Why English? | Writing an Academic Research Statement

Find the section you need

A research programme is different from a project list

Imagine receiving three paragraphs headed Project A, Project B and Project C. Each explains a technique and ends with a publication. You can see that the writer has been busy. You may still be unable to say what question connects the work or what the writer would investigate as an independent academic.

A programme supplies that connection. It can be a recurring problem, a theoretical distinction, a population, an analytical approach or a set of linked questions. The connection should explain why the selected projects belong together. It should not be so broad that almost any work would fit beneath it.

For example, “I study education” gives the reader a field. “I investigate how students transfer an explanation from a familiar example to an unfamiliar problem” gives the reader a research problem. It leaves room for different methods while making the intellectual centre easier to recognise.

Read the request as a brief

Before drafting, record the requested document, permitted length, role, department and submission format. Check whether a separate proposal, teaching statement, impact statement or supporting statement is required. A research statement may need to summarise a broad agenda, while a proposal may require a detailed method and work plan.

Do not decide the document’s purpose from its name alone. An institution may define “research plan” differently from another institution’s “research statement”. Where the instructions are genuinely unclear, ask the authorised recruitment contact a narrow question. You are clarifying the task, not asking them to write your application.

The useful test is simple: after reading the request, can you explain what decision this document helps the committee make? That answer should shape the amount of detail you include.

Contents · Next section

Start with a question that survives different projects

Take a blank page and write the questions that motivated your strongest work. Avoid publication titles initially. Titles often contain disciplinary shorthand; questions expose what you were trying to understand. Next, draw the relationship between the questions: one may refine another, test its boundary or move it into a new context.

Suppose one study examined how novices explain a worked example and another examined what they do when the surface details change. The connecting question might concern the conditions under which explanation supports transfer. The next study could then test a specific condition rather than simply repeat the earlier activity with new participants.

This is not an instruction to manufacture coherence. Sometimes a career includes genuinely separate strands. You can name two strands and explain their relationship honestly. A clear account of distinct areas is stronger than an artificial claim that everything has always belonged to one master plan.

Choose a thread at the right scale

A thread can be too narrow. “I compare the wording of item seven in one questionnaire” may identify a useful task but not the future programme. A thread can also be too broad. “I will improve education worldwide” makes the significance sound large while leaving the research mechanism invisible.

Try three levels: the broad problem, the researchable question and the particular study. The statement should move between them deliberately. The opening establishes the broad problem and researchable question; selected examples demonstrate how you investigate it; the future section shows the next studies.

Ask a colleague to name the thread after reading your first page. If they describe only a technique, add the question the technique serves. If they describe only an aspiration, add the distinction your research can actually examine.

Contents · Previous section · Next section

Give context without writing a literature review

An expert reader may know the debate already. Another committee member may need two sentences explaining what is uncertain and why that uncertainty matters. Write for both by beginning with the problem, then identifying the precise gap your work addresses.

Consider this invented opening: “Students can reproduce a worked solution yet struggle when the same relationship appears in a different form. My research examines which features of explanation help learners recognise that relationship across representations.” The reader now has a situation, a discrepancy and a question before meeting any technical vocabulary.

You can then introduce a necessary term with a short definition. The purpose is not to eliminate specialist language. It is to give the reader enough structure to interpret it. A technical term becomes useful when it names a distinction more precisely than ordinary language can.

Make the gap specific and defensible

Avoid saying “no research has examined” unless you have a basis for such a sweeping claim. Often the real gap is narrower: an existing method has not been tested under a particular condition, evidence is inconsistent, a mechanism remains unclear or a population is poorly represented.

A gap statement should tell the reader what would count as progress. “The role of preparation time remains unclear” becomes stronger when you identify the comparison, outcome and context. The statement need not include every operational detail, but it should show that the question can lead to evidence.

When a finding challenges an assumption, name the assumption accurately. Do not turn a nuanced literature disagreement into a claim that everyone else was wrong. Scholarly generosity and intellectual clarity can coexist.

Contents · Previous section · Next section

Use evidence to support a programme

The statement is not a complete publication catalogue. Choose work that demonstrates the central question, relevant capability or development of your agenda. A less recent project may deserve space because it establishes the method needed for the next phase. A prominent publication may need little space if it is peripheral to the advertised role.

