> Date: Fri, 15 Dec 2000 19:33:11 -0600 > From: Debraj <[EMAIL PROTECTED]> > Subject: help with modelling > > hi, > > I have a set of data which indicates number of correct responses on a > test (score) for 20 persons. I wanted to know if I can model the same > mathematically based on certain factors, say Score = f(factor1, > foactor2, factor3, factor4), so that I can simulate similar data with > different values of the factors. How should I go about this ? > > thanks > debraj Debra, Your response is the measurement, # correct responses. You don't say what typical scores are, so I _guess_ you have a couple options here. If the answers are more or less near 50% correct, you could do a first and second cut by assuming they were normally distributed. If they run near 20%, or quite a few above 80%, then it is more likely that you are running into a 'physical' boundary - can't go over 100%, true? Then run the first cut analysis assuming normal dist., and if there is _anything_ to talk about, do an arcsin (or Taguchi's omega) transformation to compensate for the boundary effect. As for factors, be sure that you have clearly placed each of your respondents (subjects?) into the appropriate categories or factor values, and also be sure you have a reasonable sample size in each factor group. An orthogonal array in the factors is probably a must for this, as well. Given (most of) these preliminaries, then you could do a multiple regression analysis on the whole thing. Select your model terms based on maximum model fit F ratio. If you don't have your factor levels and test conditions set up as orthogonal, most bets are off. Then drop back and look at individual single factor regressions. HOWEVER: in this instance you may have severe confounding between factors, so you must be extremely cautious about drawing conclusions - use the results as suggestions, or hints, of relationships. Good luck, Jay -- Jay Warner Principal Scientist Warner Consulting, Inc. 4444 North Green Bay Road Racine, WI 53404-1216 USA Ph: (262) 634-9100 FAX: (262) 681-1133 email: [EMAIL PROTECTED] web: http://www.a2q.com The A2Q Method (tm). What do you want to improve today? ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================
