Research questions

Why a good research question matters

A well-crafted research question does more than tell you what to investigate - it gives your whole project direction. It keeps you focused and stops your research from wandering into tangential areas. It shapes the methodology you’ll use to collect and analyse data, helps you identify relevant literature and recognise gaps in the field, sets boundaries so your project stays achievable within your timeframe and resources, and helps readers immediately understand what your research is about and why it matters.

Features of a good research question

So what makes a question capable of doing all that? A good research question is:

  • Clear and focused - A strong research question has precise, well-defined concepts that readers can easily understand.
  • Appropriately scoped - Your question should be narrow enough to be manageable within your timeframe and resources, but broad enough to be substantive. For example, “What causes stress in university students?” is too broad, while “How stressed are international students in week 1 of semester in a first-year chemistry unit?” is too narrow. If you were looking at stress in international students, you’d probably need to narrow it down by year level or stage of their studies, without going as narrow as a single week of a single unit.
  • Researchable - Make sure your question can be investigated using available research methods and data, and fits with your chosen methods. For example, “How does stress change for international students over the course of a 3-year degree?” may not be feasible if you were planning to answer it through a single round of interviews.
  • Significant - The question should address an important issue or gap in the field, contributing meaningful knowledge. “What font do students at University X prefer for the library website?” is not significant.
  • Appropriately complex - The question needs to be complex enough to require analysis and interpretation, but not overly complex, for example with too many variables. “How stressed are international students?” can be answered with a simple measure. On the other hand, if you were intending to look at workload, cultural background, family dynamics, social media use, and physical fitness, as well as comparing different age groups, you would probably need to simplify it.
  • Neutral - Your question should not contain assumptions or bias, or suggest a particular answer. For example, there is an implicit bias in the question “How does poor teaching cause stress among international students?”, since it assumes teaching quality is the cause before the research has established this.
  • Aligned with methodology - The way a research question is worded will often imply a certain methodological approach, so make sure it aligns with the research methods you intend to use. For example, “What is the prevalence of stress among international students?” would lead to a quantitative approach, while “What are international students’ experiences of stress?” would lead you to use qualitative methods.

Self-check checklist

Once you have a draft research question, check it against each of the criteria above:

☐ Is it clear and focused, using well-defined concepts?
☐ Is it appropriately scoped - not too broad, not too narrow?
☐ Is it researchable with the methods and data available to you?
☐ Is it significant - does it address a genuine gap or issue?
☐ Is it appropriately complex - neither trivial nor unmanageable?
☐ Is it neutral, free of built-in assumptions or bias?
☐ Is it aligned with your intended methodology?

If you can tick every box, you likely have a strong draft. If not, that’s useful information about what to refine next.

Common pitfalls to watch for

A few patterns come up again and again in early drafts of research questions:

  • Disguised yes/no questions - e.g. “Does stress affect international students’ academic performance?” This can usually be answered with a single word. Reframing it as “How does stress affect…?” or “In what ways does stress affect…?” opens it up to the kind of analysis a thesis requires.
  • Already-answered questions - if the literature review turns up a clear, settled answer to your question, it isn’t going to sustain a thesis. Look for what’s still unresolved or under-explored.
  • Bundled questions - a question that is really two or three questions joined together (e.g. “What causes stress in international students, and how effective are the support services designed to address it?”) is hard to answer coherently. Split it into a main question and sub-questions, or narrow your focus.
  • Loaded or leading questions - questions that presume the answer (covered under “Neutral” above) or steer the reader toward a particular conclusion.