Tidy Data and Why We Need It!

แชร์
ฝัง
  • เผยแพร่เมื่อ 16 ต.ค. 2024

ความคิดเห็น • 20

  • @adarshkannan1441
    @adarshkannan1441 ปีที่แล้ว +2

    Great explanation

  •  ปีที่แล้ว

    11:19 It's absolutely worth the time it takes to convert messy data into tidy data. Great video, thanks!

  • @nicolaieg9639
    @nicolaieg9639 7 หลายเดือนก่อน

    Thank you very much - very useful!!

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  7 หลายเดือนก่อน +1

      Greatly appreciate bro! Thanks for watching!

  • @yaoliao3517
    @yaoliao3517 2 ปีที่แล้ว

    There are 5 representive examples of messy data :
    Column headers are values, not variable names.
    Multiple variables are stored in one column.
    Variables are stored in both rows and columns.
    Multiple types of observational units are stored in the same table.
    A single observational unit is stored in multiple tables.

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  2 ปีที่แล้ว

      Yes, the 5 major ones, but I have seen over 300 "creative" tables :) if I used all of the possible mistakes, the video would be one hour :)

    • @yaoliao3517
      @yaoliao3517 2 ปีที่แล้ว

      @@yuzaR-Data-Science Yeah, thanks for your tutorial.

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  2 ปีที่แล้ว

      Welcome!

  • @dmitry503
    @dmitry503 2 ปีที่แล้ว +1

    I think a lot of statistics students would get disappointed after watching this video ))) No comments?? Thank you for the video!

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  2 ปีที่แล้ว +2

      Thanks for the feedback! :)

    • @fishfish20
      @fishfish20 2 ปีที่แล้ว +2

      #Dimtry, I beg to disagree respectfully, it depends on what you're looking for (personal preference). Bear in mind, no knowledge is useless.
      To sum it up, it's just your own hypothesis.
      Cheers 🥂

    • @dmitry503
      @dmitry503 2 ปีที่แล้ว +1

      @@fishfish20 Thanks Jonathan! That was just a reply to the author's joke in the video )) not criticizing its content )

    • @fishfish20
      @fishfish20 2 ปีที่แล้ว +3

      @@dmitry503.. 😂 😂 😂 😂 😂Noted but the first part of the statement takes the joke out of context. 😂 Thanks for the clarification

  • @yaoliao3517
    @yaoliao3517 2 ปีที่แล้ว

    Oh, this is really great!

  • @fishfish20
    @fishfish20 2 ปีที่แล้ว +1

    Greetings sir. I have a question, do we have a package in r that can compute two way anova and give post hoc at the sane time like what ggbetweenstats function does?
    Looking forward to your reply.
    Thank you

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  2 ปีที่แล้ว +1

      unfortunately not a single package, but you can easily do that:
      - first you create a (correct) model (lm, glm, glmer etc.) with two predictors (with or without interaction)
      - then you visualize it with sjPlot packge using plot_model() command
      - finally you use emmeans package and emmeans function from it, to get pairwise comparisons
      I have an article with lots of examples and code here: yuzar-blog.netlify.app/posts/2021-01-01-how-to-visualize-models-their-assumptions-and-post-hocs/
      Hope that helps and thank you for watching the videos - it's the best support! :) cheers

    • @fishfish20
      @fishfish20 2 ปีที่แล้ว

      @@yuzaR-Data-Science Thank you so much Sir for the clarification and continuous efforts.
      Regards
      Jonathan