Standard Error Calculator
Estimate how much a sample mean would vary across repeated samples using a supplied standard deviation and sample size.
Privacy: calculator inputs are processed in your browser; Lynara does not need to upload these values to calculate the result.
Why collecting twice as many observations does not halve standard error
Suppose a class uses independent observations with a standard deviation of 12 units. The standard error of the sample mean describes sampling variation, not how scattered the individual observations are. It depends on both that standard deviation and the number of independent observations.
| Sample size | Calculation | Standard error |
|---|---|---|
| 36 | 12 ÷ √36 | 2 units |
| 72 | 12 ÷ √72 | About 1.41 units |
| 144 | 12 ÷ √144 | 1 unit |
Doubling the number of independent observations from 36 to 72 reduces the standard error by a factor of √2, about 29.3%, not by 50%. At the same standard deviation, you need four times the sample size, 144, to halve the standard error from 2 to 1. Enter each row in this calculator to verify the result before using it in homework or survey planning.
Choosing the right next calculation
If your question is how much sample averages vary, use this standard-error result. If you also have a sample mean and an appropriate confidence model, the Mean Confidence Interval Calculator uses the standard error to build a z-based interval. The interval uses a normal approximation: with a small sample and an unknown population standard deviation, a Student t method may be required instead.
A very precise mean estimate can still be systematically wrong if the sample was selected with bias. Related observations, clusters or repeated measurements can also make the simple independent-observations formula inappropriate. An observed standard deviation is an estimate and may vary between samples; this illustration holds it fixed solely to show the mathematical effect of sample size.
Formula
standard error = standard deviation ÷ √sample size
Why quadrupling sample size only halves the standard error
- Suppose an independent sample has a standard deviation of 12 units and contains 36 observations. Enter 12 and 36 to calculate a standard error of 12 ÷ √36 = 2 units.
- Under a hypothetical second design with the same standard deviation but 144 independent observations, the standard error is 12 ÷ √144 = 1 unit.
- The sample size grew from 36 to 144, which is four times as large, but the standard error fell from 2 to 1, not from 2 to 0.5. This follows the square-root relationship.
What the result tells you
A larger independent sample generally gives a more precise estimated mean, but gains diminish as the sample grows. Standard error describes uncertainty in the mean rather than the spread of individual observations.
Before you use the result
These two samples are hypothetical and assume the standard deviation stays unchanged. Repeated observations on the same people, clustered sampling and selection bias can invalidate the simple independence assumption. This output is not a confidence interval by itself.
References for this method
Useful for
- Statistics coursework
- Confidence-interval preparation
- Sampling variability estimates
Questions
Why does standard error shrink with sample size?
Because the denominator uses the square root of n, larger samples generally make the estimated mean more precise.
Is standard error the same as standard deviation?
No. Standard deviation describes spread in observations; standard error describes uncertainty in an estimated mean.