Contents
- What is a representative sample, and why would I want one?
- What representative samples are available on Prolific?
- Where does Prolific get its census data from?
- How is your data stratified, and what subgroups are available?
- How much do representative samples cost?
- What are the limitations of a representative sample?

This article explains representative samples: what they are, which types are available on Prolific, where we source our data, and how we organize participant subgroups.

Once you’ve read this article, we recommend you head over to Using representative samples on Prolific - FAQs to learn more about how to use representative samples on our platform.

What is a representative sample, and why would I want one?

A representative sample mirrors the demographic distribution of a given (often national) population, making your research findings more generalizable.

• If you're a psychologist studying loneliness (or any other construct), a representative sample helps you better estimate how loneliness affects the entire country.
• If you're a market researcher gathering feedback on product concepts, a representative sample ensures your survey includes participants from diverse backgrounds, giving you clearer insights into how the general public might receive your ideas.
• If you're a political scientist forecasting an upcoming poll, a representative sample allows you to collect opinions from a group that reflects the country's demographic makeup.

Using a non-representative sample limits how confidently you can claim "my findings likely apply to populations beyond the one I tested."

What representative samples are available on Prolific?

We currently offer representative samples of two national populations: the United Kingdom and the United States. We offer standard representative samples, political representative samples, and a regional sample (UK only).

Participant eligibility for representative samples is calculated on the basis of their screening answers. Across our representative samples, we use up to 5 screeners: 'date of birth', 'sex’, 'ethnicity (simplified)’, 'U.S. political affiliation' or 'U.K. political affiliation', and ‘Current UK area of residence’ or ‘Current USA area of residence’.

Where does Prolific get its census data from?

* For Scotland we have used the available 2022 census age/sex breakdowns and overlaid the census data on ethnicity from 2011.

** The 2023 census data was not split by region.

Please note that the age brackets and age/sex/ethnicity proportions used for the UK and US non-political representative samples were updated on 27th February 2024 to use the most up-to-date data.

How is your data stratified, and what subgroups are available?

Standard representative sample

• Starting with the youngest allowable participation age on Prolific, we stratify age using five brackets: 18-24, 25-34, 35-44, 45-54 and 55+.
• ‘Sex’ is stratified into male and female.
• ‘Ethnicity (simplified)’ is stratified into the five categories recommended by the UK Office of National Statistics: White, Mixed, Asian, Black and Other.

Political representative sample

• Starting with the youngest allowable participation age on Prolific, we stratify age using five brackets: 18-24, 25-34, 35-44, 45-54 and 55+.
• ‘Sex’ is stratified into male and female.
• 'Political affiliation' for the US is stratified into Republican, Democrat, and Independent.
• 'Political affiliation' for the UK is stratified into Conservative, Labour, LibDem, Green, SNP, Reform and other. ‘Other’ is compiled of all other UK parties not listed, as well as participants who have either indicated they don’t know which party to vote for or that they don’t intend to vote.

Regional representative sample

• Starting with the youngest allowable participation age on Prolific, we stratify age using five brackets: 18-24, 25-34, 35-44, 45-54 and 55+.
• ‘Sex’ is stratified into male and female.
• Region:
  • UK - ‘Current UK area of residence’ is stratified into:
    • East Midlands, England (Derbyshire and Nottinghamshire, Leicestershire, Rutland and Northamptonshire, Lincolnshire)
    • East of England (East Anglia, Bedfordshire and Hertfordshire, Essex)
    • London, England
    • North East, England (Tees Valley, Durham, Northumberland and Tyne and Wear)
    • North West, England (Cumbria, Greater Manchester, Lancashire, Merseyside)
    • Northern Ireland
    • Scotland
    • South East, England (Berkshire, Buckinghamshire, and Oxfordshire, Surrey, Sussex, Kent, Hampshire and Isle of Wight)
    • South West, England (Gloucestershire, Wiltshire and Bristol/Bath area, Dorset and Somerset, Cornwall and Isles of Scilly, Devon)
    • Wales
    • West Midlands, England (Herefordshire, Worcestershire and Warwickshire, Shropshire and Staffordshire, West Midlands)
    • Yorkshire and the Humber, England (East Riding, North Lincolnshire and Yorkshire)
  • US - ‘Current USA area of residence’ is stratified into:
    • Midwest
    • Northeast
    • South
    • West

How much do representative samples cost?

When you use a representative sample, we don't charge any additional fees, other than our standard platform fee.

For an accurate quote, create a new study, set your sample size (minimum 300 for a representative sample), and state how many minutes your study takes to complete. There is further guidance on this here: Setting up a study on Prolific.

What are the limitations of a representative sample?

Using a representative sample means the distribution of age, sex, and ethnicity in your final sample will closely mirror the selected national population. However, remember that no sample can ever be 100% representative. While using a representative sample improves generalizability, it doesn't guarantee that your results will perfectly represent the entire population.

For example:

• Sample size affects how precisely we can match national population proportions. If a subgroup makes up 0.2% of the national population and your sample size is 300, that subgroup should technically be represented by 0.6 of a participant. Our stratification algorithm ensures at least one participant represents each subgroup, but this means smaller sample sizes won't perfectly match national proportions. In some cases, a single participant might represent an entire subgroup.
• While a representative sample matches the national population on age, sex, and either ethnicity or political affiliation, it doesn't mean the sample represents all demographic characteristics. For instance, all participants are Prolific members, which isn't true for everyone in the United Kingdom or United States. We plan to add more demographic factors to our representative samples in the future, such as socio-economic status and education level.

Want to know more?

Head over to Using representative samples on Prolific - FAQs to learn more about how to use representative samples on our platform. It covers sample sizes, prescreener compatibility, participant recruitment methods, completion timeframes, options for slow-filling samples, and how to download your data once the study is complete.

Need further help?
Click here to contact us