R demo | Chi-Square Test | How to conduct, visualize and interpret | + pairwise post-hoc tests

Поделиться
HTML-код
  • Опубликовано: 3 янв 2025

Комментарии • 36

  • @hikeaway1596
    @hikeaway1596 7 месяцев назад

    I can't stop watching your videos ;) please produce more of them, It's really fun to learn from your content.

  • @jammaningas8850
    @jammaningas8850 3 месяца назад

    Very easy to follow! Nice!

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  3 месяца назад

      Glad you think so! 🙏 you might like other videos too 😉

  • @ManuelAlejandroCastroGarcia
    @ManuelAlejandroCastroGarcia 6 месяцев назад

    Thanks for sharing your knowledge. How can I add the Yate´s correction into the plot?

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  6 месяцев назад

      Unfortunately, either not possible, or I don't know why and how. But thanks for good feedback!

  • @JohnLee-pi5qx
    @JohnLee-pi5qx 2 года назад

    Thanks for the great video! Any tips for changing colors of the bars?

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  2 года назад +2

      Thanks, mate, sure:
      library(ggplot2)
      ggbarstats(mtcars, am, cyl, results.subtitle = FALSE) +
      scale_fill_manual(values = c("#E7298A", "#66A61E"))

  • @osmanhabeeb9892
    @osmanhabeeb9892 2 года назад

    I was looking for the means separation of chi-square test in R. THANKS ALOT.

  • @ouss991
    @ouss991 3 года назад +1

    Thanks Yuri, great tutorial as usual. I have a question : can I actually use these plots in a paper or thesis, or do I have to create a conventional barplot without the stats?

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  3 года назад +2

      Hey Ouss, you can absolutely use them for a paper. They make your text cleaner, because you have all the stats on the plot, and not in the brackets, which disterbs the flow. You can do both, 1) save the plot with ggsave, where you can adjust everything, like size etc. and 2) you can create you own plot and use package "statsexpressions" from the same creator as "ggstatsplot" to add statistics to your own custom plot. Cheers.

    • @ouss991
      @ouss991 3 года назад

      Great. Thanks for the quick response. Looking forward to new videos 🙏

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  3 года назад

      you are very welcome :) more videos are on the way ;) thanks for watching!!!

  • @Ange-y1k
    @Ange-y1k 3 месяца назад

    First of all, thank you. The video may be old, but I'm only watching it now. How do you go about post-hoc testing when comparing two variables with more than two categories? (I am having an error saying: 'x' must have 2 columns)

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  3 месяца назад

      hi, your error message is most likely due to having only 1 category, not >2. sure, you can do post-hocs easily, here is how, but I think I talked about it in the video:
      install.packages("rstatix")
      library(rstatix)
      contingency_table

    • @Ange-y1k
      @Ange-y1k 3 месяца назад

      @@yuzaR-Data-Science Thank you for your answer. I used the same code, but replaced it with variables, each with 3 categories, it didn't work and gave me that code error.

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  3 месяца назад

      again, >2 categories is not the problem. here is the proof for two categorical variables with 3 categories each: ggstatsplot::ggbarstats(mtcars, cyl, gear). Something is wrong in your data, may be in your table, or some packages are not installed.

  • @andreapaolacotesperdomo2293
    @andreapaolacotesperdomo2293 Год назад

    Hi, great tutorial, many thanks!
    Any advise for comparison of one continuos vs one categorical variable?

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  Год назад +2

      thanks! sure, I have similar short video of t-test, Wilcoxon, ANOVA, Kruskal-Wallis, Repeated Measures ANOVA, Friedman .... they all deas with one continuous and one categorical variables, check them out and find what fits your need best. Cheers and thank you for watching!

    • @andreapaolacotesperdomo2293
      @andreapaolacotesperdomo2293 Год назад

      @@yuzaR-Data-Science thank you very much 😁

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  Год назад

      @@andreapaolacotesperdomo2293 you are welcome!

  • @neleluisa9210
    @neleluisa9210 Год назад

    this was very helpful, thank you! Is there a way to calculate Cramer's V for the pairwise post hoc tests?

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  Год назад +1

      Not yet! But it's a great question. I have to google it first. There is another cool package, I did not do a review yet - effectsize. Check out whether this one does the job. Thanks you for watching!

  • @dewidinevandercolff3999
    @dewidinevandercolff3999 2 года назад

    Thanks, this was really easy to follow!

  • @evansnyanney7431
    @evansnyanney7431 2 года назад

    I tried using it but I am facing lil bit difficulties- that ! Can’t convert to a symbol

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  2 года назад

      convert it to a factor: data %>% factor(bla = factor(bla))

  • @so4ragb
    @so4ragb 3 года назад

    thanks, great tut

  • @ecovisionando
    @ecovisionando 3 года назад

    Thanks 😀

  • @siriyakcr
    @siriyakcr 3 года назад

    🥰🥰 super yuri

  • @ManuelAlejandroCastroGarcia
    @ManuelAlejandroCastroGarcia 6 месяцев назад

    Thanks for sharing your knowledge. How can I add the Yate´s correction into the plot?

    • @yuzaR-Data-Science
      @yuzaR-Data-Science  6 месяцев назад +1

      Unfortunately, either not possible, or I don't know why and how. But thanks for good feedback!