What is Probability sampling?
Probability sampling is a way of selecting survey participants so every member of the target population has a known, non-zero chance of being chosen. Methods such as simple random, stratified, cluster and systematic sampling allow you to make statistically valid estimates about the whole population.
Probability sampling means selecting respondents by a transparent, repeatable rule that gives each person in the population a known chance of selection. Because selection probabilities are known, results can be weighted and used to calculate margins of error and confidence intervals. Common approaches: simple random sampling (every individual equally likely), stratified sampling (separating the population into subgroups and sampling within each), cluster sampling (randomly selecting groups, then individuals inside them) and systematic sampling (selecting every nth name from a list). Probability sampling requires a sampling frame (a list or way to identify the population) and usually more planning than convenience approaches, but it supports defensible, generalisable conclusions.
Usage example
A council wants to estimate resident support for a new park. They build a sampling frame from the electoral roll, use stratified random sampling to ensure neighbourhoods and major language groups are included, and invite the selected residents to complete a Hearo survey in their preferred language. Because selection probabilities are known, the council can report estimates with a margin of error and use weighting if some groups respond less often.
Practical application
Probability sampling matters when you need reliable, generalisable results rather than indicative feedback. It reduces selection bias, lets you quantify uncertainty, and supports official decisions or published findings. For multilingual projects it ensures different language communities are represented (or intentionally oversampled) so conclusions reflect the whole population, not only those comfortable responding in one language. The trade-offs are cost and complexity: probability samples often require a good population list and more outreach, but they make consultation defensible and comparable across groups.
FAQ
How is probability sampling different from non-probability sampling?
Probability sampling gives everyone in the target population a known chance of selection, enabling statistical inference and margins of error. Non-probability methods (convenience, voluntary panels) are cheaper and faster but cannot reliably estimate population parameters or quantify sampling error.
Do I always need probability sampling?
Not always. Use probability sampling when you must make generalisable estimates, produce defensible public consultation results, or compare groups formally. For quick user feedback, exploratory research, or early pilots, non-probability methods can be appropriate.
How large should my probability sample be?
Sample size depends on the precision you need, your population size, expected response rate, and the variability of the measure. A common benchmark for a large population is ~385 completed responses for a 95% confidence level with Β±5% margin of error, but subgroups, stratification and design effects can change that. Use a sample-size calculator or consult a statistician for project-specific estimates.
How do language barriers affect probability sampling?
Language barriers can cause underrepresentation if invitations or surveys are only in one language. To preserve probability properties, the sampling frame should include everyone in the target population, and invitations and the questionnaire should be available in the communities' languages (for example via Hearo). Track response rates by language, consider oversampling groups with lower response, and apply weighting to correct remaining imbalances.