Interdisciplinary work is now routine. Deep interdisciplinary understanding is not.
When scientists from different fields disagree, the disagreement is often framed as a failure of rigor or competence. In practice, it more often reflects different assumptions about what good evidence looks like, which risks are most serious, and which kinds of uncertainty are tolerable.
These assumptions are rarely explicit. They are learned through training, peer review, and informal norms. When they collide, scientists can talk past each other without realizing they are using different evidentiary yardsticks.
Every discipline develops its own evidentiary norms in response to the problems it can realistically study.
For example:
None of these standards are arbitrary. They are pragmatic adaptations to constraint.
Problems arise when these context-specific standards are treated as universal.
A single study can inspire confidence in one field and skepticism in another.
This often reflects different views about which errors matter most. Some disciplines prioritize avoiding false positives. Others worry more about missing real effects. These priorities influence acceptable sample sizes, tolerance for noise, and standards of evidence.
When scientists argue about rigor, they are often arguing about risk management, not quality.
Formal hierarchies of evidence can be useful teaching tools. They become misleading when applied without judgment.
A randomized trial answers some questions extremely well and others poorly. Large observational datasets can reveal patterns that experiments cannot. Qualitative work can surface mechanisms that quantitative studies depend on.
Good evidence is not evidence that ranks highest on a pyramid. It is evidence that is well matched to the question being asked.
Even core concepts like replication shift across disciplines.
In some fields, replication means repeating an experiment under identical conditions. In others, it means:
Disagreements about replication often persist because participants assume a shared definition that does not exist.
Another common source of confusion is the role of causality.
In some domains, causal inference is the central objective. In others, the goal is prediction, description, or system characterization. Problems arise when causal standards are imposed where they do not fit, or when causal language is used loosely where it should be constrained.
The question is not whether causality matters. It is whether it is the appropriate standard for the claim being made.
Evidence disputes often feel sharper than they need to be because methods are tied to professional identity.
Questioning a field’s evidentiary norms can sound like questioning its legitimacy. As a result, methodological disagreements can quickly become personal, even when no such intent exists.
Recognizing this dynamic can defuse conflict without lowering standards.
When engaging with research from outside your discipline, a useful shift is to ask different questions.
Instead of asking whether the evidence would convince your field’s reviewers, ask:
This reframing often reveals that what looks like weak evidence is actually evidence optimized for a different risk profile.
Good interdisciplinary work makes its translations explicit.
That often means:
These practices strengthen arguments rather than dilute them.
Rigor is not uniformity. It is alignment between question, method, and inference.
When scientists talk past each other, it is rarely because one side does not value evidence. It is because they value different protections against error.
Making those differences visible is not a compromise of rigor. It is a prerequisite for serious interdisciplinary science.
Have our latest blogs, stories, insights and resources straight to your inbox
