On 22 Jan 2003 09:10:50 -0800, [EMAIL PROTECTED] (dennis roberts) wrote: > At 10:22 AM 1/22/2003, Stan Brown wrote: > >Eric Lund <[EMAIL PROTECTED]> wrote in sci.stat.edu: > > > > > We really do not want to score all 1600 if we don't have to. > >[snip] > > >So, what I would like to > > >know is, can we get an approximation for the population mean without > > >having any baseline data, and if so, how many students would need to > > >be sampled? > > > i had sent a post about a short handout i did re: margin of error ... that > can be found at > > http://roberts.ed.psu.edu/users/droberts/papers/marer2.pdf > > you might find this helpful >
For my comments on a couple of gross errors in the above, see my posts of a week or so ago, under Margin of error. You could search using groups.google.com and I hope to post something more, tomorrow. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
