What makes a good research question?
This is not the same as asking, "Is the topic interesting?"
It entails particular criteria:
- Is it clear and specific? Does it define a topic that is not ambiguous, vague?
- Ambiguous - "Are little penguins at more risk than Emperor penuins?" - does this refer to the species "Little Penguins" or any penguin that is physically smaller than the Emperor penguin?
- Vague - "Are cats better pets than dogs?" What does "better" mean? If a question does not specify relevant time or space parameters, it may be too vague. e.g. "What are the main causes of poverty?" or "Why do wars occur?"
- Is it focused, and is its scope clearly defined?
- Will researchers know what research is relevant and what is not?
- Too broad a scope may be impossible to fully research in terms of time, resources, money and labour.
- Too specific a scope may mean the question is trivial and not worth investigating.
- Is there a definite sign that the question has been satisfactorily answered?
- Is there a single topic or a confused combination of issues? e.g. "What causes autism and can it be treated better?"
- Are its terms defined clearly? e.g. "How does internet use cause mental problems in young people?" - which aspects of the internet does it mean? What mental problems are included? What does "young" mean - six, sixteen, twenty-six?
- Author's note: at my age, 26 year-olds are young people.
- Is it researchable?
- Some questions, while fascinating, cannot be scientifically researched. Is there life after death? Does god exist? If a tree falls in the woods unobserved, does it make a sound?
- Other questions may have little or no reliable data available, e.g. do fish feel pain? The question should be answerable using primary and/or secondary data. If there is no data, the question is probably invalid.
- Can the question be analysed?
- Good research questions ask How...? or Why...? and have a specific answer.
- Questions that can be answered simply by describing or defining are not worthwhile. e.g. "What are the characteristics of autism?" or "How do dolphins differ from cats?"
- Questions that can be answered with just "Yes" or "No" are not good research questions.
- Compare "Does pet ownership help older people?" with "To what degree and in what ways does pet ownership help older people?"
- Questions should not be only answerable with opinions or unsupported guesses. Any question asking for a value judgement is not a valid or worthwhile research question. "Are men better than women?" is invalid because - seriously - you already know what the answer is.
- Is the question worthwhile?
- Some questions are interesting, but are not worth the time and effort of research, e.g. Why are fire trucks red? How are Adam Sandler and Bruce Willis still allowed to make movies?
- Good research questions often provoke debate and are not boring or banal. Compare "Are first-person shooter video games making players more violent?" with "Are video games becoming more violent?"
- Is the question original?
- Sure, researchers occasionally replicate previous studies to confirm or refute their findings, but this is not a good research question.
- Does your research question raise any issues that have not already been answered?
A final point: Does the question have hidden motives? This is not an issue discussed by other sources, but I think it's worth mentioning.
- What is the motive behind the question? Does it have a purpose beyond merely investigating a fact-based issue?
- Does it intend to promote a cause, advance a philosophy or religion, persuade people toward a viewpoint, advocate changes to government policy, raise money?
- Are its findings likely to be biased? Will it cherry-pick data or aim to be deliberately deceptive? Always consider the motives of any research question by an organisation that is passionately dedicated to a cause. Rarely will such findings will be published if they go against the aims of the organisation. The RSPCA may research "What are the benefits of shooting stray cats?" but will they publish any positive findings?
- A good research question leads to quantified observations on a given topic. Its findings might well be used to lead to change, but it should not - in itself - argue an opinion or promote a cause. It should ask a question, give relevant data to support a conclusion, and end. Interpretation of the research question's findings should be left to other people.
|
Problem Statements