# How to Write a Problem Statement for a Thesis or Research Proposal

> How to write a problem statement for a thesis or research proposal: what it must contain, how it differs from the gap, rules from five universities and eight steps.

Canonical: https://www.thesisroom.com/blog/how-to-write-a-problem-statement
Last updated: 2026-11-13

Published: 2026-11-13

A research problem statement names the issue your study will investigate: the [USC Libraries research guide](https://libguides.usc.edu/writingguide/introduction/researchproblem) calls a research problem "a definite, clear expression [statement] about an area of concern." Here is how to write a problem statement: say what is going wrong, prove it with current sources, show who it affects and what follows, then state what is still unknown.

This guide is about the research problem statement in Chapter 1 of a Master's thesis, a PhD dissertation or a research proposal, not the business or design kind. Rules differ between universities, so the named rules below are examples to check against your own program's handbook.

## What must a problem statement contain?

At minimum: the problem, evidence that it exists, its consequences and what is unknown. That is the four-part list in [National University's dissertation guide](https://resources.nu.edu/c.php?g=1446805&p=10752286). Ellis and Levy's [problem statement template](https://doi.org/10.28945/438), written for novice researchers, adds the cause the literature points to.

| Component | What it answers | Where the rule comes from |
|---|---|---|
| The problem | What is going wrong, in one or two sentences | National University: "Specify and describe the problem (with appropriate citations)". Ellis and Levy: "In no more than two sentences" |
| Evidence | Who says so, and what did they find | National University: "Provide evidence of the problem's existence". Ellis and Levy ask for three current, peer-reviewed references for each part |
| Impact | Who is affected, where and when | Ellis and Levy's "How, Where, and When" question |
| Consequences | What happens if nothing changes | National University: "Explain the consequences of NOT solving the problem" |
| Cause | What the literature says lies behind it | Ellis and Levy's "Why" question: the conceptual basis for the problem |
| Gap | What is unknown that keeps the problem going | National University: "what is currently unknown about the problem and its resolution" |

USC's guide adds qualities the whole passage should have. It asks for clarity and precision, "avoiding the use of value-laden words and terms", and named key concepts. It also asks for the study's "conceptual boundaries or parameters or limitations" and "more than the mere gathering of descriptive data."

Ellis and Levy also separate two terms that get mixed up. The statement of the problem is "one or two sentences" naming the problem. The problem statement is that sentence plus "the argumentation for its viability", meaning the evidence and reasoning around it.

## How is a problem statement different from the research gap, purpose and research questions?

The problem statement says why the study is needed. The gap, the purpose statement and the research questions each do a different job, and National University's guide stresses that they align.

| Part | What it answers | Its form | Source |
|---|---|---|---|
| Problem statement | Why is this research needed? | Declarative, with evidence | National University's [purpose statement page](https://resources.nu.edu/c.php?g=1446805&p=10752287): "The Problem Statement is why you are doing the (proposed) research" |
| Research gap | What do we not yet know that keeps the problem going? | Part of the problem statement's last move | National University: a gap is missing knowledge "that contributes to the persistence of the problem" |
| Purpose statement | What study will you do to address it? | "The purpose of this (qualitative or quantitative) study is to..." | National University: "The Purpose Statement is what type of study (or project) you are doing" |
| Research questions | What exactly will you ask? | Questions, narrow enough to name variables or constructs | Ellis and Levy: research questions are in question format, the problem is not |
| Significance | Who benefits if you answer them? | A separate section in the templates below | USC lists the study's "importance, benefits, and justification" as something the problem statement conveys |

National University says "lack of knowledge" or "lack of research" cannot be the problem, because "it's an academic issue or a gap". Yet the sample in [Valdosta State University's suggested Chapter 1 format](https://www.valdosta.edu/colleges/education/ltwd/documents/FormatforDissertation_ChapterOne.pdf) frames its problem as a relationship studied in one setting but not yet investigated in another.

