Statistics in a nursing research problem statement should support a precise claim, with the population, measurement and reporting period made clear. A striking number is useful only when you can trace it to a reliable source and explain what it actually measures.
This guide provides a practical checking workflow for nursing students writing proposals and dissertations. It focuses on verifying numerical evidence before it enters your paragraph. For the wider structure of the paragraph, read our nursing research problem statement guide.
Quick answer: locate the original figure, record its denominator and definition, check the data period and geographical coverage, and match your sentence to the source. Then decide whether the number establishes background importance, a local practice concern or an unresolved question. These are different jobs.
1. Decide what the statistic needs to establish
Before searching for a number, write the claim you want to support. For example, you might need evidence about the scale of a condition, variation in a care process or the number of people affected by a service problem. Each claim needs a different measure.
A count of hospital admissions cannot automatically support a statement about the number of individuals with a condition. Likewise, a workforce vacancy percentage cannot establish why particular nurses feel exhausted. Start with the meaning of the sentence, then look for suitable evidence.
Give each selected statistic a clear role. Background evidence shows why a topic deserves attention. Evidence about practice describes what is happening in a defined service. Evidence about research describes what studies establish and what remains uncertain. A large background number does not perform all three functions.
Try this drafting note: “I need this figure to establish ___, within ___, during ___.” If you cannot complete it, the figure may be decorative. Remove it until you know what contribution it makes.
For a qualitative dissertation, numerical context may be helpful, but it does not replace evidence about experiences or processes. Your problem statement can be well supported without forcing a percentage into every opening sentence.
2. Trace statistics in a nursing research problem statement to the source
When you find a useful number in a blog, news article or dissertation, use it as a lead. Follow its citation to the research paper, official publication or dataset that generated the figure. Read the accompanying explanation before copying the value.
Check whether the figure comes from the source’s own analysis or appears in its background section. A journal article may repeat a statistic from an earlier publication. If you cite that article as the original evidence, you may hide an older data period or a different population.
Record the table, figure, page or dataset sheet where the number appears. Save the source title and publication details alongside that location. This small step helps you answer a supervisor’s question without searching through dozens of downloaded files again.
If the original source is unavailable, do not pretend you have read it. Follow your university’s rules for secondary referencing and make the chain clear. Alternatively, select an accessible source that supports a narrower claim.
A reference generated by an AI tool is also a lead that requires checking. Open the publication, confirm that it exists and inspect the actual result. A plausible title and a polished citation do not verify a numerical claim.
3. Separate publication date from the data period
A report published recently may describe information collected several years earlier. Therefore, write down both the publication year and the period represented by the data. The first belongs in your reference; the second may need to appear in your sentence.
For example, an invented teaching report published in 2026 might summarise records from 2022 to 2024. Calling its results “the situation in 2026” would change the meaning. A more accurate sentence would identify the years actually covered.
Check for a newer release, correction or revised dataset. If you replace an older figure, also review the surrounding interpretation. A change in the value may result from a changed definition or reporting process rather than a changed underlying problem.
Older evidence can remain useful for explaining historical context or an established methodological issue. However, explain that role instead of presenting it as current prevalence. There is no universal rule that makes every source older than five years unsuitable.
For trends, make sure that the compared periods are equivalent. One full calendar year and one partial year should not be described as a straightforward increase or decrease without accounting for their different coverage.
4. Check the population and geographical coverage
Statistics in a nursing research problem statement need boundaries. Ask who was included, who was excluded and where the information was collected. A result from a selected patient group should retain that group’s description when you use it.
Do not turn an England-only figure into a UK-wide claim. Similarly, distinguish national statistics from regional estimates, a single hospital audit or one university’s student survey. The geographical label is part of the finding, not an optional detail.
Consider whether the population fits your proposed enquiry. A statistic covering all healthcare workers may provide broad context for a nursing topic, but it should not become a nurse-specific estimate. Keep the broader wording or locate more directly relevant evidence.
Age limits, care settings and eligibility criteria can also change relevance. An inpatient measure may not describe community services. A study involving newly qualified nurses may not represent the whole registered workforce.
If you use wider evidence because closely matched evidence is unavailable, explain the boundary. You can say that the source provides context while the proposed question concerns a more specific group. Our guide to using UK evidence explores contextual relevance in more detail.
5. Identify the denominator and unit of measurement
Before using a percentage, ask “percentage of what?” The denominator might be patients, respondents, appointments, admissions, records or eligible care episodes. These units cannot be exchanged without changing the claim.
Consider an invented audit of 120 discharge records. If 30 lack a documented item, the result is 25% of the reviewed records. It does not establish that 25% of all patients nationally missed the care, or that the care never occurred.
Distinguish people from events. One person can have several appointments or admissions. A count of events may therefore exceed the number of individuals involved. Preserve the source’s unit when you paraphrase its result.
Also check how missing responses were handled. A survey may calculate a percentage from the people answering one item rather than the full recruited sample. Your notes should identify the relevant base wherever the source provides it.
