Respondent

A respondent is a person who answers the questions in a survey, questionnaire or interview. In a research paper, the respondents of the study are the people the researcher actually collected answers from: who they are, how many there are and how they were selected.

The word comes from "respond", to answer. One person is a respondent, and the group is the respondents. A survey sent to 500 customers and answered by 380 of them has 380 respondents. The 120 who did not answer are non-respondents: they lower the response rate and can bias the results if they differ from those who answered.

Respondent, participant or subject: which word to use

Population, sample and respondents

Theses often mix up three different groups:

• Population: everyone your conclusions are about, for example all 1,200 third- and fourth-year nursing students at a university.
• Sample: the part of the population you selected and invited, for example 300 of those students.
• Respondents: the people from the sample who actually returned usable answers, for example 254 students.

How far you can generalize depends on how the sample was drawn. With probability sampling, every member of the population has a known chance of being selected. Random selection of respondents should not be confused with random assignment, which happens later and only in experiments: people already in the study are split at random into groups, such as a treatment group and a control group.

How to write the respondents of the study section

In a thesis or research paper (often in Chapter 3, Methodology), the respondents section answers five questions:

1. Who: the population and the inclusion criteria, such as "Grade 11 students enrolled in the 2025–2026 school year".
2. Who was excluded: and why, such as students on leave during data collection.
3. How many: the sample size and how you calculated it, for example with Slovin's formula or a 95% confidence level and a 5% margin of error.
4. How they were selected: the sampling technique, such as simple random, stratified, purposive or convenience sampling.
5. Ethics: how you obtained informed consent and kept answers confidential or anonymous.

A short example: "The respondents of the study were 254 third- and fourth-year nursing students of X University. Using Slovin's formula with a 5% margin of error, a sample of 300 was drawn from the population of 1,200 by stratified random sampling by year level. Of these, 254 returned complete questionnaires, a response rate of 85%."

More templates and wording for different study types are in the guide How to Write the Respondents of the Study Section in Chapter 3.

Types of respondents

By the role they play in the study:

• Individual respondents answer about their own experience, opinions or behavior. Most surveys work with them.
• Proxy respondents answer for another person who cannot answer, such as a young child or a patient.
• Expert respondents answer because of specialist knowledge, as in expert interviews or the Delphi method.
• Organizational respondents answer on behalf of a company or department, as in B2B research.

By how they answer, which matters for data quality:

• Attentive respondents read the questions and answer thoughtfully.
• Satisficers give "good enough" answers with little effort: speeding, picking the first option or the same column in every row of a matrix.
• Fraudulent respondents fake answers or eligibility, often for a reward, including bots and duplicate entries. See fraud detection.

Who should not be a respondent

• People outside the target audience. Screening questions at the start filter them out (see screenout).
• Minors without a parent's or guardian's consent. Children also give their own assent.
• People who cannot give informed consent, unless the study has a specific ethical approval for that.
• People with a conflict of interest, such as employees of the company that ordered the study, or people who work in market research or advertising. Commercial surveys often ask a screening question about this.

How respondents are surveyed

The three main methods:

• Questionnaire: a fixed set of questions answered online, on paper or by phone. It suits large groups and numerical analysis.
• Interview: a researcher asks questions face to face, by phone or by video. Interviews can be structured, semi-structured or unstructured and suit in-depth topics.
• Focus group: a moderated discussion with a small group, useful when interaction between people brings out opinions and reasons.

For where to get respondents (your own customers, employees, classmates, social media or a paid panel) and how to pick a channel, see Where to Find Survey Respondents.

Common mistakes

• Calling everyone you invited "respondents". Report invited, started and completed numbers separately.
• Describing the respondents without the selection method. "200 respondents" says little until the reader knows how they were chosen.
• Treating a convenience sample as representative. Classmates or followers who answered a shared link do not stand for a whole population.
• Ignoring non-respondents. If the people who did not answer differ from those who did, results carry non-response bias.