On 3 Jan 2004 21:55:22 GMT, [EMAIL PROTECTED] (Radford Neal) wrote: > In article <[EMAIL PROTECTED]>, > m v <[EMAIL PROTECTED]> wrote: > > >Her project was to find if short term memory was better for young > >adults than for younger or older age groups. She devised a memory > >test and tested 10 people in each of five different age groups. > > > >What would be an appropriate statistical test for this hypothesis and > >data (considering it is for a middle school project)? > > Her hypothesis is apparently that the memory is best at some > intermediate age, and worse for both smaller and larger ages. It's [ snip, reasonable stuff about how to test it.]
But, please, let us go back and write that as a sensible hypothesis. Keep in mind that this is a survey, using a "convenience sample": Even after a survey that is large and representative, you are in a shaky position for "testing the hypothesis," if you write the hypothesis as a grandiose fact about nature. I don't think that your daughter is attempting a random sample from all human beings; and she doesn't have the life-experience to say much about other cultures at all (I would bet); so, don't write the problem that way. As an example. Here is a more reasonable, limited hypothesis: "In this school/ shopping mall/ county, the several racial-ethnic groups will not differ in IQ, when measured by < ... > ." In this style, you can draw a conclusion about a small hypothesis -- without offending all the world about how wrong you are apt to be, because you have not taken *all* the possible objections into account. For your example: You test the age groups, with the hope that in your social-class/ nation/ region, this relation might be true. Assume you show that there is an apparent curve of that shape; you have achieved a little bit. At this point, you start your argument about whether the finding is apt to generalize, or to whom it is apt to generalize. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html "Taxes are the price we pay for civilization." . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
