Dissertation research questions and hypotheses state exactly what your study will find out. Every later choice, from design and sample to analysis and discussion, follows from them, so they deserve careful drafting.
This guide explains how to write questions for quantitative, qualitative and mixed methods studies, when you need hypotheses and how to phrase them, and how to check that your questions fit the rest of the dissertation. Examples show weak and improved versions side by side.
What Makes a Strong Research Question
- Focused: one clear population, setting and phenomenon.
- Answerable: data you can realistically collect will answer it.
- Aligned: it follows from your gap and suits your design.
- Significant: the answer matters to theory, practice or policy.
- Neutral: it does not assume the answer.
Quantitative studies usually pair questions with hypotheses. Qualitative studies usually do not.
Why Research Questions Carry so Much Weight
Examiners often read your research questions first and then check whether each chapter answers them. A vague question produces a vague methodology and a discussion that drifts.
Your questions also define the scope of the study. Anything they do not ask about is outside your remit, which protects you from the "why didn't you study X?" challenge in a defense.
The Alignment Chain
Research questions sit in the middle of a chain: problem, purpose, questions, hypotheses (if any), design, data and analysis. If you change one link, check the others. A new question may need a new instrument or analysis.
Writing Quantitative Research Questions
Quantitative questions ask about variables you can measure. They usually fall into three types.
| Type | Asks about | Example | Typical analysis |
|---|---|---|---|
| Descriptive | Levels or frequencies of a variable | What is the level of burnout among ICU nurses in the region? | Descriptive statistics |
| Comparative | Differences between groups | Do burnout scores differ between day-shift and night-shift ICU nurses? | t-test, ANOVA or non-parametric equivalents |
| Relationship | Associations or prediction | To what extent do workload and staffing ratio predict burnout among ICU nurses? | Correlation, regression |
Name the variables and the population explicitly. Avoid causal words such as "effect" or "impact" unless your design can support causal claims, typically an experiment or a strong quasi-experiment.
Writing Hypotheses
A hypothesis is a testable prediction about the relationship between variables. In null hypothesis significance testing, you write a pair.
- Null hypothesis (H0): no difference or no relationship exists in the population.
- Alternative hypothesis (H1 or Ha): a difference or relationship exists.
Example pair. RQ: Do burnout scores differ between day-shift and night-shift ICU nurses?
H0: There is no difference in mean burnout scores between day-shift and night-shift ICU nurses.
H1: Night-shift ICU nurses have higher mean burnout scores than day-shift ICU nurses.
Directional or Non-Directional?
The H1 above is directional: it predicts which group scores higher. A non-directional version would say only that the scores differ. Use a directional hypothesis when theory or prior evidence clearly supports a direction, and be consistent with your choice of one-tailed or two-tailed tests. Many advisors prefer two-tailed tests unless there is a strong reason not to.
Conventions on how hypotheses are presented vary. Some departments want every H0 written out; others accept only the alternative hypotheses. Follow your program's template.
Writing Qualitative Research Questions
Qualitative questions explore meanings, experiences or processes. They typically start with "how" or "what" and avoid yes-or-no phrasing.
- Central question: one broad question that captures the study's focus.
- Sub-questions: two to four narrower questions that break it down.
Example. Central question: How do newly qualified nurses experience their first six months on an acute medical ward?
Sub-questions: What do they find most challenging? How do they seek support? How do their views of their professional role change?
Match the wording to your tradition. Phenomenological studies ask about lived experience; grounded theory asks about processes; case studies ask how and why within a bounded case. Qualitative studies do not usually state hypotheses, because they do not aim to test predictions.
Writing Mixed Methods Research Questions
A mixed methods study usually has three kinds of question: a quantitative question, a qualitative question and a mixed methods (integration) question that asks what combining them reveals.
Example (explanatory sequential design). Quantitative: Is there a relationship between workload and burnout among ICU nurses? Qualitative: How do ICU nurses with high burnout scores describe their workload? Mixed: In what ways do nurses' accounts help explain the survey relationship between workload and burnout?
