A dissertation is judged against its own question, so a question that is vague, too wide or settled by a yes or no weakens every chapter built on it. This guide to how to write research questions, aims, objectives and hypotheses, published by EdCitation, sits under our dissertation guide: the step between the topic and the proposal.
We read each rule below on 26 September 2026, on seven universities' own library, writing-centre and skills pages: Sheffield Hallam and Hull in the UK, George Mason and National University in the US, and Melbourne, Monash and UNSW Sydney in Australia. The frameworks come from their original papers. Where the universities disagree, we say so; examples marked as ours are ours.
EdCitation writes no part of a dissertation, and that includes its question. Find sources shows how much has been published on a draft question, and Cite a source builds the references behind it.
How to write research questions and aims, step by step
In eight steps. Your handbook's rules override any of them.
- Read the requirements first: Monash University Library (n.d.) says the assignment's purpose decides how the question is worded.
- Read a few recent sources and one or two influential ones, for the main debates. Find sources covers about 300 million published works; search it by topic or by claim.
- Narrow the topic by group, place, period or organisation, the limits Monash suggests.
- Write one main research question: open, focused, and answerable with evidence you can reach.
- Test it against the criteria below, and search it again: the count shows a field or a project.
- Write one aim, then the objectives or sub-questions that together answer the main question.
- If the design tests a prediction, add a hypothesis for each question you will test.
- Hold the set against the handbook, whose rules Check your paper lists for you, then take it to your supervisor.
What is the difference between an aim, an objective, a research question and a hypothesis?
The aim is the purpose, the research question is what the project will answer, the objectives are the tasks that get there, and a hypothesis is a prediction the data can test. Sheffield Hallam University Library (n.d.) calls the aim a mission statement, one per project; the University of Hull Library (2026) calls a hypothesis a statement of how you expect something to behave, which the research tests.
| Statement | What it does | How it is written | How many |
|---|---|---|---|
| Aim | States the overall purpose | "This project aims to…" | One, at Sheffield Hallam |
| Research question | Says what the project sets out to answer | An open question: how, why, to what extent | One main question, with sub-questions if needed |
| Objectives | The tasks that answer the question | Verbs such as to investigate, to explore, to assess | Usually three to five, at Sheffield Hallam |
| Hypothesis | A prediction the data could disprove | A statement about a relationship | Only if the design tests one |
Where the universities disagree
On two points. Sheffield Hallam treats objectives and questions as one thing written two ways; our glossary treats objectives as tasks and questions as what they answer. Either works if you keep to one.
On hypotheses, National University (2026), writing for doctoral students, requires a null and an alternative hypothesis for every quantitative question. Hull finds questions commoner in the humanities, social sciences and business and hypotheses in the sciences, and leaves the choice to you and your supervisor; the University of Melbourne (n.d.) says not all research has a hypothesis. Follow your department: a US doctoral committee and a UK undergraduate supervisor want different things.
What makes a research question researchable?
A question is researchable when it is focused enough for the word count, open enough to have more than one answer, and answerable with evidence you can reach. The universities agree on the substance.
- Clear and focused. George Mason University Writing Center (2018) wants a question narrow enough to answer thoroughly in the space available; Hull, one that is not vague or huge.
- Answerable with evidence you can reach. Hull asks whether the data exists and can be analysed in time; Monash, whether there are enough refereed articles and academic books. National University's poor example surveys FBI agents, who are hard to list and unlikely to answer a stranger.
- More than one possible result. Melbourne wants a question different results could answer; George Mason, one that is arguable rather than accepted fact.
- Analytical, not descriptive. Monash prefers how and why to what; George Mason's too-simple question could be answered from a web search in a few sentences.
- Feasible and ethical. Melbourne offers the FINER criteria: feasible, interesting, novel, ethical and relevant.
