Geo Analysis - Demo 5

 



Demonstration -- Confidence Intervals



The variation of the unknown estimation errors is exactly the same as the variation of the sample means. The standard error of the mean (sem) is the standard deviation of the distribution of the sampling errors -- the discrepancies between the statistic and the parameter.

In a normal distribution 95% of the observations fall within +/- 1.96 standard deviations from the mean. The following relationship holds:

We are 95% certain that the true mean is less than 1.96 standard errors above the observed sample mean but greater than 1.96 standard errors below the sampel mean. There is a 5% chance that it will fall outside of this interval. We randomly sample 100 ball bearings and find a mean weight of 150 grams. The standard deviation is 3.0 grams. The sem is 0.3 grams. The 95% confidence interval is approximately: 149.4 < mu < 150.6.

Java applets are script that is downloaded to your machine. Using a browser that recognizes these applets (such as Netscape 3.0) try Globally Accessible Statistical Procedures

As alpha decreases, what happens to the width of the confidence interval -- in general? WIth an alpha of 0.05, how many samples fall outside of the confidence interval for 1000 total samples? What does theory say about this? With an alpha of 0.025, how many samples fall outside of the confidence interval for 1000 total samples? What does theory say about this?

Central Limit Applet will let you perform the analysis of rolling two die and keeping track of the sum of the spots showing.

Return to the Geo Analysis Home Page