Should You Avoid a Research Question Because the Expected Result Seems Obvious?

If You Already Know What You Expect to Find, Why Study It?

Some research questions produce an immediate reaction: Isn't the answer obvious?

Students who study more should perform better. Employees who receive more support should be more satisfied. A useful technology should improve productivity. People with better access to information should make better decisions. If the expected direction seems so intuitive, conducting a study can feel like spending months collecting evidence for something everyone already knew.

But “obvious” can mean several very different things. The answer may be strongly established by previous evidence. It may merely seem plausible. It may be conventional wisdom that has rarely been tested. Or the broad relationship may be familiar while its magnitude, mechanism, boundary conditions, or applicability to the population you care about remains uncertain.

So the relevant question is not simply whether you can predict the result. It is whether meaningful uncertainty remains despite that prediction.

An Expected Answer Is Not Necessarily an Established Answer

Ask Why the Answer Seems Obvious Before Deciding Not to Study It

“Obvious” is not an evidential category

Researchers use the word obvious loosely. Sometimes it means that a claim follows naturally from a well-supported theory and substantial empirical evidence. Sometimes it means that the claim sounds reasonable. Those are not equivalent.

Consider the statement “students learn better when they are motivated.” It is intuitively appealing, but as a research proposition it is underspecified. What kind of motivation? What measure of learning? Over what period? Under which instructional conditions? Is the relationship causal? How large is it? Does it remain after accounting for relevant differences between students?

An intuitive statement can become considerably less obvious once it is converted into a precise empirical question.

Research is often conducted precisely because researchers have an expectation. A hypothesis would be rather lonely without one. Predictability itself is therefore not a reason to reject a question.

Start by identifying where your expectation comes from

Before asking whether an expected result deserves testing, identify the basis for your confidence.

Perhaps numerous rigorous studies have already produced consistent evidence. Perhaps a well-developed theory predicts the outcome. Perhaps you have observed the pattern repeatedly in professional practice. Or perhaps the conclusion simply feels like common sense.

Those foundations carry different evidential weight.

This exercise often reveals that the real question is not whether the expected answer is obvious, but whether the evidence supporting it is already adequate.

Previous research can look more settled than it actually is

A conclusion may appear established because many papers repeat it, yet the evidence beneath those statements can vary considerably in quality.

NIH guidance on rigor and reproducibility explicitly asks researchers to examine the strengths and weaknesses of prior research supporting a proposed project. That includes considering whether earlier experimental designs were sufficiently rigorous and whether identified weaknesses or gaps need to be addressed.

This principle matters well beyond biomedical grant applications. Before declaring a question too obvious to study, examine the evidence that supposedly makes the answer obvious.

Were previous studies small? Were the measures appropriate? Were findings replicated independently? Were samples narrowly drawn? Are estimates precise? Do studies agree about magnitude as well as direction? Are there methodological weaknesses that could systematically influence the result?

An expected conclusion resting on fragile evidence can remain an important research question.

Replication is useful precisely when you already have an expected result

Replication provides perhaps the clearest counterexample to the idea that predictable results are scientifically uninteresting.

A replication usually begins with a result that has already been reported. Researchers therefore have an explicit reason to expect what they may find. The purpose is not necessarily to surprise anyone. It is to determine whether the finding is sufficiently consistent when the scientific question is examined again using new data.

The National Academies defines replicability, for the purposes of its report on reproducibility and replicability, as obtaining consistent results across studies aimed at answering the same scientific question using their own data. It also emphasizes that replication contributes to the self-correcting nature of science by allowing previous findings and inferences to be tested again.

NIH has likewise emphasized replication and reproducibility as foundational to rigorous science and, in 2026, announced an agency-wide initiative intended to strengthen and incentivize such work.

Replication therefore illustrates a broader principle: confirmation can be informative when confidence in the claim is itself scientifically important.

But repeating something is not automatically a useful replication

Recognition of replication does not mean that every repetition is worthwhile.

Suppose dozens of rigorous, independent studies already establish a relationship across the populations and conditions that matter, with reasonably precise estimates and no important unresolved methodological concern. Repeating essentially the same study once more may produce little additional information.