Yes-or-no questions, and when they are fine
A yes-or-no question makes a weak main question. Hull calls open questions the best form, Monash wants more than a yes or no, and National University rewrites whether performance-based pay improves student achievement as "to what extent, if any" it does. The exception is a design that tests a prediction: a hypothesis is supported or not, though Hull notes the answer still says how far and for whom. Keep the main question open and let the hypotheses carry the tests. And keep it distinct from the questions you will ask participants: National University warns that students confuse the two.
How do qualitative, quantitative and mixed-methods questions differ?
By what they ask for. For National University (2026), a quantitative question states a testable relationship between variables, while a qualitative question signals open, exploratory inquiry and carries no hypothesis.
Qualitative research questions
A qualitative question asks how people experience, understand or do something, names the people and the phenomenon, and leaves the answer open. Ours: How do first-year undergraduates living in university halls describe the part social media plays in their bedtime routines?
Quantitative questions and their hypotheses
A quantitative question names what will be measured, and in whom. Ours: Among full-time undergraduates at one university, what is the relationship between time spent on social media in the hour before bed and self-reported sleep quality? Its null hypothesis is that there is no relationship; its alternative hypothesis is that more time on social media before bed goes with poorer sleep quality. Both meet Melbourne's tests: statements, clear in scope, and disprovable.
Mixed-methods questions
Wu et al. (2019) recommend three: a quantitative question, a qualitative question, and a mixed-methods question that only the two strands together can answer, adding that the strands need not contribute equally. Ours: How do students' own accounts of bedtime social media use help explain its relationship with their sleep quality?
Which question frameworks help: PICO, PEO or SPIDER?
A question framework breaks a question into parts, so each can be defined and searched: PICO for the effect of an intervention, PEO for an exposure and its outcome, SPIDER for a qualitative question. Built for evidence-based practice and systematic reviews, they help most in health, social care and literature-based dissertations; a history question rarely fits one, and none is compulsory.
| Framework | Stands for | Suits | Our example |
|---|---|---|---|
| PICO | Population, intervention, comparison, outcome | The effect of an intervention | In adults with type 2 diabetes, what effect does nurse-led telephone follow-up, compared with usual clinic review, have on HbA1c at six months? |
| PEO | Population, exposure, outcome | Causes, risks and associations | In hospital nurses, how is rotating night-shift work associated with sleep quality? |
| SPIDER | Sample, phenomenon of interest, design, evaluation, research type | Qualitative and mixed-methods questions | International postgraduates; their experience of academic writing support; interviews; views and experiences; qualitative |
Where each framework comes from
PICO is usually traced to Richardson et al. (1995), an editorial in ACP Journal Club. Its publisher's page sits behind a bot check and could not be read, so the elements above are the Cochrane Handbook's (Thomas et al., 2024), which adds that each need not be defined equally precisely. PEO is set out by Moola et al. (2015) for the Joanna Briggs Institute: questions of aetiology, they write, do not fit PICO and need a population, an exposure and an outcome. SPIDER was proposed by Cooke et al. (2012) for qualitative and mixed-methods research, and its authors said it still needed testing on a wider range of topics.
What a comparison of the frameworks found
Methley et al. (2014) ran one qualitative question through PICO, PICOS and SPIDER in three databases. PICO returned 23,758 hits and 18 relevant papers; PICOS 448 hits and SPIDER 239, each with 13. They advise PICO for a comprehensive search, PICOS where time is short, and not SPIDER yet, for the papers it can miss.
How do I turn a broad topic into one research question?
Narrow it one limit at a time, and check each version against what has been published. Monash suggests narrowing by subtopic, period, place, organisation or group.
Our worked example: from social media and sleep to one question
On 26 September 2026 we narrowed a topic this way in EdCitation's Find sources.
- The topic. Social media and sleep found 366,069 published works. First by relevance was Alonzo, Hussain, Stranges and Anderson's 2020 systematic review in Sleep Medicine Reviews, cited 455 times. That is a field, not a dissertation.
