iSixSigma

help required

Six Sigma – iSixSigma › Forums › Old Forums › General › help required

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • March 9, 2004 at 9:15 am #34844

    indresh
    ★ 10 Years ★
    Participant

    dear all,
    need some help to analyse the data
    number of responses on 27 parameters taken from a sample of 30 people with rating scale 1-5 from completely dissatisfied to completely satisfied
    what all tools can be used besides pareto to get the analysis of things that need attention and have been rated low
    rgds,
    indresh

    0
    March 9, 2004 at 1:26 pm #96617

    Bob J
    ★ 10 Years ★
    Participant

    Hi indresh!
    You might consider using a chi squared contingency table or a 2 way anova…  Either will show you whether there is a significant difference between your 27 subgroups.  The contingency table may be limiting if there are cell counts less than 5 or so but the anova should be fine as long as your underlying subgroup data is reasonably normal.
    Another tool that might be helpful is the interval plot.  It will display the mean of each of your 27 subgroups with an associated confidence interval.  This makes it very easy to determine if one or more of the means is significantly different from the rest….
    You may want to also consider looking for correlation between your subgroups….
    Hope this helps…
    Bob J

    0
  • Author
    Posts
Viewing 2 posts - 1 through 2 (of 2 total)

The forum ‘General’ is closed to new topics and replies.