Exploratory Factor Analysis
ฝัง
- เผยแพร่เมื่อ 25 ต.ค. 2021
- Exploratory factor analysis (EFA) is a method that aims to uncover structures in large variable sets. If you have a data set with many variables, it is possible that some of them are interrelated, i.e. correlate with each other. These correlations are the basis of factor analysis.
The aim of the factor analysis is to divide the variables into groups. The aim is to separate those variables that correlate highly from those that correlate less strongly.
In Statistics Exploratory Factor Analysis is also called Principal Component Analysis (PCA)
More Information about Exploratory Factor Analysis
datatab.net/tutorial/explorat...
Here you can find the Factor Analysis Calculator
datatab.net/statistics-calcul...
And the Data:
datatab.net/assets/examples/P..."
If you like, please find our e-Book here: datatab.net/statistics-book 😎
I learn all this in 15 minutes what has taken 4 years and $10,000. Thanks for this. Simple,clear with no distracting loud music.
Many thanks : ) Regards Hannah
Varimax is an orthogonal rotation, which assumes that factors are unrelated to each other. In your example, you've used personality traits, which are psychological constructs and would be expected to be related to each other (in psych, constructs are rarely if ever completely unrelated). An oblique rotation, which assumes factors are related, would normally be the most sensible choice in psychology research. The appropriateness of orthogonal vs oblique rotation for their variables/discipline is something people should be aware of and consider in EFA.
Thanks !!
It's amazing how deep a rabbit hole every step of my research is taking me into, but again I'm the idiot who decided to make my own instrument for my first research project :')
I was struggling with factor analysis and I had zero background in this. You video just saved my life, explained so nicely and such a clear visual representation within such short amount of time! Can't thank you more. Keep posting more please.
Thank you so much. You provided a much clearer explanation that my lecturer or any text books!
I have spent four years in learning these kind of courses and i found you. You are a life saver. Pls keep it up
Happy to help and thanks for the nice Feedback!!! Regards hannah
This is so simple, straightforward, and helpful. Thanks a lot!
one of the best videos on explanation of FACTOR ANALYSIS. THANKS A LOT. BY HEART 🧡🧡
Great video! Great Explanations. This made me try the one month subscription to conduct an exploratory factor analysis. The software is easy to use and accessible immediately after purchase.
I can highly recommend it! Facilitated my research A LOT!
Many, many thanks for the nice Feedback!!! 😊Regards, Hannah
Easy and clear. Above my expection,very helpful. Thank you.
I understood the concept communalities after listening your explanation. Thank you so much.
One of the best lectures on EFA I have sat through
Thanks!
Amazing! This is very well and simply explained. Thank you!
Thank you, this has been so helpful to me. I'm a 1st year PhD student in social sciences working on my first literature review.
Glad it was helpful! Regards, Hannah
Loved your explanation! So easy to understand!
Thank you so much for the detailed explanation. This helped a lot!
You've done what my teacher couldn't, thank you
Glad I could help!
Thank you very much for your explanation. I appreciate your work and effort.
Omg you're so good. I literally cannot thank you enough!!! Thank you so very much.
Why was I able to understand all of this, this was so understandable, thanks!
Many thanks!
Thank you so much! love the simple and logical explanation!
You're very welcome!
Lack words to appreciate your exemplary lecture!
It's my pleasure! Thanks Hannah
I really appreciate the videos that you are covering Test Theorie und Test Konstruktion lecture.
Many thanks!!! Regards Hannah
Very easy to understand video. One of the best on TH-cam ❤
Glad you think so!
Really great. I have just started following all videos after watching this.
Great 👍Many thanks!
very nice presentation. well explained.
Thanks a trillion. Brilliantly explained.
Thanks very much for this video! Very helpful!
Glad it was helpful!
Well done!!! energetic and engaging explanation! bravo
Glad you liked it! Regards, Hannah
you are an amazing instructor, thanks a lot
I appreciate that!
A very good presentation
Woooow excellent video. Thanks
Thank you. Do we calculate the Eigean value first to assign the numbers of factor?
Veryyyyyyyyyyyyyyyyyyyyyyy helpful. that helped me so much! Thank you very much and please keep going.
Many thanks! Regards Hannah : )
thanks for the video. this is mostly PCA , rather than factor analysis. they are different.
This Video was very helpful .
So, please correct me if I'm wrong (I'm revisiting the topic). When you identify the each factor, you can create a factor score for each individual by using the relevant component values in each factor and the actual individual observations, right?
I finally understood many concepts.
Greate!
this is fantastic! thank you so much!
Glad it was helpful!
You are my savior. thank you so much
Thank you a lot!! Need to review all your videos to pass the exam😂
Hi, thank you very much! It reminded me a lot about multicolinearity in linear regression, are they related somehow?
Thank you so much for these videos
You're so welcome!
good. extremely useful for beginners
Thanks !!!
Excellent mam.Thank you so much for clear explanation
Thanks!
Great explanations (y)
Love this! thank you!
You're so welcome!
Hello Hannah,
can the underlying factors be seen as independent variables and the observeable phenomena as dependent variables?
Best regards!
Hello. thanks for your video. i have a qoestion. Is it necessary to do exploratory factor analysis to perform structural equation analysis of a theoretically determined structure?
In PCA, proportion of eigenvalues > 80% is also considered as third method.
Many thanks for the hint! Regards Hannah
Super helpful 😊
Great teaching!
Glad it was helpful!
Excellent. Kindly upload videos on Research Designs
Many Thanks!!! We will put it on our To-Do List!!! Regards, Hannah & Mathias
very helpful. thank you
Glad it was helpful!
thank you so much ... if the kink is formed below eigen value 1 then what to do?
This is great!! Thank you so much!
