For my master thesis I want to conduct a stated choice experiment. After reading about it in one of the blogs*, I have a question about the possibilities I would have when using SurveyGizmo to design the experiment.
Is it possible to use a previously obtained answer to determine attribute levels? For example, I want to base the attribute levels for travel costs in the stated choice experiment on the respondent’s current travel costs. To do that, I would ask about their current travel costs first, and then use this answer to determine the attribute levels (travel costs +/-20%, travel costs +/-10%, travel costs +0%).
Is it possible to do this with SuveyGizmo, and if so, which plan would you recommend? If it is not possible, do you have any idea how I can do it?
It depends on what you mean by the term “attribute levels”. If you are asking if the answer options that appear in a radio button or checkbox type question can be determined by the answer to a preceding question, then the answer is yes (https://help.surveygizmo.com/help/answer-option-logic). This also depends on the question type that you use to obtain your “current travel costs” response.
This type of question would require logic, but I am not sure the differences between what SurveyGizmo terms basic logic (Basic Plan) and advanced logic (Professional Plan). You might consider contacting SurveyGizmo Support for details on this (https://community.surveygizmo.com/email-a-support-hero/).
I am almost certain you could do this with a combination of hidden values, piping, and custom scripting. You would need a Professional plan to get access to the custom scripting ( https://script.surveygizmo.com/help ).
The basic plan would be to:
Create hidden values that will hold your inflated and deflated travel costs.
Use custom scripting to: get the value from a previous answer, do the math and then use sgapiSetValue to put the results in your hidden fields.
Then use piping to copy the values from the hidden fields into a question’s response options.
Since you mentioned conjoint, I recommend reading up on the Conjoint (Choice Based) question to see if this is related to what you are after:
Hope this helps!