Market Research Survey Sampling Simulator

Population

True preference for Product A 50%

Sample size (n) 30

Confidence level

Draw a Survey

Latest Sample

No sample drawn yet.

Std. error (SE)

—

Margin of error

—

Samples drawn

Mean of estimates

—

How it works

A market-research survey never asks every consumer — it draws a random sample of size n from the population and uses the sample proportion p̂ (share who prefer product A) to estimate the true population proportion p.

Standard error:  SE = √( p̂(1−p̂) / n )
Margin of error: MOE = z · SE
Confidence interval: p̂ ± MOE

z is the critical value for the chosen confidence level (1.645 / 1.96 / 2.576 for 90% / 95% / 99%).
Each dot in the population grid is one simulated consumer, coloured by their true (hidden) preference. "Draw one sample" highlights a random subset — the respondents of one survey — and computes p̂ from just that subset.
The bars on the right accumulate every p̂ you've drawn. By the Central Limit Theorem, that histogram converges toward a bell curve centred near the true p as more samples are drawn — this is why researchers trust a sample instead of surveying everyone.
Increasing n shrinks SE (and the margin of error) roughly as 1/√n: to halve the margin of error you need four times the respondents — the core cost/precision trade-off in survey design.

Real-world relevance: this is the exact statistical model behind opinion polling, product-preference surveys, A/B test sample-size planning, and any market-research report that quotes a "±X% margin of error at 95% confidence."

Prefers Product A Prefers Product B In current sample Sampling-distribution bars

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.