WTP Discrete Choice Experiment Data Coding for Pretest Parameter Estimates

[Kotryna Klizentyte]

Hello,

I am conducting a willingness to pay model (discrete choice experiment) with four attributes and an opt out option. I am currently pre-testing and received 52 responses-- I want to estimate parameters of the DCE model to ensure its accuracy. I am fairly new to STATA and would like to learn the commands to code the DCE data within STATA, instead of through Excel, as I am anticipating over 1,500 responses for the actual survey.

Here is a description of the attributes:
1. Annual tax amount (payment vehicle): 4 levels
2. Outdoor recreation: 3 levels
3. Environmental health: 2 levels
4. Water quantity: 2 levels
5. Economic impact: 2 levels

I am downloading the responses from Qualtrics. I want to create the following new rows: personid, choiceset, block, alternative, optionA, optionB, option C, choice, and then dummy code each attribute listed above (5).

I understand I will have to 'generate' new variables and then set a default value for each attribute and replace values based on where they are in the block, choice set, and option. Do any resources exist to help initial coding of data of WTP DCE within STATA from Qualtrics?


Thank you!

Kotryna Klizentyte

[Carlo Lazzaro]

Kotryna:
as a first step, you may want to take a look at the following Stata blob articles:
https://blog.stata.com/2016/06/28/fl...-model-part-1/
https://blog.stata.com/2016/06/28/fl...-model-part-2/
and at the community-contributed modules you can access just typing -search WTP-.
Last but not least, an authoritative source is represented by Arne Risa Hole 's posts on this topic.

[Kotryna Klizentyte]

Thank you, Carlo, the first article is very helpful. Do you have another link/title for the second article-- it does not seem to be working.

[Carlo Lazzaro]

Kotryna:
I've just accessed the following link: https://blog.stata.com/2016/07/05/fl...-model-part-2/