Question 215:

Asked on June 3, 2008

Tags: Confidence Intervals , Standard Normal Curve , Standard Deviation , Normally Distributed

1

Answer:

No answer provided yet.

One technique to estimate the population standard deviation when no data is available or it is unknown, is to take the likely range of values and divide by 6. In this case that's 45-30 = 15/6 = $2500. You divide by 6 since the properties of the normal curve show that 99% of the values are within +/- 3 standard deviations of the mean. If your data is not-normally distributed though, this estimate can be off by quite a bit. Salary data is often positively skewed due to very high salaries. In this case it looks like it's a reasonable approach.

See also question 197.