First steps in research

A short guide for students and doctors starting their first project.

1. Good first roles

You don't need experience to be useful. These roles teach you how a study works and are how most researchers start:

  • Screening titles and abstracts for a systematic review, with a second reviewer checking.
  • Data extraction into a shared sheet, following a written protocol.
  • Literature search support: running searches, removing duplicates.
  • Case reports: collecting the history and images (with consent) and writing the first draft with a supervisor.
  • Data collection for a retrospective study, from records you are authorised to access.

Look for projects marked Good first project: their owners expect beginners and plan for guidance.

2. How authorship works

Most journals follow the ICMJE criteria. To be an author you need all four: a real contribution to the work, help drafting or revising the paper, approval of the final version, and agreeing to be accountable for it.

  • Agree the order early and write it down; it can change if contributions change, but discuss it openly.
  • Contributions that don't meet the criteria go in the acknowledgements.
  • Each role on a project shows its expected authorship (e.g. co-author, acknowledgement) before you apply.

3. Writing a good application

  • Be specific: which role, why this project, and what you can do this month.
  • Be honest about time: say how many hours a week you can give and for how long.
  • Show what you've done: a course, a related project, a skill (Excel, R, reference managers). Little is fine; say you are new and keen to learn.
  • Keep it short: four or five sentences beat a long letter.

Example: “I'm a 3rd-year student new to research. I'd like the screening role. I can give 6 hours a week for the next 8 weeks and I've finished a Cochrane screening course.”

4. Ethics and patient data basics

  • Never share patient-identifiable information on Research buddy: no names, dates of birth, record numbers, photos of faces or anything that could identify a person. This platform is for collaboration, not clinical data.
  • Studies that use patient data or involve people usually need ethics committee approval before any data is collected. Ask your supervisor which approvals apply.
  • Case reports need the patient's written consent.
  • Keep study data in the storage your institution approves, not in personal accounts.