The integration question is the one students most often leave out, and it is the one examiners look for. It shows why you needed both strands.
A Step-by-Step Drafting Process
- Write your gap in one sentence.
- Write a purpose statement that addresses the gap.
- Draft a broad question that turns the purpose into a question.
- Specify population, setting, variables or phenomenon.
- Split it into two to four focused questions if needed.
- For quantitative questions, write the matching hypotheses.
- Check each question against your design, data and analysis plan.
- Read each aloud: could someone else tell exactly what data would answer it?
How Many Research Questions?
Most dissertations have between one and four main research questions, sometimes with sub-questions. More than five usually means the scope is too wide or some questions are really methods steps.
Every question must be answered in your results and discussed in your discussion chapter. If you cannot see where a question will be answered, drop it.
Aims and Objectives in UK Dissertations
Many UK programs ask for an aim and a short list of objectives instead of, or as well as, research questions. The aim states the overall purpose; each objective is a concrete step, usually starting with a verb such as "to identify", "to compare" or "to evaluate".
The same alignment rules apply. Each objective should map to a method and to a section of your findings, and your conclusion should say how far each one was met.
Before and After: Revising Weak Questions
| Weak question | Problem | Improved question |
|---|---|---|
| Does social media affect students? | Vague, causal, yes-or-no | What is the relationship between daily social media use and sleep quality among first-year undergraduates? |
| Why are nurses leaving the profession? | Too broad for one study | How do nurses who left acute care within three years describe their decision to leave? |
| How effective is the new policy? | "Effective" undefined | Did 30-day readmission rates change after the discharge policy was introduced? |
| What do teachers think about technology? | No setting or focus | How do secondary maths teachers in one district use tablets in lessons? |
Common Committee Objections
- "This is answerable from the literature." Your question must need new data or analysis.
- "Your design cannot answer this." Change the question or the design.
- "The variables are not defined." Add operational definitions in Chapter 1 or 3.
- "This is two questions." Split compound questions joined by "and".
- "The question presumes the answer." Remove loaded words such as "barriers" unless the gap justifies them.
How DissertationHorse Helps with Research Questions and Hypotheses
Send your topic, problem statement, the gap you have identified, your intended design and any committee feedback. A writer at your level drafts custom research questions, matching hypotheses where appropriate, and an alignment table linking each question to data and analysis, for you to study and adapt.
Free revisions within the scope of your original request are included, and your identity is never shared with the writer.
Get Research Questions Your Committee Will Accept
Send your topic, problem statement, gap and intended design. Your writer drafts custom research questions, hypotheses and an alignment table, written to that brief.
Order Research Question HelpFree revisions · Full refund if late · Written from scratch for your order
Frequently Asked Questions
Usually not. Qualitative research explores meanings and processes rather than testing predictions, so it uses a central question and sub-questions instead.
Minor refinements are common. Substantial changes usually need your chair's approval and sometimes an ethics amendment, so discuss them before you collect data.
In many dissertations they appear in Chapter 1 and are repeated in Chapter 3 with the hypotheses and analysis for each. Some templates differ, so check your program's guide.
Many committees accept it for correlational studies, though some prefer "What is the relationship..." to avoid yes-or-no wording. Follow your advisor's preference.
A question asks what you want to know. An objective states what you will do to find out. UK dissertations often use aims and objectives instead of, or alongside, questions.
If you use significance testing, the null hypothesis is the one you formally test. Some programs require it written out; others accept alternative hypotheses only.
Specific enough that a reader knows the population, the setting and what data would answer it, but not so narrow that the answer is trivial.
It states exactly how a concept will be measured or identified in your study, for example burnout as a score on a named, validated scale. It makes quantitative questions and hypotheses testable.
Yes. Many qualitative studies have one central question with sub-questions, and some quantitative studies have a single main question with several hypotheses.