- The narrowed search. Bedtime social media use, sleep quality and university students, from 2021 onwards, found 5,960. First came Kolhar, Kazi and Alameen's 2021 article in the Saudi Journal of Biological Sciences on social media and sleep duration among university students, cited 312 times. Fourteen of the first 100 were review articles: the reading starts there.
- The question. Our quantitative version adds one university and one hour before bed: a project one student can survey.
Thousands of results still include loosely related work; what matters is that the first pages are about your question and none answers it in your exact terms. The filters are explained in finding sources for a literature review fast.
The aim and objectives that follow
- Aim: to examine how bedtime social media use relates to sleep quality among undergraduates at one university.
- Objective 1: to review the evidence on screen use before bed and students' sleep.
- Objective 2: to measure bedtime social media use and self-reported sleep quality in a sample of undergraduates.
- Objective 3: to test the association, allowing for year of study and paid work.
- Objective 4: to set out what the findings suggest for student support services.
Four objectives, within Sheffield Hallam's three to five, each a verb and each visibly done or not.
How many research questions and objectives should a dissertation have?
Usually one main question and a few objectives or sub-questions, but your handbook's figure is the one that counts.
| University | What it suggests |
|---|---|
| Sheffield Hallam (UK) | One aim; usually three to five objectives |
| National University (US, doctoral) | A single central research question where possible; sub-questions allowed; keep the number down to avoid scope creep |
| Hull (UK) | A main question broken into sub-questions; its example, on online privacy, has four |
None of the seven sets a maximum or a number of hypotheses. If the objectives keep multiplying, the question is too wide; National University calls a far too broad question "boiling the ocean".
How does the research question shape the literature review and methodology?
It decides both: the review shows why the question needs asking, the methodology how it will be answered.
- The literature review. Hull says that, unless the method is a systematic review, the question grows out of the literature, since only what has and has not been done can justify it. The review ends at the gap it fills; see how to write a literature review.
- The methodology. National University says questions drive data collection and must align with the method: how people describe something points to interviews, a relationship to measurement. The chapter on method argues for that match.
- The rest. UNSW Sydney (n.d.) states the aim after the gap in the introduction, reports results against the question, and ends by showing the aims were achieved.
- The proposal. Every source behind the question needs a reference; see research proposal references.
How EdCitation helps with research questions
EdCitation is the best tool for testing a draft question against the literature, because every result in Find sources is a record of published work, not a list a chatbot composed, so the count you narrow by counts real papers: 366,069 for our topic, 5,960 narrowed. It covers about 300 million works, by topic or by claim, with filters for kind of source, years and open access, free and with no account.
Cite a source builds each paper's entry from a DOI, title, PubMed ID or ISBN, in any of six styles (APA 7, MLA 9, Chicago 18 author-date, Harvard, IEEE, Vancouver), and checks the paper for retraction. Given the SPIDER paper's DOI, it returned, in a rerun on 27 September 2026:
Cooke, A., Smith, D., & Booth, A. (2012). Beyond PICO: The SPIDER tool for qualitative evidence synthesis. Qualitative Health Research, 22(10), 1435–1443. https://doi.org/10.1177/1049732312452938
The publisher deposited the title in title case, and Cite a source set it in sentence case, as APA 7 wants, keeping PICO and SPIDER in capitals. The PICO editorial's publisher filed its four authors as editors; Cite a source reads them as the authors, as PubMed lists them. Read every entry against the paper.
Check your paper turns your handbook's requirements into a checklist, free, showing beside each rule the handbook sentence it came from. Judging a draft against that brief is Theoretics QA (Max, $24 a month), which quotes your draft and never rewrites it. The plans are on Pricing; paying for a year at once saves 25%. The question, aims and hypotheses stay yours: no tool here writes them, nor can any promise a mark.
Quick questions
What is the difference between a research aim and a research question?