For each selected example, answer four questions: What was the problem? What did you do? What did the work establish? Why does it matter for the programme? Write those answers before adding disciplinary detail.

This order helps prevent a common imbalance. Writers sometimes spend most of a paragraph describing equipment, datasets or techniques and leave the finding in an ambiguous final sentence. The reader then knows how complicated the work was but not what changed in knowledge.

Describe findings with their boundaries

Compare “This intervention improves learning” with “In this sample, the intervention was associated with more accurate explanations on the immediate task.” The second sentence may sound less dramatic, but it tells the reader the context and outcome. If the design supports a stronger causal conclusion, explain that accurately rather than borrowing certainty from enthusiasm.

Boundaries can concern population, setting, measurement, timescale or method. They need not overwhelm every sentence. Place the most important limitation near the claim it qualifies, then explain how the next study addresses it.

A well-bounded result creates a better bridge to future work. It shows what is known and what remains worth investigating. The unresolved question becomes an intellectual opportunity, not an embarrassing footnote.

Contents · Previous section · Next section

Explain what you actually contributed

Many projects are collaborative. A committee cannot infer your role from an author list alone. Use verbs that name real contributions: conceived, designed, implemented, collected, analysed, interpreted, coordinated or wrote. Select the verbs your record supports.

An invented example might read: “Working with two collaborators, I designed the comparison tasks and led the analysis of student explanations.” This gives the applicant a clear contribution without erasing the team. If the dataset or framework came from another group, acknowledge that context where it matters.

Do not write “I developed” for a tool you only used. Equally, do not hide a substantial contribution behind an uninformative “we”. Precise attribution helps others see your capability and helps you explain what you could carry into a new setting.

Show the next independent decision

Independence can be demonstrated through the question you formulate, the judgement you exercise and the programme you plan to lead. It does not require claiming that you can work without collaborators, infrastructure or mentoring.

Explain how the next phase extends or differs from your supervisor’s or team’s work. The distinction may involve a new mechanism, context, method or theoretical direction. Avoid adversarial language. Intellectual development is not a competition to minimise everyone else’s contribution.

If the programme depends on continued access to data, equipment or materials, state that dependency. Having used a resource previously does not automatically mean you can transfer it to another institution. The statement should make the capability visible while keeping permission separate.

Contents · Previous section · Next section

Connect each next question to the evidence

A future agenda becomes persuasive when the reader can see why each study follows. One finding may generate a mechanism question; a limitation may generate a boundary test; an application may require validation in a different setting.

Try writing the transition without project names: “The earlier study established X under condition Y. The next phase asks whether Z explains the pattern.” If that sentence is difficult, the future project may not yet have a clear relationship to the programme.

Two or three linked questions can be more convincing than a long list of possible topics. The list shows imagination; the sequence shows judgement. You can preserve exploratory space without presenting every interesting idea as a firm commitment.

Distinguish phases and decision points

An invented three-phase programme might first validate a measure, then use it to compare mechanisms, then test transfer in a more varied setting. Each phase should produce something that supports the next decision. A delay in one phase should not make the entire programme unintelligible.

Describe a milestone as an observable state, not merely an activity. “Conduct interviews” says what you will do. “Complete an approved interview dataset sufficient to compare the two proposed explanations” says what the activity is for. The exact sample and approval requirements belong in the detailed plan where requested.

Use conditional wording when a phase depends on a result. “If the measure distinguishes the cases reliably, I will…” is more honest than presenting every downstream step as inevitable.

Contents · Previous section · Next section

Name the resources that change the plan

Feasibility concerns time, people, access, equipment, expertise, approvals and money. You do not need a full budget unless requested, but the reader should understand the main dependencies. Separate what you already know how to do from what the proposed setting must supply.

For an archive project, the key dependency might be permission to consult a collection. For a laboratory project, it might be access to a particular instrument. For computational work, it might be lawful access to data and suitable infrastructure. Do not use a university’s reputation as evidence that a resource will be available to you.

Where a dependency is unconfirmed, identify the verification step. A realistic statement can say that a collaboration or resource is proposed. It should not represent a general conversation as an agreed partnership.

