Home / Glossary / Stratified Sampling
Stratified Sampling
Stratified sampling divides the population into mutually exclusive subgroups (strata) and samples randomly within each, ensuring proportional or oversampled representation.
Stratified sampling improves precision when the variable of interest differs across known subgroups (segments, regions, plan tiers). The researcher splits the population into strata, then draws a random sample from each — either proportionally (matching the population mix) or with deliberate oversampling of small but important segments. Compared to simple random sampling at the same n, stratified samples have smaller standard errors when within-stratum variance is lower than between-stratum variance. It's the most common probability sampling approach in B2B CX research.
Formula
n_h = n × (N_h / N) for proportional allocation, where h = stratumExample
A SaaS company has 70% SMB / 25% mid-market / 5% enterprise customers. For a sample of n = 400 with proportional allocation: 280 SMB, 100 mid-market, 20 enterprise. They may oversample enterprise to n = 80 to get usable segment-level estimates.
Run a Stratified Sampling survey free
SpaceForms includes pre-built templates with validated wording. Unlimited responses on the free tier — no credit card.
Start freeRelated terms
Random Sampling
Random sampling is a probability sampling method where every member of the target population has an equal, known chance of being selected.
Cluster Sampling
Cluster sampling divides the population into naturally occurring groups (clusters) and randomly selects entire clusters, then surveys all or some members within them.
Sample Size
Sample size is the number of respondents needed for survey results to be statistically meaningful at a given confidence level and margin of error.
Margin of Error
Margin of error is the range within which the true population value likely falls, given the sample size and confidence level.
Panel Survey
A panel survey repeatedly surveys the same pre-recruited group of respondents over time, enabling within-person change analysis.
Confidence Interval
A confidence interval is the range within which a true population parameter is expected to fall, at a specified confidence level.