Sampling, Panels & Recruitment
Sampling, panels and recruitment cover methods for finding, inviting and managing people who take your surveys and provide data.
This category groups terms about sampling approaches (probability, stratified, convenience), panels and panel management, recruitment channels and screening, quotas and incentives, and practical tactics for reaching multilingual or hard‑to‑reach communities.
Access Panel
An access panel is a pre‑recruited group of people who have agreed to be contacted for surveys, consultations or research. Organisations use access panels to recruit respondents quickly and target specific demographics or communities.
B2B Panel
A B2B panel is a pre‑recruited group of business professionals or company representatives who agree to take part in market research and surveys about products, services or sector issues. It provides a targeted audience of decision‑makers and operators rather than general consumers.
Cluster Sampling
Cluster sampling is a method where a population is divided into groups (clusters) and a sample of those clusters is selected; then everyone or a sample within chosen clusters is surveyed. It’s often used to save time and cost when a complete list of individuals isn’t available.
Consumer Panel
A consumer panel is a pre‑recruited group of people who agree to give regular feedback, opinions or test products and services. Panels are used to track attitudes, test ideas and collect ongoing insight from a defined audience.
Convenience Sampling
Convenience sampling selects participants because they are easy to contact rather than because they were randomly chosen. It's fast and inexpensive but often produces biased, non‑representative results.
Coverage Error
Coverage error happens when some people in the group you want to study have little or no chance of being included in your survey, producing a biased picture of the population. It often arises from the sampling frame, distribution channels or language and access barriers.
Demographic Quotas
Demographic quotas are rules you set in a survey to limit or target how many responses you collect from specific groups (for example by age, gender, location or language). They ensure your sample includes the mix of people you need for reliable, fair results.
Frame Error
A frame error (also called coverage error) happens when the list or method you use to reach people (the sampling frame) doesn’t match the group you want to learn from (the target population), causing some people to be left out or the wrong people to be included. This mismatch biases results and can mislead decisions.
Matched Sampling
Matched sampling is a technique for creating comparable groups in observational studies or surveys by pairing or weighting respondents with similar characteristics. It helps isolate the effect of a particular variable (for example language or treatment) when random assignment isn't possible.
Mixed-Mode Sampling
Mixed-mode sampling is a data-collection approach that uses two or more modes (for example online, phone, face-to-face or paper) to reach respondents. It helps improve coverage and response rates by meeting people where they are.
Nonprobability Sampling
Nonprobability sampling selects survey respondents who are available or choose to participate rather than being drawn randomly from the whole population. It's common in quick, low-cost surveys, outreach to specific groups, and many community consultations.
Nonresponse Bias
Nonresponse bias happens when the people who don't answer a survey differ in important ways from those who do, so the collected results misrepresent the whole group. In multilingual settings, it commonly appears when language barriers or trust issues reduce participation from particular language communities.
Online Panel
An online panel is a pre‑recruited group of people who have agreed to take part in multiple online surveys over time. Panels are used to run faster, repeatable research and track opinions or behaviours across rounds.
Oversampling
Oversampling is intentionally collecting more responses from a specific subgroup than their share of the overall population, so you have enough data to analyse that group reliably.
Panel Attrition
Panel attrition is the loss of participants from a recurring survey panel over time. It shrinks your sample and can bias results when some groups drop out more than others.
Panel Conditioning
Panel conditioning is the change in how survey panel members respond (or behave) because they take part in surveys repeatedly. Over time this can make panel answers different from those of people who haven't been surveyed as often.
Panel Fatigue
Panel fatigue is the decline in response rate and answer quality that happens when the same group of people are asked to take surveys repeatedly. Over-surveyed participants become less likely to respond and more likely to give short, careless or biased answers.
Panel Recruitment
Panel recruitment is the process of building a pre‑recruited group of people (a panel) who agree to take part in surveys or research over time. Panels let organisations reach targeted or repeat respondents quickly without finding new participants each time.
Panel Refresh
A panel refresh is the process of updating a respondent panel by replacing or supplementing members to restore representativeness, boost response rates, or improve data quality. It can be a partial top-up or a full replacement, depending on attrition and project needs.
Panel Retention
Panel retention is the proportion of people in a survey panel who continue to participate in repeated waves of research over time. High retention means fewer dropouts, more consistent samples and better longitudinal insights.
