We defined 4 attributes (type of investment, investment amount, financial performance, ecological performance) with respective levels for a CBC (maybe ACBC).
We want to test 5 different forms of presenting the last attribute, ecological performance.
Therefore, we are planning to build 5 groups of respondents, each seeing a different version of the attribute "ecological performance" - one group will get the attribute's levels as numbers, another group as text, a third group maybe as symbols (can graphics, e.g. jpg, be used as level description?) .
The idea is to analyse whether different forms of presentation lead to different part-worth utilities and thus decision behaviour.
Have you heard of studies in which one attribute "told" the same, but in different ways, across different respondent groups?
What do we have to consider when comparing the five groups' responses?