So ask your supervisor which logic your program uses. If it follows National University's line, lead with a real-world problem and keep the gap as the last move. Our guide to [finding a research gap](https://www.thesisroom.com/blog/research-gap-how-to-find-one) covers the gap itself, and how to test that it is real.

Keep significance short here: a sentence or two on why the problem matters. The full argument for who benefits belongs in its own section, if your template has one.

## Where does the problem statement go, and how long should it be?

In four of the five examples below (all but National University), it sits in Chapter 1 or the introduction. Where a template gives an order, the problem comes before the purpose and research questions. Length runs from a one-sentence opening to three paragraphs. Rules differ between universities, so treat this table as five examples, not a standard.

| University and document | Where it goes | How long |
|---|---|---|
| [National University](https://resources.nu.edu/c.php?g=1446805&p=10752286) (US), dissertation and applied doctoral project guide | Listed first among the components, before the purpose statement | "Do not exceed 250-300 words", with a three-paragraph option |
| [USC Libraries](https://libguides.usc.edu/writingguide/introduction/researchproblem) (US), social sciences writing guide | Under the introduction | "The statement need not be lengthy"; "usually described in a short paragraph" |
| [Howard University, Department of Political Science](https://politicalscience.howard.edu/sites/politicalscience.howard.edu/files/2019-02/Recommended%20Dissertation%20Format.pdf) (US), a recommended proposal format posted by the department | Second section, after the introduction | "This section must begin with a one sentence statement of the problem" |
| [Valdosta State University](https://www.valdosta.edu/colleges/education/ltwd/documents/FormatforDissertation_ChapterOne.pdf) (US), suggested Chapter 1 format | After introduction and background, before conceptual framework, purpose and research questions; significance comes later | Described as "clearly and concisely detailed"; no word limit given |
| [MJP Rohilkhand University](https://mjprudor.ac.in/RMS//Uploads/Website/News/06222022102254.pdf) (India), PhD synopsis guidelines for Arts, Humanities, Languages, Social Sciences, Law and Education | Inside Chapter 1, after "Need and Significance of the study", before operational definitions and research questions | "State the problem to be undertaken in the statement form" |

Notice that Valdosta State puts significance after the problem, while MJP Rohilkhand puts it before. Neither is wrong. Follow your own template, and if there is none, ask your supervisor before you draft.

In the recommended proposal format posted by Howard's political science department, it comes right after the introduction. Our [PhD research proposal format guide](https://www.thesisroom.com/blog/format-for-phd-research-proposal) shows the sections universities ask for around it. For how Chapter 1 hands off to the literature review and method, see [how to write thesis chapters 1-3 as one argument](https://www.thesisroom.com/blog/thesis-chapters-1-3).

## How to write a problem statement in eight steps

Write it after a first pass through the literature, not before. Ellis and Levy warn against basing a problem "solely on personal observations and/or experiences", so the reading comes first.

1. **Start from a real situation in the literature.** Look for something documented as going wrong in practice, or results that conflict. A hunch from your own job is a starting point to check, not the problem itself.
2. **Write one sentence that names what is going wrong.** National University suggests opening with "The problem to be addressed in this study or project is...". A proposal format posted by Howard University's political science department asks for a single opening sentence. It warns that if it takes more, "you are probably unclear about the nature of the problem."
3. **Back it with current, specific evidence.** For each source, say what it found, not only that it exists: Ellis and Levy call this answering "based on what?". National University asks for citations within the last five years; your program's rule may differ.
4. **Show who is affected, where and when.** Name the population and setting. This is Ellis and Levy's "How, Where, and When" part, kept to two sentences in their template.
5. **State the consequences of doing nothing.** National University asks: "Who will be affected? How will they be affected?" Its guide adds that consequences you can measure weigh more than those you cannot.
6. **Name the cause the literature points to, then the gap.** Say what the literature gives as the reason for the problem. Then say what is not known that stops it being solved. That last sentence is your gap.
7. **Line it up with the purpose and research questions.** Read the problem, purpose and questions in a row and check that each names the same population, setting and concept. Once the method is drafted, [Scope Check](https://www.thesisroom.com/room#scopeCheck) quotes pairs of passages from your proposal where what it asks and what it plans do not obviously meet, each with one question about how they join.
8. **Cut it to your program's length and test it.** Trim to your template's limit, then ask USC's "So What?" question of every sentence. If you will defend the proposal, the problem is open to questioning; see our [proposal defense questions, section by section](https://www.thesisroom.com/blog/proposal-defense-questions).