Offer alternatives that preserve the question

A useful contingency changes the route while retaining a meaningful purpose. If one dataset is unavailable, what other evidence could examine the question? If recruitment is slower than expected, which phase remains viable? If a technique fails validation, which conclusion becomes unavailable?

Avoid the promise that every setback will produce the same result. A smaller sample may change precision; a different archive may change coverage; a substitute method may answer a narrower question. State the consequence of the alternative, not just its existence.

This is where English supports research judgement. Words such as provisional, contingent, exploratory and validated tell the reader how much weight to place on a plan. They are not signs of weakness when they accurately describe its state.

Contents · Previous section · Next section

Find actual connections

Read the official department, programme, centre and facility pages relevant to the vacancy. Choose a few connections that genuinely affect your agenda. A long list of faculty names can look like browsing rather than thinking.

For each connection, explain both directions: what the setting could enable and what you could contribute. A shared method may support collaboration; a complementary field may create a useful comparison; a curriculum need may let research inform teaching. Keep these possibilities distinct from existing agreements.

An invented fit sentence might say: “The department’s documented work on classroom interaction would provide a relevant setting for my proposed comparison of spoken and written explanations.” It should not say that the department has agreed to run the project unless that is true.

Avoid prestige as a substitute for fit

“Your world-renowned institution is the ideal place” gives almost no decision-relevant information. Replace it with the resource, intellectual connection or student context that actually matters. Verify current names and affiliations immediately before submission because people and programmes move.

If the fit is partial, explain the transfer and the learning required. You do not have to pretend you already know a new institution’s working practices. A thoughtful candidate can offer a credible starting point and a plan for understanding local context.

The broader career lesson is useful for students too: a famous name is not the same as a suitable environment. The quality of a decision depends on the connection between your purpose and the actual opportunities available.

Contents · Previous section · Next section

Mark the time of every claim

Readers need to know whether work is completed, ongoing or planned. Keep accepted, published, submitted, under review and in preparation distinct. Similarly, distinguish awarded funding from an application or a potential funding source.

Past tense can describe completed actions. Present tense can describe a programme or established relationship. Future and conditional forms can describe plans. The important issue is not a rigid grammar rule; it is whether the sentence gives the reader a truthful status.

An invented contrast makes the point: “I secured a grant” is different from “I submitted a grant application” and different again from “I plan to seek funding from a scheme whose remit fits this question.” No amount of fluent editing should collapse those states.

Choose verbs that fit the evidence

Demonstrates, suggests, supports, estimates and is consistent with carry different strength. Select the verb after checking the design and evidence, not before. Be especially careful when moving from a specific result to a broad application.

“The method could inform assessment design” names a possibility. “Schools use the method in assessment design” names uptake that needs evidence. “The method improves assessment” names an outcome that needs another kind of evidence. These may all be worthwhile statements, but they are not interchangeable.

During revision, highlight every strong evaluative word. Ask what the reader could inspect to justify it. If the answer is unclear, replace the adjective with a concrete description or reduce the claim’s scope.

Contents · Previous section · Next section

The first version

Consider a researcher who writes: “My research revolutionises learning through innovative explanation technologies. I have worked on several highly successful projects and will develop a world-leading programme with major educational impact.”

The problem is not simply tone. The reader cannot identify the question, method, contribution, result, future sequence or boundary. The paragraph asks the committee to accept evaluations that it cannot examine. It also combines past and future claims without revealing their evidence.

Start the repair by deleting the evaluations temporarily. Write only what happened and what you intend to investigate. This usually exposes the underlying argument more quickly than searching for modest synonyms.

A more useful version

An invented revision might say: “My research examines how learners recognise a scientific relationship when an explanation changes form. In a recent collaborative study, I designed comparison tasks and analysed how students linked diagrams to written claims. The findings identified a distinction between repeating an explanation and applying it to a new case. My next phase will test whether contrasting representations make that distinction more visible to learners.”

This is still only an outline. The actual applicant must insert accurate evidence, context and limitations. Yet the reader can already see a question, a contribution, a result and a next step. The paragraph gives the committee something to evaluate.

Add feasibility and fit after the logic is clear

The next paragraph could explain the measure, proposed setting and principal dependency. For example, a new study might require an approved classroom partnership and a validated assessment task. Those conditions should be named as planned requirements, not completed arrangements.

