Discrete Choice Experiment: None of the proposed options

Поделиться
HTML-код
  • Опубликовано: 11 дек 2024

Комментарии • 8

  • @adamacoulibaly812
    @adamacoulibaly812 2 года назад

    Greetings dear Prof this is very interesting video you have made. I struggle a lot to find such video elicitating how to administer DCE questions with electronic data collection tools. I am PhD student in applied agricultural economics working on DCE and I really want to collect my data with mobile phone or tablets but I have no idea about how to proceed because most the prior studies used paper-and-pen data collect methods and I am not familiar with this. On the other hand I have time constraint going through the paper-based data collect. So I am familiar with the ODK or Kobotoolbox survey design platform. Then I wanted to know 1) if there is speficific platform (e.g. XM as seen from your video) for designing DCE questions, 2) or if the ODK or kobo can also perform the DCE questions. I would like to kindly request your mail address for further questions. Thank you.

    • @eugenleo
      @eugenleo  2 года назад

      Thank you for the comment. 1: yes, there are several tools for DCE development. There is even a manual way to do it. Maybe once I'll make a video for this. Otherwise, simply google for the tools - there are plenty of them. I use Qualtrics as we have an institutional account. And there is a plugin that makes DCE easy to implement. Yet, if you develop the items manually, you may use ANY platform. Once I'll explain how. 2: I've never worked with the platform. But go for it, if it works. My email: yevgen.bogodistov(g)mci.edu

  • @lelavanosselaer8884
    @lelavanosselaer8884 Год назад

    Dear Prof Bogodistov, I use a status quo option in my experiment. What is the effect on the restructuring of the data in SPSS?

    • @eugenleo
      @eugenleo  Год назад

      Hello. Could you explain, what you mean by "status quo"? Randomized attributes?

    • @lelavanosselaer8884
      @lelavanosselaer8884 Год назад

      @@eugenleo What you called 'none of the proposed options' is called company car in my study, because it was the status quo (if they didn't change their choice in comparison with the original situation).

    • @eugenleo
      @eugenleo  Год назад

      @@lelavanosselaer8884 you will need to calculate the final_decision variable differently. All other restructuring remains the same. I think...

    • @eugenleo
      @eugenleo  Год назад

      @@lelavanosselaer8884 it is hard to say without being at my PC and seeing the file. Let's think about it in the following way. If option 1 is preferred it gets a 1 (decision = 1 and index2 = 1) and option 2 gets a 0. If option 2 is preferred, option 1 gets a zero and option 2 gets a one (decision = 2 and Index2 = 2). If "none of the options is selected (decision = 3), then both options get a zero (bone of them is preferred). Then you don't need a 3 in the final_decision. The variable is then binary and you run a logistic regression. I hope it helps. Have a nice evening!

    • @eugenleo
      @eugenleo  Год назад +1

      @@lelavanosselaer8884 you can do, or just arrange a meeting with me. I should have time this week. calendly.com/yevgen-bogodistov Yet, I won't be able to look at it before Monday.