Hi,

If you want to validate the tool between raters, then you may go with what Julia said.

If your quetion means how to analyze Likert scale among your sample, then you may need to calcualte the means for each item and sort the means from either ascending or descending. This will show how much agreement or similarities are there among the items.

and if you want to compare between groups, you do total score and use it as DV, and if you have 2 groups run ttest, and if you have more than 2 groups you run ANOVA.

Good luck 

I think you need Kendall's coefficient of concordance.  This is a test to determine agreement across different people on ranking order.  Borrowing from the stats Wiki:

Assume there are m raters rating k subjects in rank order from 1 to k. Let rij = the rating rater j gives to subject i. For each subject i, let Ri = . let be the mean of the Ri and let R be the squared deviation, i.e. 

   R - SUM (from i=1-k) of {Ri - mean(R)}squared

 Now define Kendall’s W = {12R} / (m^2)(k)(k^2-1)

There is an SPSS help page on it (actually a couple) - if you click on "Help" and enter "Kendall's coefficient of concordance" in the search window you will find it.  It is part of Nonparametric Tests algorithms for related samples tests.  You will need to do some math once SPSS gives you the info, unless they have improved things from version 18 (last time I ran it).

Hi,

If you want to validate the tool between raters, then you may go with what Julia said.

If your quetion means how to analyze Likert scale among your sample, then you may need to calcualte the means for each item and sort the means from either ascending or descending. This will show how much agreement or similarities are there among the items.

and if you want to compare between groups, you do total score and use it as DV, and if you have 2 groups run ttest, and if you have more than 2 groups you run ANOVA.

Good luck