## What does a strong problem statement look like?

Here is an invented example for a Master's thesis in education, labeled with Ellis and Levy's template parts. The citations are shown as placeholders in brackets because the case is invented. In your own draft, each bracket becomes one or more real, current sources.

- **What (the problem):** The problem to be addressed in this study is that rural public secondary teachers in [region] use the shared lesson platform mainly for attendance and announcements. It was introduced for teaching two school years ago.
- **Who says so (evidence):** [The education office's platform use report] and [a recent peer-reviewed study of platform use in similar schools] both describe lesson materials as rarely posted.
- **How, where and when (impact):** Students in these schools still learn from the same printed modules. The platform's main aim, lessons students can revisit at home, has not reached them.
- **Consequences:** If the pattern continues, the office will keep funding a platform that has not changed teaching, and students will keep learning from the same limited materials.
- **Why (cause in the literature):** [Two or three studies of school technology adoption] point to teacher workload and training as likely causes, but they studied urban schools.
- **Gap (what is unknown):** What is not known is how rural public secondary teachers in [region] choose platform features when they plan lessons, and what stops them.

The purpose and first research question then follow from that last sentence:

- **Purpose:** The purpose of this qualitative study is to explore how rural public secondary teachers in [region] decide whether to use the lesson platform when planning lessons.
- **Research question:** How do rural public secondary teachers in [region] describe their decisions about using the lesson platform for lesson planning?

Compare a weak first draft of the same idea: "There is little research on online lesson platforms in [region]." It names a gap, not a problem, which National University rules out. It gives no evidence, no one affected and no consequence, so it fails USC's "So What?" test as well.

## What weaknesses should you check for in a problem statement?

Each weakness below is named in one of the guides cited in this post. Check your draft against all of them.

- **A gap posing as the problem:** "No one has studied X" is an academic gap. National University asks for a real-life issue that exists "regardless of whether research is conducted or not."
- **Personal experience as evidence:** National University's example of a weak problem starts "Based on the researcher's experience", and it says "Anecdotal evidence does not qualify."
- **A comparison or correlation dressed as a problem:** Ellis and Levy say comparing groups "is not a research-worthy problem; rather it is a methodology to address a problem." The same goes for a correlation on its own.
- **Citations that do not say what they show:** Ellis and Levy's faulty example lists sources but gives "no indication of what was contained in the articles cited". Name the finding, not only the author.
- **Circular reasoning:** USC warns against stating the problem "as simply the absence of the thing you are suggesting", using the example "there is no hospital in this community".
- **A situation with nothing going wrong:** Ellis and Levy's faulty example "does not really establish the presence of a problem". Describing a situation is not enough; say what is going wrong and for whom.
- **A question instead of a statement:** Ellis and Levy say the statement of the problem "should not be noted in a question format". Questions belong in the research questions.
- **Too big to study:** National University warns that a problem "too big or too vague" makes it hard to scope a manageable purpose. Narrow the population, setting or time frame.
- **Loaded or vague wording:** USC's list rules out value-laden words and "unnecessary jargon or overly complex sentence constructions."

Boundaries are related but separate. If your committee asks what you chose to leave out, see our guide to [delimitations vs limitations](https://www.thesisroom.com/blog/delimitations-vs-limitations).

