Writing the Background Literature in a PhD Application Proposal: Gap Claims, Theoretical Frameworks and Referencing

Applicants writing a PhD proposal commonly treat the background as the least demanding section, aiming to cite sufficiently, cite recent work and demonstrate breadth. It is, however, the section on which the proposal is most often decided, because it is where the research gap must become visible. Reviewers of research proposals record an incomplete literature review, an underdeveloped scientific foundation and a missing theoretical framework among the recurring weaknesses they identify, and a standard rejection checklist includes a literature review indicating that the applicant is unfamiliar with the field. All three are failures of argument rather than of reading.

Purpose of the background section

The background of a PhD application has a single purpose: to make a particular absence in the literature visible and consequential 32. Its other functions — demonstrating that the applicant can read critically, knows the methods and understands the current state of the field — are by-products of fulfilling that purpose well. Reviewers do not need to be persuaded that the field exists. They need to be persuaded that the applicant has identified the specific matter that is not known and that it merits three or four years of a researcher’s time.

This reframes the writing task. The relevant question is not which topics should be covered, but what a reader needs to accept before the research question becomes unavoidable. The background should be constructed backwards from the question, with each paragraph establishing one premise on which the question depends. A paragraph that establishes no premise does not belong in the proposal; it belongs in the thesis introduction, where space is available.

A common failure is a background that reads as a field survey, ordered chronologically, with every subtopic given a paragraph and the gap appearing in the final three lines as a non sequitur. Reviewers record an incomplete literature review and an underdeveloped scientific foundation among their recurring criticisms, and in practice both are often diagnosed from the same document — one that covers a great deal and argues nothing 26,35. Our pages on the role of the literature review in biomedical writing and the five commonest literature review mistakes cover the mechanics in more depth.

Three forms of a gap claim

Gap claims come in three kinds, and they are not equally credible.

The absence claim — nobody has studied X. It is the form to which applicants most often resort and the one reviewers trust least, because the most common explanation for an apparent absence is an incomplete search. If it is made, it must be substantiated: the databases and search terms should be named, together with the adjacent work that would have identified such studies had they existed, and the plausibility of the absence should be explained rather than merely asserted. In most cases the defensible version is narrower — nobody has studied X in this system, or at this scale, or with a method capable of resolving it.

The inconsistency claim — the existing work disagrees, and the disagreement is tractable. This is the strongest form available to most PhD applicants, because it is verifiable and it supplies a study design. The conflicting findings should be stated specifically, a reason for the conflict proposed — different populations, different assays, different endpoints, an unmeasured moderator — and the proposed work shown to discriminate between the explanations. A reviewer can evaluate this in thirty seconds, which is an advantage.

The inadequacy claim — the question has been asked but the available methods could not answer it. This is the natural form for a biotechnology or methods-driven proposal: the measurement was too coarse, the model system lacked the relevant feature, the sample was too small, or the readout was indirect. It requires specificity about the methodological limit, which entails engaging with how the prior studies were conducted rather than with what they concluded.

Write your gap in all three forms and keep the strongest. Applicants who do this almost always find they had reached for an absence claim when an inadequacy claim was available and better.

The theoretical framework in laboratory science

A missing theoretical framework is among the 19 weaknesses peer reviewers commonly identify in proposals, and the evidence that it is widely missing is unambiguous: in a review of 235 implementation studies, fewer than 25 per cent used theory in any way and only 6 per cent were explicitly theory-based 26,32. That prevalence is sometimes interpreted as permission, on the reasoning that an element omitted so widely cannot matter. A more appropriate interpretation is that including a framework is an inexpensive means of distinguishing a proposal.

In laboratory and biotechnology proposals the word “theory” causes confusion, because applicants expect it to mean a named social-science model. What a reviewer expects is more modest and more useful: a statement of the mechanism that the applicant believes to be operating, in sufficient detail that it predicts something 18,32. “We hypothesise that flux through pathway A limits product formation because the intermediate accumulates under all tested conditions” is a framework. It specifies what should be observed if it is correct and what should be observed if it is incorrect. A background that establishes this allows the aims to follow logically rather than appear abruptly 18.