I was just wondering why you used 3 as the amount of factors at the beginning? How did you know to use that instead of 2,4,5 etc.
This is explained in the video. Because the model output shows three factors have Eigenvalues greater than 1.
You can use the eigenvalue criterion or the elbow method. I think they are also explained in the video! Regards Hannah
Many thanks Quant Quill!
@@datatab I think that the question was: why do you set the factor levels at 3 as the first step in the PCA. In other words, why would you set the factor levels at 3 prior to creating the eigenvectors etc.? I am also curious - because in the Explained Total Variance table you have 6 factors ("components") listed rather than the 3 you selected.
Thanks. This is awesome
Thanks for the nice Feedback! Hannah & Mathias
Thank you very much for your clear & precise explanation. Is there any difference(s) on conducting PCA, EFA and CFA?
Yes there is a small difference!
@@datatab Thank you for making this easy-to-use software and the training material. It helps a lot.
One Suggestion: please add a button of "Select ALL" for lazy people like me. I am doing PCA & Reliability of 40 variables; but have to click them one by one after switching to one another analytical tool. thanks again.
Thanks for the wonderful lecture.
I have a question in EFA:
The assumption that the measurement errors are not correlated between them?
This assumption isn't valid in reality ?
Hmm, I can't answer that for you unfortunately! It is true that in reality the requirements are often not taken quite so strictly!
I wish my stats professor in undergrad was even a tenth as good as you.
Glad you liked it!
Thanks ma'am it really helped me a lot
Most welcome 😊
Perfect explanation
Glad it was helpful!
She's a life saver.
Thanks!
I have a query, the variance of the variables, for each factor, in the component matrix, is same, as the variance of the variables for each factor in the rotation matrix, I don't see any difference between the two methods, what is the difference between the two methods?
Beautiful talk
Thank you
Excellent....
Many thanks : ) Regards Hannah
Sorry I have a question here....
If I want to perform bivariate correlations of the factors after EFA, how do I transform data from the multiple variables into data for a single factor for the correlation analysis with other factors?
Thanks for your time and attention.
Sorry for the late reply! Unfortunately, I can not answer you in a hurry, I would have to read up first! Regards Hannah
Really great
Thanks!
Thank you.
Thanks : )
thank you
You're welcome
quite useful
Glad you think so!
Thanks
which one is 1 and which one is 5 is not mentioned. some times scalling starts from 5,4,3 ..1 or some times , in ascending order. which order has been followed here? i think 5,4,3,2,1 . Am I correct?
hi, and thanks for the video . i wanted to ask what variance is. plz use an example . Thank u
Thanks for your question, we have a video on Variance: th-cam.com/video/jx8a_jdlxAQ/w-d-xo.html
Hi can you also create a video on confirmatory factor analysis please?
You just saved a life.
Many thanks for the feedback : )
Can you Do a video for principal axes and maximum likelihood?
thank you for your feedback! I write it down but can not promise, there are so many topics : )
Nice....Thank you so much..How many respondents are needed to do this EFA? Is there any literature to support this?
Im planning to do pilot test on factors that affect staff retention.But for sure, its a lot.So, I would like to lessen the factors.Thank you in advance
Many thanks. Unfortunately, I can not answer that right away! I would also have to do a literature search. Sorry!!!
pls if you are online please help me by how to analysis logistic regression with more than 13 independent variable and how to check and write interpreted report
What about factor loading, how to calculate that, am I missing something ?
Is it possible to conduct Strucutural Equation Modeling in datatab?
No sorry, at the moment this is not possible!
And you follow the steps to buy in the end DATAtab... Good Marketing
what you talked about was principal component analysis instead of factor analysis.
what is identity matrix? is it component matrix?
I appreciate the simplification but EFA and PCA are two different tecniques based on different extraction methods. The data used as example require EFA but your procedure is about PCA....
Many thanks for your feedback!!! I will have a closer look at it!!!
What's the difference between factor analysis and PCA?
It's a good question. In the presentation, I got the feeling that the term "components" (i.e. the term used in PCA) is used to mean "factors" here. I think that the difference is that PCA is seeking to reduce dimensionality of an analysis through creating components that "summarise" a greater number of variables, whereas for EFA the goal is to elucidate explicitly latent variables.
Exploratory factor analysis seeks for underlying dimensions that explain correlations among the variables, whereas PCA reduce variables into sets of factors (Principle Components) to explain the dataset with a fewer number of variables.
At 9:22 figures do not match your explanation. This creates confusion. Can you please check it?
I just dont understand the difference between EFA and PCA then? Isnt this PCA?
🙏🙏🙏🙏
🙂
Can i have only one factor?
Am I missing something here??? Why is the entire video on PCA...
Many thanks for your feedback! What are you missing?
PCA is not EFA. You said "In Statistics Exploratory Factor Analysis is also called Principal Component Analysis (PCA)" but that's not true. Those are two different analyses. That's a common confusion, but pls, check the differences.
help...
lol need payment
sadistic calculator
Present Yogmath
1 second ago
th-cam.com/video/3oLt6KaJ8w8/w-d-xo.html
Great video! As someone who has used SPSS for data analysis, I can definitely attest to its power and usefulness. The user-friendly interface and wide range of features make it a go-to choice for researchers and analysts alike.
I appreciated how the video showcased the various capabilities of SPSS, from data visualization to hypothesis testing. It's impressive to see how quickly and easily SPSS can produce meaningful insights from raw data.
If you're new to SPSS or considering switching to it for your data analysis needs, this video is a great introduction. And for those who are already familiar with SPSS, it's a helpful reminder of all the amazing things this software can do.
Thanks for sharing this informative video!
hi can i have your email address please? i have really struggling with my results and dont know how to correct them.