In doing a factorial survey, the first 70 or so respondents used a very narrow range of the 1 to 7 Likert scale, mostly 6's and 7's, seldom going below 5. This pilot study is using students who agree to fill out a 15 minute survey. What would you suggest to discern a very narrow range, where most respondents might use 6 and 7 and a lesser number use 5 and 6? Are there tests for such within respondents? Normalize the data somehow so they are more in tune with each other? Suggestions? Is this a common problem in statistics with surveys? Adam Sundor ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================