Principal components analysis using SPSS (Oct 2019)

Поделиться
HTML-код
  • Опубликовано: 26 дек 2024

Комментарии • 18

  • @liabogitini3950
    @liabogitini3950 4 года назад

    Thank you so much Mike. I have been looking for videos that explains how to interpret my pca outputs in simple terms but with no luck until I came across your video. So informative and easy to understand. Definitely recommending to my friends

  • @saeedahmad6827
    @saeedahmad6827 4 года назад +1

    Thank you very much Respected Sir for your brief description.

    • @mikecrowson2462
      @mikecrowson2462  4 года назад

      You are very welcome Saeed! Thanks for visiting!

  • @nehadave7773
    @nehadave7773 Год назад

    Thank you so much. It is very useful and you explained it really well.

  • @negusuworku1871
    @negusuworku1871 2 года назад

    Thank you so much Mike. It is very useful. I learnt a lot.

  • @thulfiqaral-graiti7131
    @thulfiqaral-graiti7131 2 года назад

    Thank you, it was very nice and easy to follow!

  • @cocosimba9178
    @cocosimba9178 4 года назад +1

    Hi Mike! Very useful video here. Yours is definitely one of the most resourceful and clear PCA videos I have ever come across (with reference too - which is awesome!) Highly recommend this video!!
    Just one quick question: when I was doing the rotation component matrix, some of the items appeared do not contribute to any component at all (e.g. the item instructor being sensitive to students does not have any number to either component 1 or 2 when I set the suppression value to 0.4. Does it mean that I have done something wrong? or I can just take out that item when I consider naming my components?)
    Thank you so much! :)

    • @mikecrowson2462
      @mikecrowson2462  4 года назад +1

      Here there. When running a PCA (or EFA for that matter), you often will have items that do not meet loading criteria. This does not mean you did anything wrong. Those items that do not do a good job of defining a given component are basically not used in the naming of the component. If you are forming scales based on the pattern of loadings, then you usually just include those items that met the loading criteria (and that were used in the naming of the component). Cheers!

  • @lucastheworldis1336
    @lucastheworldis1336 3 года назад

    Fantastic and simple explanation, thanks.

  • @aayushmak.c.7586
    @aayushmak.c.7586 4 года назад

    Hey Mike, thank you so much for uploading this, it was very helpful! Very clearly explained :)

  • @yulinliu850
    @yulinliu850 5 лет назад +2

    Thanks for teaching!

    • @mikecrowson2462
      @mikecrowson2462  5 лет назад +1

      You're welcome, Yulin. thank you for visiting!

  • @ghasemyadegarfar9736
    @ghasemyadegarfar9736 4 года назад

    Thank you Mike. It was very useful.

  • @garvitbhomia
    @garvitbhomia 4 года назад

    In output, I am not getting my rotation matrix coz it says only factor was extracted
    The problem fixes when I change eigen value from 1 to 0.5 in extraction.
    Is this legit way to do....?

  • @irfanullah7561
    @irfanullah7561 4 года назад

    Mike Crowson!
    Can you give the short description about this dataset?
    What was the purpose of collecting this dataset?
    I'll be very thankful.

    • @mikecrowson2462
      @mikecrowson2462  4 года назад

      Hi there. Actually, I didn't collect the data. I downloaded it at:
      stats.idre.ucla.edu/spss/output/principal-components-analysis/
      The data appears to be instructor ratings, and there's a better description of the individual items at: stats.idre.ucla.edu/sas/output/principal-components-analysis/
      hope this helps!

  • @mikecrowson2462
    @mikecrowson2462  5 лет назад

    Hi everyone, be sure to also check out this video on PCA using the open-access (freely downloadable at www.jamovi.org/download.html ) program, jamovi: ruclips.net/video/fLeXEG2XUPI/видео.html

  • @nourbhr2746
    @nourbhr2746 3 года назад

    Thank you so much for this useful explanation. Could I have your email please ? . Cheers