Post-stratification
Post-stratification is a survey weighting technique that adjusts collected responses so the sample matches known population characteristics (age, gender, language, region). It reduces bias from unequal response rates by giving underrepresented groups more weight and overrepresented groups less.
Pre-screening
Pre-screening is a short set of questions asked at the start of a survey to determine a participant's eligibility, route them to the right questions, or place them into quota groups. It filters or segments respondents so the main survey only reaches the people it’s intended for.
Probability Proportional to Size (PPS)
Probability Proportional to Size (PPS) is a sampling method where units (usually clusters like villages or schools) are selected with a probability proportional to a known size measure (for example population or number of households). It helps ensure larger clusters are more likely to be chosen when cluster sizes vary widely.
Probability Sampling
Probability sampling is a method of selecting survey participants so that every member of a defined population has a known, non-zero chance of being chosen. It enables statistically valid estimates and calculation of margins of error for the whole population or subgroups.
Propensity Score Matching
Propensity score matching (PSM) is a statistical method that helps make fair comparisons between two groups in non-randomised studies by matching individuals who have similar characteristics. It reduces bias from observable differences so you can estimate how an intervention or treatment is associated with an outcome.
Purposive Sampling
Purposive sampling is a non‑probability method where researchers deliberately select participants who have specific characteristics or experience relevant to the study. It’s used to ensure the sample includes the people or perspectives needed to answer a particular question.
Quota Sampling
Quota sampling is a non‑probability method that sets target numbers (quotas) for specific subgroups of a population so a survey includes enough respondents from each group. It’s used to ensure representation of key categories (e.g., language, age, location) when random sampling is impractical or too costly.
Raking
Raking (also called rim weighting or iterative proportional fitting) is a survey-weighting method that adjusts sample responses so their marginal distributions match known population totals (for example by age, gender, region). It’s used to reduce bias when the people who answered a survey don’t perfectly reflect the population you care about.
Random Sampling
Random sampling is a probability-based method for selecting people from a defined population so each person has a known (usually equal) chance of being chosen. It’s used to produce survey results that can be reasonably generalised to the whole population.
Replacement Sampling
Replacement sampling (also called sampling with replacement) is a method where each selected unit is returned to the sampling pool and can be chosen again. It keeps selection probabilities constant across draws but can produce duplicate selections in a sample.
Response Rate
Response rate is the proportion of people invited to a survey who actually provide usable answers. It’s usually expressed as a percentage and helps measure how much of your target audience you reached.
Sample Frame
A sample frame (or sampling frame) is the list or database of people, households or units you can actually contact when running a survey. It should closely match the population you want to learn about so your results are representative.
Sampling Error
Sampling error is the random difference between a survey result from your sample and the true value in the whole population. It arises because you ask a subset of people, not everyone.
Sampling Frame Construction
Sampling frame construction is the process of building a usable list or map of people (or households, organisations, etc.) from which you will select survey participants. A good sampling frame matches the group you want to hear from and helps you run a fair, efficient survey.
Screening Questionnaire
A screening questionnaire is a short set of questions used at the start of a survey or form to check whether a respondent meets the rules for participation. It quickly filters or routes people so only eligible participants continue to the main questions.
Selection Bias
Selection bias occurs when the group of people who take part in a survey or study are not representative of the population you want to learn about. That mismatch skews results and can produce misleading conclusions.
Snowball Sampling
Snowball sampling is a non‑probability recruitment method where existing study participants recruit future participants from their networks. It’s commonly used to reach hard‑to‑find or tightly knit groups when a sampling frame is unavailable.
Stratified Sampling
Stratified sampling is a method that splits a population into distinct subgroups (strata) and draws samples from each so that key groups are represented. It improves representativeness and precision compared with simple random sampling when groups differ on the things you care about.
Systematic Sampling
Systematic sampling picks every k-th item from an ordered list after a single random start. It's a simple, practical alternative to drawing a fully random sample when you have a complete list (sampling frame).
Weighting
Weighting adjusts survey results so they better reflect a target population when some groups are over- or under-represented in your responses. It gives each respondent a weight that increases or decreases their influence on reported totals and averages.
No glossary terms yet