Looking for significance when taking overlap into account

I am currently working on completing my dissertation. I am looking for a way to test for distinct but overlapping phases. I have run ANOVA's and used Games-Howell Post Hoc's to determine which phases show significant differences in their means. But, since the variable being used to compare means per phase uses a seven point likert scale, and most participants used the upper end of the rating scale, there is a lot of overlap. The particular phase theory being tested predicted overlap in the phases. Is there a method available in SPSS that will look for significance while accounting for the overlap? I hope this makes sense.