Excel for Scientists & Engineers

Often times, we need to estimate the characteristics of a population on the basis of information about a sample. For instance, we use the sample's mean to estimate the population's mean.
The amount of confidence we can put in an estimate varies. To deal with this problem,
we set a confidence level (usually 95%) and
we determine a confidence limit (around the mean).
Look at the example shown in Table - 11. Based on a sample's mean of 44 (and a standard error of 6), we have 95% confidence that the population's mean is between 32 and 56. Consequently, we take a 5% risk ( ?) of being wrong!
| -3 | -2 | -1 | 0 | 1 | 2 | 3 | Z-value | |||
| 0% | 2% | 16% | 50% | 84% | 98% | 100% | probability | |||
| 26 | 32 | 38 | 44 | 50 | 56 | 62 | mean | stdev | ||
| 44 | 6 | |||||||||
| -DIST functions find the probabilities (for a certain distance) | ||||||||||
| -INV functions find the distance (for a certain probability) |
If we were to plot "cumulative" means for one up to 20 samples of 10...