Selecting a Rotation in a Factor Analysis using SPSS

[@truegrit4752]

I learn so much more from you than I learn in my graduate level classes. Thanks for sharing!

[@wimamaa9273]

This video, like many others on your channel, is a life-saver. Thank you!

[@DrGrande]

Thank you!

[@barrieahmed3233]

I'm lucky to get your video explanations on the rotation in factor analysis... thank you very much, Dr. Todd... this was very much helpful.

[@thomasstarr6433]

Two questions: First where does the absolute of .32 come from regarding correlation values? Second, for the Oblique rotation you examined the Component Correlation table and for the Orthogonal Rotation you examined the Rotated Component Matrix table, so, why the different tables?

[@-yt5258]

We may not see each other but your lessons are very helpful. My sincere respect LOVE from Odisha, India

[@2736492821]

This is gold! I got some introduction on general PCA and it was missing in rotation. This tutorial helped me grasp the concept of rotation and I will be ready to apply it in my analyses, thanks!

[@DrGrande]

You're welcome!

[@javisfernandes9468]

THANK YOU SO MUCH FOR BEING SUCH A BEAUTIFUL HUMAN BEING!!! THIS REALLY HELPED ME!!!

[@khoiandhannah4362]

You are always the best! keep going Dr. Grande!

[@stratpap637]

Very nice explanation! Thank you Dr. Grande

[@Seanv12345]

This is really well done. Thank you.

[@theresiabusagara7909]

Thank you again for very educative clip.

[@theresiabusagara7909]

Please Can you provide reference used for the selection of the rotation method. I refer the .32 as the referred loading for the choice to be made.

[@riksawibawa4630]

Hello Dr Grande. My name is Riksa. I'm PhD Student, and I'm working on PCA. Thank you for your explanation related to selecting rotation in factor analysis. I followed your steps, and I'm using direct oblimin methods to analyze because I believe that every items are correlated, but how about if the results in the rotated component matrix mentioned "only one component was extracted. The solution cannot be rotated." What should I do Dr. Grande? Please kindly your information. Thank you

[@tonytaioftimestreamer2616]

These videos are very helpful thank you!!

[@DrGrande]

You're welcome!

[@dilrubabasser8724]

Thank you very much for this informative video! I had a question. I used an oblimin rotation. In output where can I see the rotated factor loadings? In Pattern Matrix Table or in Structure Matrix Table?

[@svetlanabesklubova6362]

Dr. Todd Grande, thank you for the video, very good explanation! I conducted the varimax rotation. In some cases, I got complex variables with loading 0.466 and 0.455 (example) to different groups. What I should do in this case? Should I leave this item in a group with loading 0.466?

[@leonardonevarez2522]

where did you get the basis of .32 when identifying if it's oblique or orthogonal?????

[@SudarshanBaurai]

Thank you Prof. 🙏

[@aymanzein7]

In depth video, thanks 1- If the complex variables persist ( whenever I delete one , another one pops up ) can I keep them ? 2- Can I use any Extraction method for Oblimin ?

[@mahasarwar5513]

Hey. Did you get your answer from anywhere?

[@aymanzein7]

@@mahasarwar5513 Yes , I recommend : Hair, Multivariate Data Analysis 7th edition Ch.4 Exploratory Factor Analysis

[@iPsychlops]

Does it matter that this is PCA? I'm doing PAF; would I look at the factor correlation matrix? Is the value |.32| to determine if I'm running oblique vs. orthogonal?

[@MartinaHertl]

Great video, very useful. Thank you.

[@DrGrande]

You're welcome, thanks for watching -

[@MsStinaB]

Do you have any good references on which rotation method to use and where does the absolute of .32 come from regarding correlation values? Need this for a publication.

[@sebi1988777]

Tabachnick and Fiddell (2007, p. 646) argue that “Perhaps the best way to decide between orthogonal and oblique rotation is to request oblique rotation [e.g., direct oblimin or promax from SPSS] with the desired number of factors [see Brown, 2009b] and look at the correlations among factors…if factor correlations are not driven by the data, the solution remains nearly orthogonal. Look at the factor correlation matrix for correlations around .32 and above. If correlations exceed .32, then there is 10% (or more) overlap in variance among factors, enough variance to warrant oblique rotation unless there are compelling reasons for orthogonal rotation.”

[@gideonvictor1490]

Thank you!

[@halilemrekocalar6537]

thank you for explanation which is so useful for me!

[@DrGrande]

You're welcome, thanks for watching -

[@DrHanjabamBarunSharma]

references for significant loading, zero loading, complex items etc..plz

[@newgeneration8390]

you hit the bull's eye... awesome

[@mohammadmasroorzafar9728]

How we decide from 0.32? What is the basis? Can you share the name ot book Or research paper Please

[@youthf7c343]

What is rotation? I want to visualize that concept. What components on the axis are rotated? And under what conditions do we choose varimax or quartimax, equimax . thanks :)

[@abyarthagoswami7663]

How are factors formed in factor analysis

[@khalilsaleh2984]

rotate the items to form the factor select type of rotation depend on the correlation status if it is above .32 or less than it ,, insignificant differences among subcategory rotation

[@2736492821]

www.theanalysisfactor.com/rotations-factor-analysis/ I reckon this help in visualizing the concept of rotation, cheers

[@naveens806]

How to write hypothesis for above problem

[@neuroscience5994]

Why correlations of specifically 0.32 for direct oblimin to be useful? Is there a reference for this?

[@cintiacampos3454]

Did Todd Grande answer your question? I also could not understand where is from de the magic number 0.32

[@sebi1988777]

Tabachnick and Fiddell (2007, p. 646) argue that “Perhaps the best way to decide between orthogonal and oblique rotation is to request oblique rotation [e.g., direct oblimin or promax from SPSS] with the desired number of factors [see Brown, 2009b] and look at the correlations among factors…if factor correlations are not driven by the data, the solution remains nearly orthogonal. Look at the factor correlation matrix for correlations around .32 and above. If correlations exceed .32, then there is 10% (or more) overlap in variance among factors, enough variance to warrant oblique rotation unless there are compelling reasons for orthogonal rotation.”

[@patrickdi910]

lol the questions here are dumb as hell. thx for the video it was quite helpful

[@biniambekele7267]

your explanations are fantastic but the videos are not good. they all lack visibility!