The relevant issue is marginal contribution.

Would the new study materially change confidence in the result? Test an important population? Address a weakness in previous research? Examine robustness under a meaningful condition? Improve precision? Resolve conflicting findings?

If not, the study may merely reproduce an expected result without contributing enough to justify the effort.

This is the same distinction involved in deciding whether a very small contribution is still worth making. Incremental research can be useful, but “another study exists” is not itself a contribution.

The direction may be obvious while the magnitude is not

Researchers sometimes dismiss a question because everyone expects the direction of the relationship. Yet direction is only one part of an empirical result.

Suppose everyone expects a particular educational intervention to improve performance. Even if that prediction is correct, important questions remain.

Does performance improve by a negligible amount or a substantial one? How precise is the estimate? Does the improvement justify the intervention's cost and effort? Is the effect similar across students? Does it persist over time? Are there unintended consequences?

A result can therefore be predictable in direction while highly uncertain in magnitude and practical significance.

Moving from the first question to the second can turn an apparently obvious study into a more informative one.

An obvious relationship may conceal an uncertain mechanism

Sometimes researchers are reasonably confident that two things are related but do not know why.

Imagine strong evidence that timely feedback is associated with better student performance. A new study asking only whether timely feedback is again associated with performance may add little. But researchers might still disagree about the mechanism.

Does timely feedback help because students correct misconceptions sooner? Because it increases engagement? Because it changes study behavior? Because students receiving timely feedback differ systematically in some other way?

The broad pattern can be familiar while the explanation remains unsettled.

A more useful research question may therefore shift from whether the expected relationship occurs to how or why it occurs.

The expected result may fail under conditions nobody has tested

A claim that is well supported in one setting does not automatically apply under every relevant condition.

The National Academies distinguishes replicability from generalizability, defining the latter as the extent to which study results apply in other contexts or populations that differ from the original one. This distinction matters when a seemingly obvious result is being extended beyond the conditions under which it was established.

Suppose an instructional technique reliably improves learning in highly structured laboratory tasks. Will the same advantage appear in authentic courses where students have competing demands, variable prior knowledge, and less controlled exposure?

You may expect that it will. Testing that expectation can still matter if the new conditions represent a meaningful boundary of existing evidence.

This is also why simply moving a study elsewhere is not enough. As discussed in the guide on whether a local research question has broader value, a new setting contributes when the contextual difference addresses a consequential uncertainty, not merely because the location has changed.

Common sense can be an especially unreliable reason to stop asking questions

Some claims seem obvious because they fit familiar narratives.

Technology saves time. More choice improves satisfaction. Smaller classes improve learning. More information improves decisions. Flexible work improves employee well-being.

Each statement can sound plausible while concealing conditions, trade-offs, nonlinear relationships, measurement problems, or competing mechanisms.

Common sense can be useful for generating hypotheses. It is weaker as a substitute for evidence.

Indeed, opposing predictions can sometimes both sound obvious after the result is known. A flexible policy improves satisfaction because employees gain autonomy. Or it reduces satisfaction because boundaries become blurred. Both stories can sound intuitively persuasive.

A research question becomes stronger when the expected result follows from an explicit theoretical or empirical rationale that could, in principle, be wrong.

A predictable result can still challenge an exaggerated claim

Sometimes the contribution lies not in discovering that an effect exists, but in establishing that it is smaller or more conditional than people assume.

Imagine a widely promoted technology that almost everyone expects to improve productivity. A rigorous study may indeed find improvement, but only under certain tasks and by a modest amount.

The direction of the result was predictable. The evidence still matters because it replaces an imprecise claim such as “this improves productivity” with a more defensible conclusion about magnitude and conditions.

This is especially important when practical decisions depend on whether an effect is large enough to matter, not merely whether its estimated direction is positive.

Your study should remain informative if the obvious result does not appear

One of the strongest tests of the question is to imagine the expected result fails to appear.

Would that be informative?

If a well-supported theory predicts a relationship and a rigorous study fails to observe it under conditions where it should occur, the result may expose a boundary condition, measurement problem, methodological issue, or theoretical weakness.

