This small "n" observational study used a concurrent mixed methods approach to address a void in the literature with regard to the qualitative meaningfulness of the data yielded by absolute magnitude estimation scaling (MES) used to rate subjective stimuli. We investigated whether respondents' scales progressed from less to more and provided for meaningful discriminations, similarities and differences in how they defined their scales, and whether their definitions were congruent with how they used their scales. The respondents completed a think-aloud task while using MES and a follow-up interview regarding how they used MES and defined their scales. The respondents' scales did not yield meaningful data. Implications for survey research and the use of mixed methods in rating scale construction are discussed.