How much data is enough? The sampleA subset of the population that you actually collect and analyze, ideally representative of the whole. Click term for full definition size formula gives you a clear answer, so you collect neither too little nor too much. Watch the short video, then work with the toolkit PDF: it explains the method, shows a worked example and gives you a blank template to fill in.
Tool video
The short tool video will appear here.
Sample Size in short
- Choose the confidence level. 95 % is the usual standard. Then Z = 1.96.
- Set the margin of errorYour desired precision — the maximum acceptable distance between your sample result and the true population value. Click term for full definition. How close must the result be? For example ± 5 percentage points.
- Estimate p. Use an earlier rate if you have one, otherwise 0.5 for the safe maximum.
- Calculate n and round up. Put the numbers into the formula and always round up.
- Plan a random sample. Spread the sample over days, shifts and products so it represents the whole.
Key points
- More certainty or precisionHow much variation exists when the same item is measured multiple times. Click term for full definition needs more data.
- 95 % confidence: Z = 1.96.
- Unknown share: use p = 0.5.
- The sample must be random and representative.