Use this plain-language check: “Among [the stated denominator], [the measured outcome] occurred or was recorded at [the reported value].” If that sentence exposes an unsupported assumption, revise the wording before including the number in your dissertation.
6. Preserve the outcome definition
A statistic has a definition as well as a value. Read the source’s methods, glossary and table notes to understand what qualifies as the reported outcome. Similar labels can hide different measurements.
For example, an intention to leave measure should remain a statement about intention. It should not become evidence of actual departures. Likewise, dissatisfaction with one aspect of care does not automatically mean dissatisfaction with the whole service.
Documentation and delivery also require separate wording. A record review may show that an action was not documented. Unless the study establishes more, it cannot tell you with certainty that the action did not happen.
Write down any threshold, questionnaire item or classification rule relevant to your claim. If a score indicates elevated symptoms, keep that description rather than introducing a diagnosis the study did not make.
Finally, reread your paraphrase against the result. Ask whether you have changed “reported,” “recorded,” “screened” or “estimated” into an absolute assertion. Accurate verbs help the reader understand what the evidence can support and where its limits begin.
7. Distinguish percentage points from percentage change
The Office for National Statistics explains that percentage points describe the difference between percentages (ONS, n.d.a). Using the wrong expression can make a change look much larger or smaller than intended.
In an invented example, a recorded proportion rises from 20% to 25%. The difference is 5 percentage points. The relative percentage increase is 25%, calculated as the difference divided by the original value, multiplied by 100.
These statements describe the same change from different perspectives. Choose the expression that fits your purpose and make it explicit. Do not write “a 5% increase” when you mean a rise of 5 percentage points.
For your checking notes, retain the original values as well as the calculated change. This makes the arithmetic easier to verify and helps prevent a secondary source’s ambiguous wording from entering your paragraph.
If a calculation adds nothing useful to the problem statement, use the original result. The purpose is to establish a credible context, so an extra transformation should clarify the argument rather than make the number more dramatic.
8. Read uncertainty without overstating the estimate
Survey estimates may be accompanied by measures of uncertainty. ONS explains that these help readers understand how an estimate may differ from the population value (ONS, n.d.b). Read the uncertainty information alongside the headline result.
A problem statement does not usually need a full statistics tutorial. However, it should avoid presenting an estimate as an exact count of everyone affected. Keep words such as “estimated” where they are needed.
Where the source reports a confidence interval, check whether its width affects the claim you plan to make. Use the publication’s own interpretation of precision and statistical comparisons rather than deciding from the headline value alone.
A small observed difference between two releases is not automatically evidence of an established trend. Check whether the source supports that interpretation and whether the figures are comparable. When the evidence is uncertain, retain proportionate wording.
For a short opening paragraph, you may decide that a simpler, directly relevant finding is preferable to a complex estimate requiring several qualifications. That is an editorial choice about clarity, not permission to omit a limitation that changes the conclusion.
9. Check whether two statistics are comparable
Before writing that a problem is increasing, compare the definitions behind the numbers. Ask whether both figures refer to the same population, measurement, period length and data collection process.
Two studies can use the same outcome label while asking different questions or applying different thresholds. In that case, their percentages should not become a simple before-and-after comparison. Describe the variation and examine what explains it.
Likewise, a national survey and a single-service audit may provide different types of evidence. One may estimate a population characteristic; the other may describe records reviewed in a local project. Place each in its appropriate role.
If you cannot establish comparability, do not calculate an average or difference merely because both results are percentages. Instead, choose one suitable source or explain why the evidence cannot be directly combined.
Keep a short comparison note: “These figures differ in ___, so they can support ___ but cannot establish ___.” This helps you retain useful evidence while limiting the conclusion to what the sources actually show.
10. Use a verification table before drafting
A compact evidence-checking table helps you organise statistics in a nursing research problem statement. Complete it in your working notes before deciding which figures belong in the final paragraph.
| Check | What to record | Decision it supports |
|---|---|---|
| Source | Original publication and exact result location | Can the figure be verified? |
| Time | Publication date and data period | Does the wording describe the correct period? |
| Coverage | Population, eligibility and geography | Does the claim stay within the evidence? |
| Measurement | Outcome definition, unit and denominator | Does the sentence preserve the finding? |
| Interpretation | Limitations and uncertainty relevant to the claim | How much confidence does the wording imply? |
| Purpose | Background, local concern or research uncertainty | Why does this number belong here? |
Add a final column for your proposed sentence. Comparing that sentence with the recorded details often reveals a population switch, an outdated period or an exaggerated interpretation.
Keep the table as a drafting aid unless your brief requests it as an appendix. It is not a replacement for a reference list or critical appraisal. Its purpose is to preserve the details that are easily lost when several sources are condensed into one paragraph.
11. Statistics in a nursing research problem statement: worked example
The following figures are invented for teaching and must not be cited as real findings. Imagine a service audit reviewed 80 eligible discharge records from one ward over three months. Twenty records lacked documentation of a specified education item.
