Developing Research Questions

7.2.1 What is a research question?

A research question is the specific question a study intends to answer. It guides the data, the analysis and the interpretation. “How can healthcare improve?” is too broad. It does not name the service, the people, the setting or the problem.

A stronger example: what proportion of monthly outpatient reports submitted by selected departments at a teaching hospital contained all required fields during the first quarter of 2026? That names the measure, the records, the setting and the period. An actual study must use its real setting and period.

7.2.2 Five characteristics of a useful research question

• Clear: the reader understands what is being asked.
• Focused: it addresses a manageable problem.
• Answerable: suitable data and methods exist.
• Relevant: the answer could inform knowledge, practice or a decision.
• Ethically appropriate: it can be investigated without unjustified risk to people or to confidential information.

7.2.3 Types of research questions

Descriptive questions ask what, how many, how often, or what proportion. Comparative questions ask how groups differ, and they need consistent definitions. Exploratory questions investigate experiences and possible explanations, often through interviews, discussions or open responses. Evaluative questions assess a programme or process. A before-and-after comparison can show a change. It does not automatically show that the programme caused it.

7.2.4 Turn a broad problem into a research question

“Our reports are not good enough” is a problem statement. More specific questions include: what percentage of expected reports was received last quarter; which departments submitted after the deadline; which required fields were most often incomplete; and what challenges staff report when preparing the reports. The first three can be quantitative. The fourth needs qualitative information.

7.2.5 Define the scope

• The population or records of interest.
• The location.
• The reporting period.
• The variables or indicators.
• Inclusion and exclusion criteria, where they apply.
• The data sources.
• The method of analysis.

7.2.6 Example

Problem: recurring delays in monthly reporting. Question: what proportion of expected monthly reports was submitted by the deadline, and how did timely reporting vary across departments during the defined six-month period? Useful variables are department, month, expected date, actual date and submission status. Calculate the timely rate, compare departments, and look at the trend. A later study can ask why the delays happened.

Before you begin, you should be able to say exactly what you intend to find out.