness. A university's culture is not simply the average attitude of its students. System reliability is not a psychological perception merely because users report it.
When concepts belong to different levels, specify the relationship explicitly. Otherwise, the question can contain what appears to be one coherent chain of variables while actually mixing units of explanation.
Theories can make incompatible causal assumptions
Two theories may use related constructs while assuming different causal structures.
One framework may treat attitudes as causes of behavior. Another may emphasize that behavior and context recursively shape attitudes. A third may reject a simple individual-level causal model because the phenomenon is understood as emerging from social practices.
Combining variables from these frameworks into one directional path model can quietly privilege one causal account while citing the others as though they support the same structure.
This is one reason interdisciplinary integration cannot be accomplished simply by collecting useful constructs from multiple theories.
Different standards of evidence can pull the design in opposite directions
Interdisciplinary conflict can also arise over rigor itself.
Research on disciplinary epistemological differences has documented disagreements concerning facts, causal explanation, research goals, and standards of rigor. More recent work similarly argues that disciplines may value different methodologies, aims, and epistemic goods.
Suppose one part of a project requires estimating whether an intervention produces an average effect, while another seeks to understand how participants interpret and enact that intervention differently across contexts. These are not necessarily contradictory aims. They become problematic if the study treats one form of evidence as capable of answering the other's question without justification.
Integration may require preserving different kinds of evidence while explaining how their claims relate rather than forcing them into one evidentiary hierarchy.
A method can silently import one discipline's assumptions
Researchers sometimes believe they have created a balanced interdisciplinary framework because the literature review includes several perspectives. Then the method operationalizes the problem entirely according to one of them.
For example, a framework may describe a phenomenon as dynamic, relational, and context-dependent, but the study measures it once as a stable individual score and interprets that score as representing the full construct.
The contradiction is not necessarily that standardized measurement is inappropriate. It is that the operationalization may no longer correspond to the phenomenon the conceptual framework claims to study.
Watch Out
Check whether the research method quietly resolves a disciplinary disagreement by adopting one field's definition of the phenomenon. If so, make that choice explicit and ensure the resulting claims do not continue to rely on incompatible definitions from the other perspective.
Not every disciplinary difference needs to be reconciled away
A common reaction to conflicting assumptions is to search for a compromise framework in which everybody agrees.
That may not always be necessary or desirable.
Interdisciplinary scholarship has debated the tension between integration and epistemic pluralism. Talbi and van Woerden argue that interdisciplinary work can face conflict precisely because different disciplines bring distinct epistemic systems, while meaningful pluralism may require retaining rather than eliminating some differences.
The objective can therefore be to make assumptions explicit, establish enough common ground for the research question, and specify where differences remain consequential.
Integration does not have to mean pretending that disciplinary disagreements have disappeared.
Sometimes the inconsistency is the research problem
An especially productive possibility is that conflicting disciplinary assumptions become something to investigate rather than something to hide.
Suppose one discipline treats successful educational technology adoption as frequent continued use, while another questions whether continued use indicates educational benefit at all. Instead of selecting one definition silently, the researcher might ask under what conditions continued use corresponds with meaningful learning outcomes.
The disagreement has now generated a sharper interdisciplinary question.
This is one reason a research gap can emerge between disciplines. What is missing may be evidence capable of clarifying the relationship between two disciplinary accounts.
Choosing a primary framework does not erase incompatible assumptions
A study may legitimately have a conceptual anchor. But assigning one discipline primary status does not make contradictions disappear.
If another disciplinary contribution is genuinely necessary, its assumptions still need to fit the explanation or be handled explicitly. Otherwise, the secondary perspective becomes a source of concepts stripped from the intellectual conditions that gave them meaning.
The decision about which discipline should provide the primary conceptual framework should therefore occur alongside an assumptions audit rather than before it.
Look for common ground rather than superficial agreement
Work on interdisciplinary integration has emphasized the importance of establishing common ground among disciplinary insights. Common ground does not require the perspectives to become identical. It provides enough shared conceptual structure for their contributions to be related coherently.
You might clarify that two disciplines use different definitions but agree on a narrower observable phenomenon. You might specify that one explains individual variation while another explains contextual conditions. You might retain different standards of evidence because they answer different subquestions.
The goal is to know exactly what has been integrated and what has not.