Only then should the writer explain why the advertised department is suitable. A verified programme connection can support the fit argument. It should not replace the research question or make the statement depend on invented enthusiasm from prospective collaborators.

The order matters. First the reader understands the intellectual programme. Next they see why it is feasible. Finally they see why this setting is relevant. A different order may work, but the three jobs must remain visible.

Contents · Previous section · Next section

Argument before sentences

Read only the opening sentence of each paragraph. Together, they should form a coherent route through the problem, selected evidence, current work, future agenda and fit. If they jump unpredictably, reorganise before polishing grammar.

Next, check what each paragraph contributes. Two paragraphs may both establish technical competence while neither explains the next research question. A beautiful sentence should not survive merely because it is beautiful. Its place depends on the document’s purpose.

Ask a reader outside your subfield to summarise the programme in three sentences. Do not immediately correct them. Their summary reveals what the document actually communicates. A missing distinction is a revision target, not proof that the reader should have tried harder.

Evidence before confidence

Audit names, dates, numbers, publication states, funding states and contribution verbs. Compare them with the CV and supporting documents. A mismatch can arise from an old draft rather than dishonesty, but it still needs correction.

Then check citations and links. Use the requested convention; avoid turning a short statement into a bibliography unless instructed. When you refer to a particular work, make it identifiable enough for the reader to locate it.

Finally, inspect the exported file, not just the editable document. Look for clipped text, unreadable figures, broken characters, missing pages and an incorrect filename. Good reasoning must survive the submission format.

Contents · Previous section · Next section

For students: practise the evidence-to-next-question move

You do not need a doctorate to practise this kind of English. After a science investigation, write what the evidence shows, what it does not show and what you would test next. After reading two sources, explain where they agree and which difference deserves another question.

This exercise develops the underlying capability without pretending a school task is an academic job application. The level of technical detail should match the learner. A younger student may use one observation and one next question; an older student can distinguish method, inference and scope.

The aim is to make thinking visible. A teacher can then respond to the actual reasoning instead of guessing from a confident final sentence. That is a practical route from ordinary English learning to research communication.

For parents and advisers: ask for the connection

A useful supporting question is “How does this next step follow from what you found?” Another is “Which part of that claim comes from evidence, and which part is your idea?” These questions encourage precision without demanding adult academic language from a child.

Avoid replacing the learner’s words with a polished paragraph they cannot explain. Help them identify the missing relationship, then let them rebuild it. Multilingual discussion can be valuable for thinking through a difficult idea; the final English explanation should preserve the meaning developed in that discussion.

For adult applicants, feedback works similarly. A mentor can challenge an unclear connection or excessive promise. The applicant remains responsible for the record, intellectual choices and final wording.

Contents · Previous section · Next section

Begin with a one-page map

Write the central question at the top. Beneath it, create three columns: established work, current investigation and next questions. Add only items that help the reader understand the programme. Beside every item, note your contribution and the status of its evidence.

Now draw the transitions. Which result makes the next question possible? Which limitation requires a boundary test? Which technique provides the capability for the future phase? The map should reveal relationships that a chronological CV does not.

If an item has no connection, decide whether it is a separate strand or should be omitted from this statement. You are selecting for one document, not declaring that the work has no value.

Write two versions of the opening

In the first version, explain the programme to a colleague from another department. In the second, explain it to a specialist who knows the debate. Compare the two. Keep the first version’s orientation and the second version’s necessary distinctions.

Then draft the middle using one or two strong examples. End with a future agenda that follows from them. Add verified fit and major dependencies after the argument is visible. Use the actual length limit to decide how much detail each part receives.

Do not shrink the font to rescue a crowded statement. Remove a weaker example, shorten background or move requested technical detail into the appropriate separate document. Readability is part of the communication task.

Finish with a boundary audit

Underline statements about novelty, effectiveness, adoption, independence, funding and institutional access. For each, ask what evidence supports it and whether the verb is appropriately strong. Check that plans are still plans and collaborative work is still collaborative.

Read the final version aloud. Long sentences sometimes conceal a missing actor or a causal jump. Break them at the point where the reasoning changes. A short sentence can carry a complex idea when its nouns and relationships are precise.

