DOE Binary Choice Design with option "none"

[#1]

I am currently trying to build a reduced design for a discrete choice experiment with JMP. In the end I want to have a binary choice design where you can either buy or not buy the product. Is there a way to create a reduced design? Unfortunately, JMP does not give me the option of setting the "number of profiles per set" to 1, as it always automatically corrects itself to 2. Does anyone know how to add a non-purchase option here?

Thank you for your feedback.

[#2]

Hello,

With choice designs you can get the willingness to pay as an output.

In a different approach, have you tried using a logistic regression model?

[#3]

Hi @LatinBiasesOwl7,

Welcome in the Community !

Choice Designs are helpful for determining which factors/features of a product are the most important for a consumer. In order to determine which product attribute(s) should be prioritized for the consumer, you need to provide to the rater/consumer a set with different products (at least 2), where the rater will compare and choose one or several products among the ones proposed in each set. At the end, a utility (desirability) score is provided, which helps to find the best product attributes combination.

You can look at an example that may be close to your expectations here : Example of a Choice Design with Analysis

Several sets of two products are created, and respondents have to select one of the two products.

I'm not sure about your goal in this topic. Choice design might be a good solution, but there might be other options as well, like using factorial design with a categorical response Yes/No.

Since your factors "Price" seems to be a continuous factors and Nutriscore could be seen as an ordinal numerical/categorical factor, a factorial design could be an easy solution, and provide more insight about the influence of continuous factor, like how changing the price by a certain amount may change the probability to pay for this product.

I hope this answer will help you. Please provide more context and objectives if you would like to dive deeper in the topic and have more recommandations,