RESEARCH / BLOG

Research,
Explained.

Read the research question, follow the authors’ methods and explore the evidence behind the article.

Research Blog

Research methodology · Research insights

Research Gap vs Research Problem: A Clear Guide for PhD Scholars

Distinguish a research problem, literature gap, question and objective with practical examples. Build a focused PhD proposal grounded in evidence and research methodology.

By NPT Research8 min read
The novelty toolkit · S201 / 09
Separate the problem from the gap.An observation starts the inquiry. Evidence defines the uncertainty.
NPT Research

The Novelty Toolkit · Series 2 · Part 1 of 9

Imagine that the ceiling in your classroom leaks whenever it rains. Everyone agrees that the leak is a problem. But does anybody know whether the water enters through damaged tiles, a blocked drain, or condensation? The problem is visible; the missing explanation is a gap in knowledge. A scientist must separate the two before deciding what to investigate.

Many dissertations collapse the problem, gap, question, objective and hypothesis into one sentence. The result sounds academic but leaves reviewers unsure what evidence is missing or why the proposed study is needed. This tutorial gives each component a distinct job.

Start with four different questions

A practical problem is an undesirable condition or unmet need: a low-cost monitor fails to give consistent readings outdoors. A research problem is an intellectually or empirically answerable uncertainty arising from that situation: the determinants of long-term measurement drift remain insufficiently understood in the specified setting. A research gap describes a supported shortfall in existing knowledge, methods or evidence. A research question asks exactly what a study will investigate to address that uncertainty.

A research problem need not originate in a practical inconvenience. It may arise from an unexplained observation, two conflicting theories, a mathematical inconsistency or an unresolved empirical relationship. Equally, a practical problem may already have a well-established scientific explanation but lack resources for implementation; that does not automatically create a novel research question.

The novelty toolkit · Series 2Give each research decision its own place.A concern becomes an investigation through evidence and scope.
  1. 01
    ObservationRecord what happened without assuming its cause.
  2. 02
    Research problemName the uncertainty worth investigating.
  3. 03
    Evidence gapExplain what relevant studies leave unresolved.
  4. 04
    Research questionSelect a specific uncertainty your study can address.
A learning framework, not a measured result. Verify evidence before making a research claim.
Figure 1. Moving from an observation to an investigation requires distinct decisions, not interchangeable labels.

Use the leaky-roof test

Suppose three engineers agree that a roof leaks. Engineer A says, 'The research gap is that the roof leaks.' That is merely a restatement of the problem. Engineer B says, 'Nobody has studied my exact classroom.' That is an absence of local publications, not necessarily a meaningful gap. Engineer C says, 'Existing drainage models do not explain why water accumulates at this roof joint during wind-driven rain; I will measure pressure and drainage under controlled conditions.' This is much closer to a researchable problem because a specific uncertainty and investigation are visible.

Try the same test on your own topic: if the supposed gap could be expressed without discussing any existing evidence, you probably have not demonstrated a gap yet.

One water-monitoring example, five different statements

Observation: inexpensive field sensors sometimes appear to disagree after prolonged outdoor use.

Practical problem: operators may struggle to trust readings when maintenance visits are infrequent.

Possible research problem: it is uncertain which mechanisms explain observed variation under the local deployment conditions.

Candidate gap: the studies retrieved so far may not directly separate sensor ageing from temperature effects while evaluating prolonged offline operation. This is an unverified candidate, not a literature conclusion.

Research question: among defined sensors and field conditions, how much of the measurement error is associated with temperature variation after controlling for sensor age and calibration history?

Objective: quantify error by temperature band, sensor age and calibration condition using predefined measurement procedures.

Observe the shift. We began with a concern and ended with an answerable question, not a grand claim that we invented a new sensor.

Learn the difference between a gap and an empty shelf

A library may have no book on 'solar-powered underwater robots in my district'. That empty shelf does not prove originality. A closely related study may use a different term, exist in a thesis, or already explain the relevant mechanism. Similarly, a location change may be worthwhile for external validity without constituting a major theoretical advance.

A defensible gap statement should name what is known, what remains uncertain, the evidence supporting that uncertainty, and why resolving it matters. The middle two elements are frequently missing.

There are several legitimate kinds of gaps: conflicting findings, missing mechanism, weak measurement validity, insufficient replication, poor generalizability, or limits in current methods. Each produces a different kind of study. For example, contradictory results may call for better controls; a missing mechanism may need an experiment; uncertain generalizability may need replication in a carefully chosen new setting.

The novelty toolkit · Series 2Similar words. Different scientific jobs.Use the right statement for the decision you are making.
  1. 01ProblemWhy does the uncertainty matter?
  2. 02GapWhat does the existing evidence not establish?
  3. 03QuestionWhat exactly will the study ask?
  4. 04ObjectiveWhat will the researcher measure or explain?
A learning framework, not a measured result. Verify evidence before making a research claim.
Figure 2. A quick desk-side distinction that prevents research statements from collapsing into each other.