A weak draft might say: “A quarter of UK patients receive no discharge education, proving that nurses need more training.” This changes the population, geographical coverage and outcome. It also introduces a causal explanation the audit did not investigate.
A defensible description of the imagined result would be: “In a three-month audit of 80 eligible discharge records from one ward, 20 records (25%) lacked documentation of the specified education item.” The sentence now retains the unit, period and setting.
The next sentence must explain the implication carefully. For example, the result could justify examining documentation practices locally. It would not, by itself, establish national prevalence, the absence of all education or the reason for the missing entries.
If the proposed dissertation concerns patients’ understanding after discharge, additional evidence would be needed. Missing documentation and patient understanding are different measurements. The example shows why verifying the number also helps test whether it belongs in the chosen research problem.
| Available evidence | Suitable wording | Unsupported extension |
|---|---|---|
| Records missing an education entry | The item was not documented in the reviewed records | No patients received education |
| Respondents expressing an intention to leave | Respondents reported an intention to leave | Those respondents have left their jobs |
| One ward’s three-month audit | The result describes that ward and reporting period | The result represents all UK wards |
12. Connect the verified figure to the unresolved question
Once a statistic has passed these checks, explain why it matters to the proposed enquiry. Avoid placing several figures together and assuming that their size establishes a research gap.
A numerical result may show that an issue affects a particular group. The literature still needs to establish what remains unresolved about that issue. The uncertainty could concern an experience, a process, a comparison or an explanation that your method can investigate.
Use a transition that makes the reasoning visible: “This finding establishes [the supported concern]; however, it does not explain [the specific uncertainty].” Complete the second part only after reviewing relevant evidence, not simply because the selected statistic lacks that information.
For example, an administrative count does not explain patient experience, but qualitative studies elsewhere may already address it. Check the wider literature before claiming the experience is unknown.
UKRIO’s Code of Practice stresses accurate reporting and warns against misleading selection of findings (UKRIO, 2025). Applying that principle here means choosing evidence for relevance, keeping important qualifications and allowing contrary findings to change your proposed argument.
Final checklist before submission
Read every numerical sentence in isolation, then compare it with its source. A citation at the end of a paragraph should not be expected to support unrelated claims about prevalence, cause, staffing and intervention effectiveness.
- Can I locate the exact figure in the source?
- Have I preserved the data period and geographical coverage?
- Does the population match my wording?
- Have I retained the correct denominator and unit?
- Does my verb describe what was measured?
- Are percentage points and percentage changes labelled correctly?
- Have I included limitations that affect the interpretation?
- Does the number contribute to the proposed research problem?
If a figure fails the check, correct it, qualify the sentence or remove it. Do not replace it with an unsupported estimate merely to keep the paragraph looking evidence-rich.
For feedback on your own draft, review our nursing dissertation proposal support guide. Bring your brief, proposed question and source details so feedback can examine the research logic. Use support within your university’s academic integrity rules and retain responsibility for your own submitted work.
Frequently asked questions
Does a nursing research problem statement always need statistics?
No. The evidence should fit the claim. Qualitative findings, a synthesis of studies or relevant policy may support a problem more directly than a percentage. Use numerical evidence when it contributes to the reasoning.
Can I use a statistic quoted in another dissertation?
Follow its reference to the original publication and verify the figure. The dissertation may have copied an outdated number or altered its meaning. If you cannot access the original, follow your institution’s secondary referencing rules.
How recent should my statistics be?
Use evidence suitable for the claim and check whether a newer release exists. Record the data period separately from the publication date. Historical figures should be labelled as historical context rather than presented as current estimates.
Can national statistics establish a problem on my placement ward?
They can provide wider context, but they do not demonstrate the local situation. A local claim needs appropriately authorised local evidence. Keep patient and service information confidential and follow the relevant governance requirements.
What should I do when two sources report different percentages?
Compare their populations, definitions, periods and methods before treating the difference as a contradiction. Report the distinction where it matters, or select the source most suitable for your specific claim. Avoid an unsupported combined estimate.
Explore specific questions in more depth
- Absolute vs Relative Risk in Nursing Problem Statements: How should a nursing problem statement use absolute and relative effects to describe the scale of a problem fairly?
- How to Find Local Statistics for a Nursing Research Problem: Where should nursing students look for local numerical evidence, and what should they do when suitable local data are unavailable?
References
Office for National Statistics (ONS) (n.d.a) Percentages and percentage points. Available at: ONS content style guide (Accessed: 13 October 2026).
Office for National Statistics (ONS) (n.d.b) Uncertainty and how we measure it for our surveys. Available at: ONS uncertainty guidance (Accessed: 13 October 2026).
UK Research Integrity Office (UKRIO) (2025) Code of Practice for Research. Version 3.5. Available at: UKRIO Code of Practice (Accessed: 13 October 2026).