A useful test of a draft is whether a reader, having read the background, could predict what is to be proposed. If so, the framework is serving its purpose; if the aims arrive as new information, the background has summarised a field rather than constructed a position. Where the proposal involves a comparison or an outcome in people or samples, the four PICO (Population, Intervention, Comparison, Outcome) elements give the question its structure — the Pubrica page on the PICO framework describes how, and where it ceases to be the appropriate tool 2.

Weaknesses recorded by reviewers

Two catalogues are informative because they indicate which criticisms reviewers are predisposed to make. A health research agency lists 19 recurring weaknesses that peer reviewers identify, among them an incomplete literature review, an underdeveloped scientific foundation and a missing theoretical framework 26. Separately, a widely used rejection checklist gives 15 reasons proposals fail, one of which is a literature review that shows the applicant is unfamiliar with the field 35.

The overlap is instructive. Three of the recorded weaknesses are about the background, which is a larger share than the background’s word count would suggest 26. The three are also distinct, which matters because they require different remedies. Incomplete indicates that work known to a specialist has been missed — the remedy is a better search and, usually, a specialist reader. Underdeveloped foundation indicates that the relevant work has been cited but nothing has been built from it — the remedy is an argument, not more citations. Missing framework indicates that the mechanism or model is unstated — the remedy is a paragraph that commits to one.

The weakness to which a draft is exposed should be diagnosed before revision, because adding references to a proposal with an underdeveloped foundation makes it worse: it lengthens the section without strengthening the claim, which is how a background becomes both long and unconvincing. The Pubrica page on what to do and avoid in a health care literature review is the companion piece here.

Referencing as a substantive skill

Among 120 novice researchers, 73 per cent reported difficulty referencing material, and the surrounding evidence indicates why early-career applicants are at a disadvantage in this respect: in the same body of work, only 4 per cent of medical students had published an article against 26 per cent of residents 27. Citation practice is learned through supervised correction, and PhD applicants have usually not had that opportunity.

The difficulty lies not in formatting but in the mapping between a claim and the evidence for it, and three specific errors recur. The first is citing a review for a primary finding, which leaves a reader unable to verify the claim and suggests that the original was not read. The second is citing a paper for something adjacent to what it reported, a common drift when a sentence is written first and a citation found afterwards. The third is bundling four references after a broad sentence so that no individual claim is attributable, which reviewers interpret as a reluctance to commit to anything.

The remedy is procedural: the claim should be written first, and the paper then consulted to confirm that it supports the claim. Where a claim is an interpretation rather than a finding, the sentence should say so. This verification should be conducted separately from the writing, because the two tasks cannot be performed well simultaneously. Reference validation is this pass done systematically; literature search and citation support covers the search side.

Bounding the literature to a defensible scope

PhD applicants commonly over-scope the background, usually because they have not yet settled the research question and use breadth as insurance. The result is a section that cannot be completed and a proposal whose aims are not yet determinate.

The scope should be bounded by three explicit decisions, recorded before searching. First, the unit of relevance: the criterion that places a paper within scope. For a mechanistic proposal this is usually the system plus the process; for a methods proposal it is the measurement problem, across systems. Second, the temporal and quality floor: how far back the account needs to extend, and how older work will be treated — usually compressed into one or two sentences establishing where the field settled, with the space devoted to the current disagreement 16. Third, the deliberate exclusions: adjacent literatures not addressed, named explicitly in one sentence. Naming an exclusion demonstrates awareness without consuming space, and it pre-empts a reviewer who might otherwise raise it.

One contextual figure is relevant to scope planning. In a study of student research projects where 358 of 652 eligible students returned both questionnaires — a 55 per cent response rate — 61 per cent undertook clinical projects, 22 per cent basic science and 10 per cent epidemiological 20. If you are proposing basic science, you are in the smaller group, and the readers assessing you may be more accustomed to clinical proposals. That is not a reason to dilute the proposal; it is a reason to make the mechanism legible to a non-specialist in the first paragraph.

