What a Research Problem Is

In one line

A research problem is a specific difficulty or gap in knowledge that can be resolved by investigation, and it is not a subject, not a title and not an area of interest.

In the wording a student can write in an exam: a research problem is a clearly stated difficulty, doubt or gap in existing knowledge which the researcher proposes to resolve, expressed with sufficient precision that it is possible to say what evidence would resolve it; it is distinguished from a subject or area, which merely names a field, and from a research question, which is the interrogative form in which the problem is put for investigation.

Topic, problem and question distinguished

A subject or area names a field. Legal aid. Clinical legal education. Examination reform. It has no boundaries, no question and no answer, and there is nothing that would count as having finished.

A research problem names a difficulty within that field. That the entitlement under section 12 is very wide while the number of applications is small, and it is not known where between entitlement and application the loss occurs.

A research question puts the problem in the form that will be investigated. At what point do people who are entitled to legal services stop, and in what proportions: not knowing the entitlement, not reaching the office, being refused at scrutiny, or being deterred by what they heard?

All three are needed and in that order, and MU's own design components require the last two: the research methodology is to include the identification of the research problem and the research questions, chapter 1310.

What makes something a research problem

There is a difficulty or a gap. Something is not known, is disputed, or does not work.

It is specific enough to bound. A reader can say what is inside it and what is outside.

It is capable of being resolved by evidence. Doctrinal evidence, empirical evidence or both. A problem no evidence could settle is not a research problem, chapter 1050.

It has not already been resolved. Which is why the review of literature comes early, chapter 1290.

And it matters to somebody. Not necessarily to the world; but a problem whose resolution would change nothing for anybody is a puzzle.

The five defects of a badly stated problem

Too wide. Legal aid in India. Nothing bounds it and nothing would finish it.

Too narrow to be worth doing. The number of chairs in one clinic.

Not a problem at all, but a request for a description that already exists in every textbook.

Not resolvable. Whether legal aid is just. That is a normative question, and it requires argument from a value premise rather than investigation, chapter 1050.