Overview
A 99% confidence interval gives a range for the population mean that you can be very confident contains the true value. The price is a wider range than a 95% interval.
Enter the sample mean, standard deviation and size. Keep the confidence code at 3 for 99%.
Worked examples
| Inputs | Result |
|---|---|
| Sample Mean54.2Sample Standard Deviation12.4Sample Size (n)120Confidence level99% | 99% CI: [51.28, 57.12]Margin of error: 2.92 |
Show 2 more examples
| Sample Mean100Sample Standard Deviation15Sample Size (n)50Confidence level99% | 99% CI: [94.54, 105.46]Margin of error: 5.46 |
| Sample Mean3.2Sample Standard Deviation0.4Sample Size (n)200Confidence level99% | 99% CI: [3.13, 3.27]Margin of error: 0.07 |
How it works
The standard error is the standard deviation divided by √n. For 99%, it's multiplied by 2.576 to get the margin of error, which is added to and subtracted from the mean. The calculation uses z values, so small samples under about 30 would need a slightly wider t-based interval.
The formula
| 2.576 | z value for 99% confidence |
Practical tips
- A mean of 54.2 with s = 12.4 and n = 120 gives 51.28 to 57.12.
- A 99% interval is about 31% wider than a 95% one from the same data.
- Use 99% when a wrong conclusion would be costly, such as safety testing.
Frequently asked questions
What does 99% confidence mean?
If you repeated the sampling many times, about 99% of intervals built this way would contain the true mean.
Why not always use 99%?
The wider range is less precise, and you need much bigger samples to keep it narrow.
How do I narrow it?
Increase the sample size. Quadrupling n halves the width.