Structuring the background to present the research gap

Two structures are effective and one is not. The ineffective structure is chronological: a survey from the earliest relevant work to the most recent. It appears thorough but obscures the argument, because the reader cannot tell which developments the gap depends on.

The funnel is suited to a focused proposal. It opens at the level of the real-world problem, in no more than one paragraph 16. It then narrows to the scientific question to which that problem reduces, and again to what is known about that question, organised by claim rather than by paper. Next comes the limit: what the existing work cannot resolve, with the methodological reason. Finally comes the research question, which by this point should follow as the obvious next sentence. The funnel requires each level to be narrower than the one above it, which prevents the section from expanding laterally.

The contested-claims structure is suited to a gap that takes the form of an inconsistency. It is organised by the disagreement: the claim, the evidence for it, the evidence against it, and what would discriminate between them. This structure is more difficult to write and considerably more convincing when executed well, because it demonstrates adjudication rather than reporting. It also shows most clearly that the applicant has read the methods sections and not only the abstracts.

In either structure, the gap should occupy a distinct paragraph — identifiable, self-contained and placed immediately before the aims 31. Reviewers look for it. Make it findable.

Specific considerations for biotechnology applications

There is no peer-reviewed evidence on biotechnology-specific background writing, so the following guidance reflects professional practice. This is stated explicitly because presenting domain advice in the same register as the measured findings above would be misleading.

Three aspects appear to differ in practice. The evidence being adjudicated is often a mixture of published work and unpublished or preprint methods development, so the background must be explicit about provenance: a yield figure from a preprint is usable provided that it is identified as such. The relevant prior art may include patent literature as well as papers, and a background that ignores it may be exposed by a reviewer with commercial knowledge of the field. Finally, the comparator is frequently a process or a platform rather than a finding, which means that the gap claim is often an inadequacy claim about measurement or scale rather than an absence claim about knowledge.

The structural requirement does not differ. The background must still establish the premises on which the question depends, commit to a mechanism, and end with a readily identifiable gap paragraph. For decisions on the structure of the proposal itself, the Pubrica page on structuring a biotechnology research proposal covers the page budget and the aims page, and biotechnology subject support is where the domain reading gets done.

Demonstrating fit with the programme through the background

A PhD application is also an argument that the project belongs in the host group. A crude form of that argument cites the supervisor’s papers conspicuously, which is transparent and counterproductive. A more effective form uses the background to show that the identified gap is reachable with the group’s capabilities.

This is best achieved through the inadequacy claim. If the limit on the existing work is methodological, and the group has the method, the fit is established without flattery, and the proposal simultaneously acquires a feasibility argument 36. Where the group’s prior work is essential to a premise, it should be cited as any other evidence would be, at the point where the claim requires it; where it is not, it should not be cited.

Two practical points follow. The group’s recent methods sections, not only their abstracts, should be read, because a background that reflects how the group works reads very differently from one assembled from titles. If the gap requires a capability the group lacks, this should be stated, together with the intended source of that capability, because a reviewer will notice the deficiency and silence suggests that the applicant did not. For the wider submission, the Pubrica quality checklist for doctoral writing and thesis and dissertation support cover the later stages.

Key points

Application to your study

Frequently asked questions

@misc{research_proposal_for_phd_application__2026,
  author       = {{Pubrica Academy}},
  title        = {Writing the Background Literature in a PhD Application Proposal: Gap Claims, Theoretical Frameworks and Referencing},
  howpublished = {Pubrica},
  year         = {2026},
  month        = {oct},
  url          = {https://pubrica.com/academy/research/research-proposal-for-phd-application-biotechnology-background-literature-writing/},
  note         = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}

BACKGROUND LITERATURE OUTLINE

STEP 0. WRITE THE QUESTION FIRST  (the whole section is built backwards from this)

  My question is: ____________________________________________
  Expressed as a gap claim of type:  absence / inconsistency / inadequacy
  After reading my background, a reader should be able to predict
  this question without being told it.   [ ] yes  [ ] not yet


STEP 1. SCOPE DECISIONS  (write these down before searching)

  Unit of relevance - a paper is in scope if it: ______________________
  Temporal floor - the account starts at: ______________  because ______
  Databases and terms searched: ______________________________
  Date of last search: ______________
  Adjacent literatures DELIBERATELY EXCLUDED (one sentence each, and
  this sentence goes into the proposal):
    1. ______________________  excluded because ______________
    2. ______________________  excluded because ______________


STEP 2. PREMISE LIST  (the spine of the section)

  List every premise a reader must accept before my question becomes
  unavoidable. One paragraph will install each. If a premise has no
  evidence beside it, the paragraph cannot be written yet.

