Factor Analysis: Illustration with Practical Example in Minitab

Factor Analysis, Factor Analysis in Minitab, Multivariate Analysis, Multivariate Tools

Hello Friends,
In the last video, we saw the 1st multivariate analysis tool in Minitab software, i.e. Principal Components Analysis, with the help of a practical example.
In this video, we will learn the 2nd tool of multivariate analysis in Minitab software, i.e. Factor Analysis, with the help of a practical example for easy understanding and better clarity.
This video consists of the following topics:
• What is Factor Analysis (FA)?
• Data considerations for Factor Analysis (FA)
• Example of Factor Components Analysis
• Conduct Factor Analysis (FA) in Minitab with a practical example, including:
1. Number of factors to extract
2. Method of Extraction: Principal components and Maximum likelihood
3. Type of Rotation:
 None
 Varimax
 Quartimax
 Equimax
 Orthomax with γ:
4. Graphs:
 Scree Plot,
 Score plot for the first 2 factors,
 Loading plot for first 2 factors, and
 Biplot for the first 2 factors
• Detailed interpretation of results from Principal Component Analysis (PCA), including:
 Unrotated factor loadings
 Rotated Factor Loadings and Communalities
 Large loading and Small loading of factors on variables
 Categorization of data
 % of variation explained by each factor
 % of variation explained by all factors together
 Detailed interpretation of Loading plot for the first 2 components, and
• The conclusion from the analysis.
I am sure you will be liked it.

The Important links about LEARN & APPLY:

0:00 Introduction of Factor Analysis
1:57 Data Considerations for Factor Analysis
2:55 Example for Factor Analysis in Minitab
3:44 Conduct a Factor Analysis in Minitab
4:53 Selection of all options while performing Factor Analysis in Minitab
8:19 Interpretation of Factor Analysis in Minitab

And finally, thank you for watching…

#factoranalysis #FactorAnalysisMinitab #Multivariate #MultivariateAnalysis #MultivariateTools #leansixsigma #learnandapply #successfulcareerhub #sixsigma

[kouroumajeanmoussa]

You videos are so much helpful and very easy to understand. Thanks and it up 🙏!

[SaudaraLink]

If you want to use factor loadings in a regression, do you use all the scores from each variable on all loadings, or do you use the two, three, or four values you identified belonged to each variable? Is there an academic source that says the right way to do it?

[maaleem]

Thanks sir, that was indeed one of the simplest explanation.
Although sir I got a doubt.
What should be done after doing the factor analysis.
I mean now that we have got four factors and we identified which factor explains nest which three variables.
So what should we do after that?

[Shabbir]

Very nice presentation with clear explanation of each steps. This is my 1st experience easily explaining the table results of multivariate . Thanks.

[vikasarora]

Thankyou very much! Nice example and your speed is very comfortable.

[nikitamhatre]

I never usually comment. But this helped me a lot. Thanks a TON!

[hymavathi]

Excellent !!! Now only clearly understood this concept for me thanks

[savithamkumar]

simple and clear explanation thank you

[AbeelRao]

hi sir, hope you are good, sir i tried many times to do factor analysis with maximum likelihood method, when i put total 10 variables in variables box and then all i did u said in this vedio but when i click ok to execute with same settings you did, then appears a message of error "*ERROR* MUST HAVE MORE (NONMISSING) OBSERVATIONS THAN VARIABLES. so i tried with 9 out of 10 variables in variables box, then it performs, but results are not matching to book example, so kindly tell me where i am doing wrong

[raghavexplore]

Hello Sir, Thanks you for all your videos. I have a question, I have all my 6 subjects marks from 1st std to 10th std. I want to analysis that which subjects I have performed better. Which Analysis method should I use? Thank you

[abbasuleiman]

I tried Factor analysis using my data and keeps giving me an error message as "* ERROR * Must have more (nonmissing) observations than variables". my question is how do I add more observations when I don't have none in the data set? ( urgent help needed please)

[maybellemarie]

I like your videos! However, I think it would be better if you would give your viewers the copy of the data. It can be in either Google Sheets or what. In this way, we could also solve it on our own while watching the tutorial.

[georgie]

Thank you very much for your very informative video! I have one question regarding transformation of the original data (at the individual level) based on the factor loadings. After performing factor analysis on my data (100 participants, 12 variables), I extract four factors that seem to comprise the 12 variables. Now I wish to return to the original data and transform the individual responses based on the factor loadings so that I can see if individual demographics, such as education level or age, affects the factor structure. Can you tell me how I might go back and transform each individual's data based on these factor loadings/scores? Thank you very much in advance for your help.

[anuraupasantakumarawickram]

Thank you very much, nice explanation with video, and very clear

[maheesha]

Thank you very mch for this clear clarification🤗🤗🤗

[rumeeranisavapandit]

Can't we have any manually done problems?

[anuraupasantakumarawickram]

When I tried Factor analysis, it gave an error message as "* ERROR * Must have more (nonmissing) observations than variables". Anyone could kindly explain what is the reason for having such an error message?

[Shabbir]

Please make videos on NMD, ANOSIM, ANCOA. and GLM
Please explain the output results of diagram also which revealed in PCA/CCA/NMD etc

[pervezalamkhan]

make a video on kansei engineering plz

[dtox]

you keep pronouncing 'condensing' as 'considering' but otherwise good explanation