Daily research tips

Write three versions of your question: one broad, one narrow, and one in between. Discuss all three with a mentor and choose based on data access and available time.

Ask your supervisor what data already exist in your department before inventing a question from scratch. Matching the question to accessible data is the fastest route to a finished project.

Size the question to your timeline. A resident with six months should ask something answerable with existing data or a small prospective sample, not something that needs years of follow-up.

Start from the outcome. Decide exactly what you will measure and how, and let that decision shape the rest of the question rather than the other way around.

Use the PICO structure of population, intervention, comparison, and outcome to turn a vague clinical curiosity into an answerable question. A question that will not fit a frame like PICO is usually not ready for a protocol.

Before committing to a research question, write it as one sentence and test it against the FINER criteria: feasible, interesting, novel, ethical, and relevant. If it fails on feasibility, fix the scope before anything else.