  P1: ______________________________  evidence: ______________
  P2: ______________________________  evidence: ______________
  P3: ______________________________  evidence: ______________
  P4: ______________________________  evidence: ______________
  P5: ______________________________  evidence: ______________

  [ ] Every premise is needed by the question
  [ ] No premise appears twice
  [ ] Each piece of evidence is a PRIMARY source, or is a review
      cited only for a review-level claim


STEP 3. PARAGRAPH PLAN  (funnel structure)

  PARA 1 - THE PROBLEM, one paragraph only
    The problem in the world: ______________________
    Who is affected and how: ______________________
    Installs premise: ____

  PARA 2 - THE SCIENTIFIC QUESTION THE PROBLEM REDUCES TO
    Reduces to: ______________________
    Why this reduction is legitimate: ______________________
    Installs premise: ____

  PARA 3 to 5 - WHAT IS KNOWN, ORGANISED BY CLAIM NOT BY PAPER
    Claim A: ______________________
      Evidence for: ______________  Evidence against: ______________
    Claim B: ______________________
      Evidence for: ______________  Evidence against: ______________
    Claim C: ______________________
      Evidence for: ______________  Evidence against: ______________

  PARA 6 - THE MECHANISM I AM COMMITTING TO  (the framework)
    I believe the operating mechanism is: ______________________
    If this is right, we should observe: ______________________
    If it is wrong, we should observe instead: ______________________
    Installs premise: ____

  PARA 7 - THE GAP PARAGRAPH  (self-contained; place immediately
           before the aims; a reviewer will look for this)
    What HAS been established: ______________________
    What the existing work CANNOT resolve: ______________________
    The METHODOLOGICAL reason it cannot: ______________________
    Why that limit is removable: ______________________
    Therefore the question is: ______________________


STEP 4. GAP CLAIM STRESS TEST  (write all three, keep the strongest)

  As an ABSENCE claim:
    Nobody has studied ______________________
    Search that would have found it: ______________
    Adjacent work that exists but does not cover it: ______________
    Weakest point a reviewer will attack: ______________

  As an INCONSISTENCY claim:
    Account 1 holds that ______________  based on ______________
    Account 2 holds that ______________  based on ______________
    They differ methodologically in: ______________________
    The observation that would discriminate: ______________________
    Has it ever been reported?  yes / no

  As an INADEQUACY claim:
    The question was asked by ______________________
    The method could not resolve it because ______________________
    My approach removes that limit by ______________________
    Evidence that it does: ______________  [your own preliminary data?]

  STRONGEST FORM: ______________   Reason: ______________________


STEP 5. CITATION VERIFICATION PASS  (separate sitting, papers open)

  For each citation:
    [ ] I have opened the paper, not only the abstract
    [ ] The paper states what my sentence claims it states
    [ ] A primary source is cited for every primary finding
    [ ] Where my sentence is an interpretation, the sentence says so
    [ ] No sentence carries a bundle of citations that makes no single
        claim attributable
    [ ] Numbers quoted match the paper exactly, including units


STEP 6. PRE-SUBMISSION DIAGNOSIS  (which weakness am I exposed to?)

  [ ] INCOMPLETE - a specialist would name work I missed
      Fix: documented search + a specialist reader. NOT more citations.
  [ ] UNDERDEVELOPED FOUNDATION - cited well, argued nothing
      Fix: reorganise by claim; one premise per paragraph.
  [ ] NO FRAMEWORK - mechanism never stated
      Fix: write Para 6 above and make the aims follow from it.

  Specialist reader approached: ______________  by date ______________
  [ ] Draft complete early enough that their reading can change it