Can a Theory That Seems Wrong Be a Better Starting Point Than a Missing Literature Gap?

What if the interesting problem is not what researchers have missed, but what they may have explained incorrectly?

You read the literature expecting to find an empty space: a population nobody has studied, a variable nobody has measured, or a relationship nobody has tested. Instead, you find plenty of research.

But something about the explanation bothers you.

An established theory predicts one pattern while credible observations suggest another. A mechanism that researchers routinely invoke does not seem capable of explaining an important case. A theory works under some conditions but repeatedly struggles under others. Perhaps a competing explanation seems to account for the evidence more convincingly.

This can be a stronger research starting point than discovering that “few studies have examined X.” Research does not advance only by filling empty spaces. It can also advance by asking whether existing explanations are adequate. The challenge is that thinking a theory is wrong is not itself evidence that it is wrong. You need to identify precisely what the theory predicts, where the difficulty lies, and what evidence could distinguish the theory from credible alternatives.

Yes, when the theoretical problem is real, consequential, and empirically testable

A research opportunity can come from an inadequate explanation, not just missing evidence

A literature gap and a theoretical problem are not the same thing

A literature gap generally refers to something important that remains insufficiently addressed in the existing body of research. That absence can take different forms: a phenomenon may be understudied, evidence may be inconsistent, a population may be poorly represented, a method may be inadequate, or an important question may remain unanswered.

A theoretical problem is more specific. Existing research may be abundant, but the explanation used to make sense of the phenomenon appears incomplete, inaccurate, overly broad, or unable to account for important evidence.

These categories can overlap. Conflicting studies might expose a theoretical boundary condition. A methodological improvement might reveal that a familiar theoretical relationship is weaker than previously believed. A new technology might create circumstances the theory did not anticipate.

The useful question is not which label sounds more sophisticated. It is what kind of uncertainty actually exists.

An empty space in the literature is not automatically an important problem

Suppose nobody has examined whether a familiar theoretical relationship holds among students at a particular university. That is technically an absence in the literature. But why should anyone expect the relationship to differ there?

The fact that something has not been studied does not establish that studying it would materially improve knowledge. Some absences exist because the question is unimportant, redundant, infeasible, or already answerable from closely related evidence.

By contrast, evidence that a widely used explanation repeatedly fails under theoretically important conditions can create a more consequential question because resolving it may change how a larger body of evidence is understood.

This is why the starting point matters less than the intellectual problem it reveals.

Before challenging a theory, make sure you understand it

The easiest theory to refute is a theory nobody actually proposed.

Before arguing that an explanation fails, return to its foundational sources and important later developments. Determine the theory's intended scope, definitions, assumptions, mechanisms, and predictions. Check whether later researchers have already revised the theory in response to the problem you noticed.

You may discover that the apparent contradiction disappears because the theory never claimed to apply under those conditions. Alternatively, you may find that researchers routinely use the theory more broadly than its original scope supports. That itself can become an interesting problem, but it is different from demonstrating that the original theory is wrong.

Specify what observation is difficult for the theory to explain

“I don't think this theory works” is too vague to generate a rigorous study.

A stronger argument takes the form:

The theory predicts or implies ________. Under ________ conditions, however, we observe or have credible reason to expect ________. These cannot both be explained easily under the current account because ________.

This identifies the point of theoretical tension.

For example, a theory might predict that greater perceived usefulness increases voluntary adoption of a technology. But imagine a professional setting in which use becomes mandatory. High use under those conditions cannot necessarily be interpreted as evidence that perceived usefulness drove adoption because the decision process has changed.

The theoretical problem concerns the mechanism and scope, not simply whether a familiar correlation reaches statistical significance.

Try to explain the apparent failure without changing the theory first

When evidence conflicts with a theoretical prediction, the theory is only one possible source of the problem.

The measurement may poorly represent a construct. The sample may be unusual. Implementation may differ from what the theory assumes. Statistical estimates may be imprecise. The relevant mechanism may require conditions absent from the study. A result may be exploratory or unstable.

These alternatives should be taken seriously. If a methodological limitation explains the apparent theoretical failure, fixing the method may be more informative than immediately proposing theoretical revision.