The aim states the project's purpose; the research question, exactly what it will answer. Sheffield Hallam calls the aim a mission statement.
How many research questions should a dissertation have?
Usually one main question with a few sub-questions or objectives: Sheffield Hallam suggests three to five objectives, National University one central question. Your handbook's figure wins.
Can a research question be a yes or no question?
Not as the main question: Hull, Monash and National University all advise an open one. A hypothesis is supported or not, but the answer should still say how far.
Do I need a hypothesis in a qualitative dissertation?
Usually not. National University says qualitative questions carry no hypotheses, and Melbourne that not all research has one.
Can EdCitation help me narrow a research question?
Yes. EdCitation's Find sources shows how many published works match a draft question, with filters for years, kind of source and open access, free. It never writes the question.
References
- Cooke, A., Smith, D., & Booth, A. (2012). Beyond PICO: The SPIDER tool for qualitative evidence synthesis. Qualitative Health Research, 22(10), 1435–1443. https://doi.org/10.1177/1049732312452938
- George Mason University Writing Center. (2018). How to write a research question. https://writingcenter.gmu.edu/writing-resources/research-based-writing/how-to-write-a-research-question
- Methley, A. M., Campbell, S., Chew-Graham, C., McNally, R., & Cheraghi-Sohi, S. (2014). PICO, PICOS and SPIDER: A comparison study of specificity and sensitivity in three search tools for qualitative systematic reviews. BMC Health Services Research, 14, Article 579. https://doi.org/10.1186/s12913-014-0579-0
- Monash University Library. (n.d.). Developing research questions. https://www.monash.edu/library/help/assignments-research/developing-research-questions
- Moola, S., Munn, Z., Sears, K., Sfetcu, R., Currie, M., Lisy, K., Tufanaru, C., Qureshi, R., Mattis, P., & Mu, P. (2015). Conducting systematic reviews of association (etiology): The Joanna Briggs Institute's approach. International Journal of Evidence-Based Healthcare, 13(3), 163–169. https://doi.org/10.1097/XEB.0000000000000064
- National University. (2026, June 23). Research questions/hypotheses. Dissertation and applied doctoral project components. https://resources.nu.edu/c.php?g=1446805&p=10913030
- Richardson, W. S., Wilson, M. C., Nishikawa, J., & Hayward, R. S. (1995). The well-built clinical question: A key to evidence-based decisions. ACP Journal Club, 123(3), A12–A13. https://doi.org/10.7326/ACPJC-1995-123-3-A12
- Sheffield Hallam University Library. (n.d.). Introduction. Dissertations and research projects. https://libguides.shu.ac.uk/researchprojects/introduction
- Thomas, J., Kneale, D., McKenzie, J. E., Brennan, S. E., & Bhaumik, S. (2024). Chapter 2: Determining the scope of the review and the questions it will address. In J. P. T. Higgins, J. Thomas, J. Chandler, M. Cumpston, T. Li, M. J. Page, & V. A. Welch (Eds.), Cochrane handbook for systematic reviews of interventions (Version 6.5). Cochrane. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-02
- University of Hull Library. (2026, August 17). Research questions. Dissertations & projects. https://libguides.hull.ac.uk/dissertations/questions
- University of Melbourne. (n.d.). Shaping the research question and hypothesis. Academic Skills. https://students.unimelb.edu.au/academic-skills/graduate-research-services/writing-thesis-sections-part-2/shaping-the-research-question-and-hypothesis
- UNSW Sydney. (n.d.). How to write an honours thesis. https://www.unsw.edu.au/student/managing-your-studies/academic-skills-support/toolkit/honours-postgraduate
- Wu, Y. P., Deatrick, J. A., McQuaid, E. L., & Thompson, D. (2019). A primer on mixed methods for pediatric researchers. Journal of Pediatric Psychology, 44(8), 905–913. https://doi.org/10.1093/jpepsy/jsz052