Research essentials
What research is and why it is done
Questions, designs, data, analysis and outputs: the vocabulary you need before you recruit anyone.
You do not need to become a researcher to be a link worker. You do need to know what the team is doing, why each step happens, and where your part fits. This page gives that vocabulary in plain terms.
What research is
Research is a planned attempt to answer a question with evidence, in a way that another person could check. The check is the difference between research and opinion. If you cannot say how you would know you were wrong, you do not yet have a research question.
Three things that make it research
- A question that can be answered. Not "how do we improve care", but "what stops people from attending a first appointment".
- Evidence collected in a stated way. Who was asked, how, when, and what happened to people who did not take part.
- A record that can be inspected. Consent, data, decisions and changes all leave a trail.
The four families of question
- What is happening? Description and prevalence. Surveys, audits, routine data.
- Why and how does it happen? Meaning, experience and mechanism. Interviews, focus groups, observation, photovoice.
- Does it work? Effect. Trials and comparisons.
- How do we make it work here? Implementation. Mixed methods, process evaluation, co-design.
Link workers most often work on the second and fourth. Both depend on access and interpretation. Both are damaged by recruitment from a convenience sample that the team calls a community.
Involvement, engagement and participation are not the same thing
Three terms NIHR separates
- Involvement means working in partnership with the public on the research: shaping questions, design, delivery and dissemination. A link worker is an involvement role, and more than that: a team member.
- Engagement means sharing information and listening. A talk, a stall, a newsletter.
- Participation means taking part in a study as a participant: giving data.
Costs follow the same distinction. NIHR guidance says clearly that a focus group of study participants is a participation cost, not an involvement cost, and the budget must separate them. A link worker who recruits participants is doing involvement work to enable participation.
The life of a study
Seven stages, and where you can contribute
- Question and priorities. What should be studied. Link workers can change this, and it is the highest-value place to be.
- Design. Who is eligible, what is asked, where, when, in what language, by whom.
- Approvals. Ethics and information governance. The team writes, the community shapes the wording and the process.
- Recruitment. Who is invited and how. This is usually the link worker's main task.
- Data collection. Interviews, groups, questionnaires, photographs, measurements.
- Analysis. Turning records into findings. Link workers contribute cultural interpretation and challenge.
- Reporting and dissemination. What is said, to whom, in what form, in which language.
Qualitative and quantitative, in one page
Quantitative work counts. Qualitative work explains. Most community research uses both, because a number without a reason is hard to act on, and a reason without a number is easy to dismiss.
What good qualitative work needs
- A stated sampling approach: who was chosen, why, and who was not.
- An interview or group guide, used flexibly.
- Recording and a written record.
- A documented method of analysis.
- Enough participants to reach saturation, with the reasoning stated.
Words you will hear
Twelve terms, defined once
- Protocol. The written plan for the study, approved before work starts.
- Eligibility criteria. The rules for who can take part.
- Consent. A person's informed, voluntary agreement to take part, recorded.
- Participant information sheet. The document that explains the study.
- Sampling. How people are chosen to be invited.
- Saturation. The point where new interviews stop producing new themes.
- Coding. Labelling pieces of data so that similar pieces can be grouped.
- Theme. A pattern across data that answers part of the question.
- Pseudonym and anonymisation. Removing identifying detail so a person cannot be recognised.
- PPI or PPIE. Patient and public involvement and engagement.
- Lay summary. A plain-language account of the study and its findings.
- Governance. The rules and records that keep the study lawful and accountable.
What you are not expected to know
You are not expected to write the protocol, run the statistics, manage the ethics application or hold the data protection officer's role. You are expected to know enough to ask a good question, spot a problem that the team has missed, and refuse a task that is outside your training. That last one is a competence, and it is assessed.
The test of understanding
A researcher says: "We need 20 people from the community for a focus group next month. Can you get them?" What is missing from that request?
At minimum: the question being asked, why a group rather than interviews, who is eligible, what participants will be told, how consent will be taken, what payment and expenses apply, and what the data will be used for. A link worker who accepts the request as written cannot recruit well. A link worker who asks these questions is doing the job.