If the unexpected result would be dismissed immediately as “the study must be wrong because everyone knows the answer,” then the hypothesis may not be functioning as a genuinely testable proposition.

Do not equate statistical significance with proof of the obvious

A predictable hypothesis can also encourage a mechanical research strategy: collect data, obtain a statistically significant result in the expected direction, and declare the obvious confirmed.

That is rarely the most informative use of the evidence.

Researchers should consider effect estimates, uncertainty, design quality, assumptions, measurement, alternative explanations, and the relationship between the observed result and previous evidence. A statistically significant result does not establish that an effect is large, practically important, causal, or universally applicable.

Likewise, a non-significant result does not automatically prove that nothing happens.

The latter issue deserves separate treatment because research can be worthwhile even when you expect to find no difference.

Sometimes the obvious answer means you should ask a better question

Not every apparently obvious question needs to be rescued.

If strong evidence already establishes the broad relationship, asking the same broad question again may indeed be unnecessary. The appropriate response may be to move one level deeper.

The problem may therefore be the level at which the question is framed rather than the topic itself.

Predictability can be a strength when theory makes a precise prediction

There is another reason not to equate surprise with scientific value.

Strong theories should generate expectations. If a theory makes a clear prediction and the evidence supports it under a demanding test, the fact that the result was predicted does not make the study pointless. Prediction is part of what gives the test intellectual structure.

The question is whether the test was informative. Did competing explanations make different predictions? Were the conditions capable of exposing failure? Did the result reduce meaningful uncertainty?

A study can therefore be unsurprising and scientifically useful at the same time.

Surprising research is not automatically better research

The mirror-image mistake is to prefer a question merely because its possible result sounds counterintuitive.

Surprise attracts attention, but surprising findings can arise from noise, analytical flexibility, measurement error, unusual samples, or genuine phenomena. Their novelty does not establish their reliability or importance.

Choosing questions according to how surprising the result might look can also recreate the publication incentives discussed in the guide on whether publishability should determine what you study.

Research should not become a competition to produce the least expected headline.

The final test is marginal knowledge gain

Before rejecting or accepting an obvious-looking question, compare what is reasonably known before the proposed study with what could be known afterward.

If the study confirms an expected result, would confidence increase meaningfully? Would the magnitude become clearer? Would an important population become represented? Would a theoretical prediction receive a stronger test? Would an unresolved boundary become clearer?

If the expected result fails, would that challenge an important assumption or redirect subsequent research?

If neither outcome changes much, the question may genuinely be too settled to deserve another study.

The criterion is not surprise. It is how much useful uncertainty the study can remove.

When an Obvious Answer Still Leaves an Important Question

The broad prediction remains intuitive. What makes the study worthwhile is not proving that intuition can guess the direction correctly. It is resolving uncertainty that remains underneath the intuitive claim.

Now imagine that numerous rigorous studies have already compared the same forms of feedback, in similar students, using comparable tasks and measures, with highly consistent findings. Repeating the study solely because it has never been conducted at the researcher's own university would be much harder to justify.

The expected result did not become the problem. The remaining uncertainty disappeared.

Common Mistakes About Predictable Research Findings

Do Not Ask Whether the Result Is Obvious; Ask What Remains Uncertain

When someone tells you that your expected result is obvious, resist both immediate reactions: abandoning the question and defending it merely because “research is still needed.”

Inspect the evidence.

A particularly useful test is to complete two sentences before committing to the project:

“I expect the answer to be ______ because ______.”

“The study is still necessary because we remain uncertain about ______.”

If you can complete both precisely, the predictable result may not be a problem at all.

If the second sentence collapses into “because nobody has studied it exactly this way before,” reconsider the contribution.

The broader task is to balance scientific importance, practical relevance, and personal interest without adding surprise as an artificial fourth requirement. Good research does not need to astonish you. It needs to tell you something worth knowing with evidence you did not previously have.

Before Rejecting a Question Because the Answer Seems Obvious

Questions About Research With Predictable Results

Do Not Confuse Predictability With Knowledge

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.