On 11 May 2002 05:47:25 -0700, [EMAIL PROTECTED] (Mobile Survey) wrote: > I had a few questions and would be grateful for any answers: > 1.If I have non-normal distributions for variables measured using 7 > point Semantic differential scales, can I use normalized scores for > the analysis? Are there any objections to doing this.
'Normalizing' a seven-point score is usually going to be a bad idea. It would just about have to be a confession that someone erred in designing the data collection. Unless they already do that (something I have not heard of), the folks who do Semantic differentials will probably object fiercely, for that reason and more. > 2. I have a dichotomous variable in my model. While validating the > measurement model, which software would be the best for me to use? I > understand that AMOS only returns Pearson correlations and may not be > appropriate? I do most of my scale validating with Pearson correlations, which are the ones that people are most apt to understand. But I can't say I understand your assumptions or what you have in mind - perhaps because I never have created Semantic differentials. -- 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/ . =================================================================
