**LANCER STATS PAGE **

Point Estimates

and

Confidence Intervals

Back to Lancer Stats Home Page

- Read the Hyperstats overview on Point Estimates . Read carefully the qualities of a good statistical estimator.
- This is an overview of the ideas of confidence intervals.
- Hyperstats section on One Proportion Confidence intervals
- This section of Hyperstat covers confidence intervals when the standard deviation is known.... Ok, that almost never happens, but the idea will help you see the difference between Z and t-tests.
- And of course if you don't know the standard deviation, and you almost never do, then we need to read the section on confidence intervals with sigma estimated by s
_{x} - Hyperstats also has a nice summary of the general formula for finding cofidence intervals.
- Here are three java applets to remind us that the 95% confidence interval means we will only capture the true mean 95% of the time. (Of course we never know which time we fall in that other 5%.) Do each of these as they each address the idea in slightly different ways

Applet one

Applet two

Applet three - Here are some questions on Confidence Intervals (with the answers, or at least SOME have answers) to test you understanding.
- You can sign in at the Yates/Moore/Starns web site and take Practice tests. (Leave instructor blank when asked and "continue"). For this chapter you want their chapter 10 quiz. http://bcs.whfreeman.com/yates2e/default.asp
- This link describes an experiment on the effects of calcium for treating blood pressure in African American males. The data is at the bottom. Use the data for "Begin" for both groups to construct a 95% confidence interval on the true mean of an untreated population. Use the "END" data from and on the true mean of a population if calcium treatment was used. Be sure to verify all necessary conditions. This data is also available at Cacium data on StatCrunch

Here is a video on a step by step approach to writing an acceptable Confidence interval statement.

and this one has a little history and information on uses of confidence intervals.

Unsure about finding critical values, here is a brief cover

A glossary of Statistical Terms from SticiGui

Hyper-Stats Glossary