The finished statement should leave the reader with three answers: what you investigate, why your record supports the next phase and how the proposed setting connects to the programme.

Contents · Previous section · Next section

Exercise A: a method without a question

An invented applicant writes: “I use interviews, surveys and machine learning to study students.” Ask what comparison each method supports. Interviews might examine how learners describe a difficulty; a survey might estimate a pattern across a larger group; a computational model might classify an observable feature. None of those purposes follows automatically from the method’s name.

Rewrite the paragraph around the question. Then explain why the selected evidence can address it. If the methods answer separate questions, make that division explicit. If one method is exploratory and another tests a pre-specified claim, preserve the distinction. The repair is conceptual before it is grammatical.

A useful check is to remove the method names temporarily. Can the reader still identify the problem and the proposed comparison? If not, the paragraph needs a clearer intellectual centre. Put the technical names back only after that centre is visible.

Exercise B: a future agenda without a bridge

Another applicant describes a completed study of classroom discussion, then announces a future programme on digital assessment with no transition. The projects might connect, but the reader cannot see how. Write the missing sentence: which finding, limitation or capability from the discussion study makes the assessment question worth pursuing?

One possible bridge is that the earlier study revealed a difference between what learners say in discussion and what they demonstrate individually. A future assessment project could examine that discrepancy. This is an invented example of a relationship, not a finding to copy into an application.

Now check whether the future method actually examines the discrepancy. A bridge that sounds elegant but has no methodological consequence is still weak. The statement becomes stronger when the relationship affects the next research decision.

These exercises teach two complementary English skills: naming the purpose beneath a technical label and explaining the connection between two pieces of work. Both are useful in school investigations, university assignments and professional proposals long before an academic appointment is in view.

Contents · Previous section · Next section

Is a research statement the same as a research proposal?

Not necessarily. A statement commonly connects a broader record and agenda, while a proposal may specify one project in more detail. Read the application definitions and required headings. If both are requested, give each a distinct job rather than submitting the same text twice.

Should I include every publication?

Use the CV for the full record unless instructed otherwise. In the statement, select work that supports the programme, relevant capability or future direction. Explain what the selected work established and what you contributed. A list of titles alone does not do that explanatory work.

How technical should the statement be?

Technical enough to preserve the research question and method, readable enough for the intended committee. Define terms that carry essential distinctions and remove jargon that merely labels familiar ideas. Ask both a specialist and an informed non-specialist to read it.

What if my research has several strands?

Name the strands honestly and explain their relationship where one exists. You may have a methodological connection, a shared question or a complementary programme. If there is no genuine connection, use clear headings rather than inventing one. Select the strands relevant to the role.

Can I name possible collaborators?

Yes, where current official information supports the intellectual connection. Describe a potential collaboration as potential unless it is agreed. Verify affiliations and avoid implying access, funding or commitment from someone who has not provided it.

Can AI help with editing?

It can identify unclear sentences or suggest an outline, subject to the application’s rules. It can also invent findings, references and commitments. Keep confidential material within approved arrangements and verify every change. Your research judgement and factual record must remain yours.

Final research-statement checklist

• The opening names a research question rather than only a field or technique.
• Selected past work explains a finding and your actual contribution.
• Current work is distinguished from completed and planned work.
• Each future phase has a reason to follow from the earlier evidence.
• Resources, permissions and potential collaborations carry their correct status.
• Institutional fit uses current verified particulars, not reputation alone.
• The exported statement matches the requested length and agrees with the CV.

Use this list after substantive revision. It is not a substitute for reading the argument from beginning to end, because an individually accurate sentence can still be placed where it creates a misleading connection.

Contents · Previous section · Next section

Use Cornell’s research-statement guidance for an official university view of the document’s purpose. Read Oxford’s academic-application guidance when coordinating research and teaching materials.

For connected eduKate routes, the teaching-philosophy statement guide owns the teaching document, while the research-fellowship interview guide helps turn a written case into spoken reasoning. Return to Why English Matters for the broader importance discussion.

Clear research writing is a generous act. It lets another person see the question, examine the evidence and understand the next step without having to guess what your confidence means.

Contents · Previous section · Why English Matters