About the Stratified Sample Allocation
This calculator determines how many survey or sample participants should be drawn from each subgroup (stratum) of a population so that every subgroup is represented in proportion to its actual size. It is designed for planning stratified sampling, where the goal is a sample that mirrors the population's composition rather than one that is fully random or evenly split.
How It Works
You enter the total sample size you want, a comma-separated list of stratum labels, and a matching comma-separated list of each stratum's population size. The calculator sums the stratum sizes to get the total population, computes each stratum's proportion of that total, and multiplies the total sample size by each proportion to get that stratum's allocated sample count, rounded to the nearest whole number.
Formula & Methodology
The number of labels entered must match the number of stratum sizes entered, and the total population across all strata must be greater than zero, or the calculator returns an error. Each stratum's allocation is rounded independently with Math.round, so the individual allocated samples may not sum to exactly the requested total sample size due to standard rounding behavior.
Examples
Allocating a market research sample across three regions
With a total sample size of 400 and stratum populations of 12,000, 8,000, and 5,000 (total population 25,000), Region A gets round(400 * 12000/25000) = round(192) = 192, Region B gets round(400 * 8000/25000) = round(128) = 128, and Region C gets round(400 * 5000/25000) = round(80) = 80.
Allocating across two unequal strata
With a total sample size of 150 and stratum populations of 9,000 and 1,000 (total population 10,000), the larger stratum gets round(150 * 0.9) = 135 samples and the smaller stratum gets round(150 * 0.1) = 15 samples.
Advantages
- Handles any number of strata at once through comma-separated lists, rather than requiring one calculation per subgroup.
- Automatically shows each stratum's proportion of the total population alongside its allocated sample count in a single table.
- Ensures the sample composition matches the population's actual subgroup structure rather than an arbitrary even split.
Common Mistakes
- Splitting the total sample evenly across strata regardless of their actual size, which over-represents small subgroups and under-represents large ones.
- Entering mismatched numbers of labels and stratum sizes, which the calculator will reject rather than silently misalign.
- Expecting the allocated sample counts to sum exactly to the requested total, when independent rounding per stratum can cause small discrepancies.
Edge Cases to Watch For
- If the count of labels doesn't match the count of population sizes, the calculator rejects the input rather than guessing an alignment.
- A total population of zero across all strata triggers an error since the proportions would be undefined.
- Because each stratum's allocation is rounded separately, the sum of all allocated samples can be off by a point or two from the requested total sample size, especially with many small strata.
Common Use Cases
- Market researchers designing a survey that needs proportional representation across regions, age groups, or other segments.
- Public health or social science researchers planning a stratified sample study across demographic subgroups.
- Analysts auditing an existing sample to check whether it matches the population's true subgroup proportions.