as most know, the concept of power is one of the most quagmiriest (i am sure that is not a word) in all of statistics ... particularly when trying to teach it to students the last chapter in my book has a shot at this ... about 20 pages worth ... and tonight, i (finally ... after some bad file moments) was able to convert to a pdf file (about 80 kb) and shoot it to my server at http://roberts.ed.psu.edu/users/droberts/power.pdf please feel free to have a look ... and use if you think it would be helpful the points i discuss in this chapter are: 1. basic types of decisions ... type 1 and 2 errors, power ... the infamous 2 by 2 box 2. how power is calculated (i use a bit of minitab to do this) 3. factors influencing power ... a. distance between null and alternative b. minimum value of power (and max too and conversely, for beta) c. impact of n on power d. type i error rate and power e. and an attempt to show the steps involved for finding the n needed to build in an approximate value for power ... in a simple investigation any comments would be appreciated ============================================================== dennis roberts, penn state university educational psychology, 8148632401 http://roberts.ed.psu.edu/users/droberts/drober~1.htm ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================