A stronger challenge usually includes an alternative explanation

Showing that one theory has difficulty with an observation is useful. Explaining the observation better is often more informative.

John Platt's account of strong inference emphasizes the value of formulating alternative hypotheses and designing tests that can discriminate among them. This changes the research problem from:

“Is Theory A wrong?”

to:

“Does the evidence better support Explanation A or Explanation B under conditions where their predictions differ?”

That is a much more demanding question. It also protects the study from becoming an exercise in collecting evidence against a theory you have already decided to dislike.

Look for conditions where competing explanations make different predictions

Suppose two theories both predict that students who receive more feedback will revise more. Observing greater revision does not distinguish them.

But Theory A predicts that feedback works primarily by increasing information about errors, while Theory B predicts that its effect depends on students actively evaluating and interpreting the feedback. A study could create or observe conditions where the amount of feedback remains similar but opportunities for evaluation differ.

If the theories make divergent predictions under those conditions, the study becomes more informative.

A theory can be useful precisely because it can be challenged

Scientific explanations gain value partly by making claims that expose them to empirical evaluation. The National Academies describes scientific theories as predictive and provisionally accepted, with research continually testing whether observations are consistent with theoretical expectations.

Philosophical accounts of falsifiability have likewise emphasized that scientific statements must, in principle, be capable of conflicting with possible observations. Contemporary scientific practice is more complicated than a simple rule that one contrary observation destroys a theory, but the underlying principle remains useful: a theory that accommodates every possible result provides little leverage for distinguishing explanations.

If every finding in your study can be narrated as confirmation, the theory is not doing much empirical work in the design.

The goal does not have to be destroying the theory

The language of “proving a theory wrong” can make theoretical research sound like intellectual combat. More often, useful research produces refinement.

You may discover that a theory works well only when participation is voluntary, when information is scarce, when a particular mechanism is present, or among populations with certain characteristics. The theory becomes narrower but more precise.

Alternatively, evidence may suggest that a proposed mediator is unnecessary, that another mechanism needs to be incorporated, or that two apparently competing theories explain different stages of the same process.

A study can therefore challenge an explanation without requiring a ceremonial theoretical funeral.

Negative evidence can be theoretically productive

A well-designed study that fails to support a theoretically important prediction can generate new questions, particularly when the evidence is sufficiently informative to make the discrepancy difficult to dismiss.

The next step may involve examining boundary conditions, alternative mechanisms, measurement, or competing theories. This connects theoretical problem finding with the broader possibility that a null result can generate a new research question.

The important issue is not whether the result is statistically non-significant. It is whether the evidence meaningfully challenges what the theoretical account led researchers to expect.

Sometimes the literature gap and theoretical problem reinforce each other

You do not have to choose permanently between gap-driven and theory-driven research. A strong research rationale can contain both.

Perhaps a theory predicts a relationship under a particular condition, observational evidence suggests the prediction may fail, and surprisingly little research has directly tested that theoretical boundary. The missing empirical evidence then matters because of the theoretical problem.

This is stronger than identifying an empty space first and searching afterward for a theory to attach to it.

The more general lesson is that good research ideas can originate from several kinds of intellectual tension. A missing study is one possibility. An explanation that no longer seems adequate is another.

When an established explanation stops fitting the behavior

The research does not begin from “nobody has studied this software in my organization.” It begins from a more consequential question about whether changed technological conditions alter the explanatory mechanism of an established theory.

Challenging theory requires more than finding an inconvenient result

Choose the starting point that creates the sharper intellectual problem

You do not need to decide that every study must begin either with a literature gap or with a theoretical challenge. Instead, ask which route produces a question whose answer would meaningfully change what researchers know.

Try writing both rationales:

Gap version: “Previous research has not adequately examined ________.”

Theory version: “Existing theory predicts ________, but cannot readily explain ________. Distinguishing between ________ and ________ would clarify ________.”

Then ask which version explains more clearly why the study needs to exist. Sometimes the gap wins. Sometimes the theoretical tension wins by quite a distance. Peer reviewers, inconveniently, are allowed to notice the difference.

Before building a study around a theory you suspect is wrong

Common questions about theory problems and literature gaps

An inadequate explanation can be a more important gap than an empty space

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.