A five-sentence diagnostic worksheet

Complete these sentences before writing a proposal:

  1. In [specific setting], I observe [phenomenon or concern].
  2. Existing evidence establishes [two or three findings with citations].
  3. It does not yet adequately establish [specific uncertainty and basis for saying so].
  4. Therefore, I ask [one answerable research question].
  5. The answer matters because [theoretical, methodological or practical consequence].

A statement such as 'there is little research' should be followed by the search approach used to support that assessment. The absence of an article in a single database is not enough. Likewise, a strong problem statement does not require an exaggerated 'world's first' claim.

Avoid these common traps

  • Trap: describing a societal concern as a gap. Correct it by identifying the unanswered scientific question behind the concern.
  • Trap: treating a new technology label as a discovery. Correct it by checking what the underlying method already does.
  • Trap: equating an unstudied population with novelty. Explain why that population could change the expected result or test transferability.
  • Trap: making the question too large. Replace 'How can we solve pollution?' with a bounded question using defined variables and outcomes.

Practice in fifteen minutes

Take a topic you already know. Write one sentence each for observation, research problem, candidate gap, research question and objective. Swap only the 'gap' sentence with a colleague and ask them to identify the evidence needed to justify it. If they cannot tell what papers must be checked, rewrite the gap more precisely. Now test the question against the FINER criteria: feasible, interesting, novel, ethical and relevant (Stanford’s research-question guidance).

Three quick examples across disciplines

Biomedical research: 'Many patients miss follow-up appointments' is a practical problem. 'The reasons for missed appointments in our clinic are incompletely documented' is a possible research problem. A genuine gap might concern whether previously validated explanations apply when transport constraints differ; first check what studies and records already show.

Engineering: 'Batteries degrade' is a problem statement so broad that it cannot guide a dissertation. 'The contribution of a specified charging pattern to capacity loss under defined operating temperatures is uncertain after accounting for cell variability' creates a tractable question. A novel controller is not justified until the specific uncertainty or performance limitation is demonstrated.

Education: 'Students dislike a course' is a concern. A research problem could ask how students experience a new teaching format. A qualitative gap would concern underexplored perspectives or mechanisms, not simply whether the researcher can find the exact course name online. The question may require interviews and thoughtful sampling, not a laboratory hypothesis.

A small test of scientific honesty

Imagine an examiner asks: 'If I show you one paper answering your question, what will you do?' A defensible response is to compare its context, methods and findings with the proposed work. You might discover that your question is already answered, or that a useful boundary condition remains. Either way, the paper is evidence to learn from, not an obstacle to dismiss.

Now reverse the test: 'If I show you no papers with those exact keywords, what will you do?' The correct answer is to broaden terminology, follow citations, search relevant grey literature and examine whether the question is scientifically worthwhile. Empty search results can reflect an indexing or phrasing problem, not originality.

A short self-assessment

Can you define the problem without naming your preferred solution? Can you describe the gap using evidence rather than intuition? Can a colleague infer what kind of data would answer your question? Can you name one result that might weaken your proposed explanation? Four 'yes' answers are a promising start—not proof of novelty, but evidence that your research statement has become more rigorous.

What to remember

The problem tells you why you care; the gap tells you what knowledge is missing; the question tells you what you will ask; the method tells you how you might answer. A strong dissertation links them, but never confuses them. The next tutorial begins with the most human part of this chain: noticing a problem worth investigating.

Methodological context and limits

For a methodological grounding, Farrugia and colleagues’ question–hypothesis–objective guide distinguishes the study’s question, hypothesis and objectives in clinical research. Gutiérrez Ubeda’s research-question framework offers a structured question-development process; apply it with attention to your discipline rather than treating a template as proof of novelty.

Continue the research novelty series

Series 1: originality, evidence and scientific judgment

Begin with what makes research novel, then find the closest prior research and build a novelty matrix. Before using an assistant, examine the limits of AI novelty judgments and why the researcher owns the question. Develop the conversation with scientific prompting and six-round brainstorming, then challenge the idea against evidence. The complete worked research dialogue connects those decisions.

Series 2: from a problem to a defensible proposal

First separate a research gap from a research problem and identify a genuine problem worth investigating. Next search published literature for the gap and align the question, objectives and hypothesis. For AI-assisted work, clarify the problem without outsourcing judgment, verify the literature search and its references and compare papers in an evidence matrix. Then stress-test the research hypothesis before following the complete observation-to-proposal workflow.

Sources and further reading

  1. Stanford University. Selecting Your Research Question: FINER and study planning. Institutional guidance.
  2. Farrugia et al. Research questions, hypotheses and objectives. Canadian Journal of Surgery 53, 278–281 (2010).
  3. Gutiérrez Ubeda. How to Build and Assess the Quality of Healthcare-Related Research Questions. Global Journal on Quality and Safety in Healthcare 5, 39–43 (2022). Published 2 Jun 2022