## Printable problem statement planner

Fill in one row per part in pencil, then tick the checks before you write the paragraph.

| Part | Your sentence (one or two) | Sources and what each one found |
|---|---|---|
| What is going wrong | | |
| Who is affected, where and when | | |
| Consequences if nothing changes | | |
| Cause the literature points to | | |
| What is unknown (the gap) | | |
| Purpose (one sentence) | | |
| First research question | | |

- [ ] The problem is a real situation, not "a lack of research" (if my program follows this rule)
- [ ] The first sentence names the problem on its own
- [ ] Every source line says what the source found
- [ ] The sources meet my program's recency rule
- [ ] The population, setting and key concept match across problem, purpose and question
- [ ] No sentence is phrased as a question
- [ ] No comparison or correlation is presented as the problem
- [ ] The length fits my program's template (limit: ____ words)
- [ ] Each sentence passes the "So what?" test
- [ ] My supervisor has seen this version (date: ____)

## Frequently asked questions

### What is a problem statement in research?

It is a short passage in Chapter 1 or a proposal that names an issue needing investigation. It gives evidence that the issue exists, says who it affects and what follows if nothing changes, and states what is still unknown. USC Libraries defines a research problem as a clear statement about an area of concern.

### How long should a problem statement be?

It depends on your program. National University's guide says not to exceed 250 to 300 words. USC calls it usually a short paragraph, and a proposal format posted by Howard University's political science department asks it to begin with one sentence. Check your own handbook or template first.

### Is a research gap the same as a problem statement?

No. The problem is what is going wrong; the gap is what is unknown that keeps it going. National University says a lack of research cannot be the problem itself, while Valdosta State's sample frames its problem around a relationship not yet investigated in a new setting. Ask your supervisor which logic your program follows.

### Can a problem statement be written as a question?

Ellis and Levy say no: the statement of the problem should be declarative, and research questions carry the question form. Write the problem as a statement of what is going wrong, then let the research questions ask what you will find out.

### Where does the problem statement go in a thesis?

In the Howard, Valdosta State and MJP Rohilkhand templates cited here, in Chapter 1 or the proposal's introduction, and before the purpose statement and research questions where an order is given. Order varies: Valdosta State's suggested format puts significance after the problem, while MJP Rohilkhand University's synopsis guidelines put need and significance before it. Follow your own template.

### How many sources should a problem statement cite?

No single rule applies. Ellis and Levy's template asks for three current, peer-reviewed references for each part of the statement, and National University asks for citations within the last five years. What matters most is that each source's finding is stated, not only its author.

## Sources

- [The Research Problem/Question, Organizing Your Social Sciences Research Paper (USC Libraries)](https://libguides.usc.edu/writingguide/introduction/researchproblem)
- [Problem Statement, Dissertation and Applied Doctoral Project Components (National University)](https://resources.nu.edu/c.php?g=1446805&p=10752286)
- [Purpose Statement, Dissertation and Applied Doctoral Project Components (National University)](https://resources.nu.edu/c.php?g=1446805&p=10752287)
- [Framework of Problem-Based Research: A Guide for Novice Researchers on the Development of a Research-Worthy Problem (Ellis and Levy, Informing Science)](https://doi.org/10.28945/438)
- [A Recommended Format for Dissertation Proposals (Howard University, Department of Political Science)](https://politicalscience.howard.edu/sites/politicalscience.howard.edu/files/2019-02/Recommended%20Dissertation%20Format.pdf)
- [A Suggested Format for Chapter 1 of the Dissertation (Valdosta State University)](https://www.valdosta.edu/colleges/education/ltwd/documents/FormatforDissertation_ChapterOne.pdf)
- [Guidelines for Preparation of Ph.D. Synopsis (MJP Rohilkhand University)](https://mjprudor.ac.in/RMS//Uploads/Website/News/06222022102254.pdf)

How this guide was made: Thesisroom drafted this guide with the help of AI writing tools. Every fact, figure and source in this post was checked against its original source before publishing.

## Related

- [All guides](https://www.thesisroom.com/blog)
- [Check references free](https://www.thesisroom.com/citation-check)
