Exploratory Factor Analysis (EFA): Concept, Terminologies, Assumptions, Running, Interpreting - SPSS

[@valley-of-paradise1]

You have simply made life of doctorate scholars easier. Sincere gratitude

[@researchwithfawad]

Thanks. I am glad you liked it.

[@ishararathnayake6159]

Yes, Simply amazing..

[@n.yogalakshmiShasun]

A CLEAR CUT EXPLANATION ABOUT FACTOR ANALYIS FROM START TO END. UNDERSTOOD WELL.

[@researchwithfawad]

Thank you. I am glad you liked it.

[@B.Serrano-mi8xl]

Fawad, I'm an undergraduate trying to propose a scale development study for my senior thesis. Thank you so much. WIth no prior teaching on this, I was able to understand very easily with an undergraduate level of statistics knowledge. You made so many of the papers I'm using for research comprehensible. You are doing great work and I am very grateful. Resources like this are so helpful. Thank you thank you thank you

[@researchwithfawad]

Thank you. I am glad you liked it.

[@emmanuelonoja6251]

You have given a good lecture on this. Thank you for sharing your thoughts

[@AyeshaImran-w8x]

Its a wonderful explanation of EFA that made my research life easy. Thank you

[@researchwithfawad]

Glad it was helpful!

[@JuberTheLegend]

The most helpful and concise video I found on KZbin. thanks!

[@researchwithfawad]

Thanks. I am glad you liked it.

[@sandradsa1340]

Yes me too

[@darefadare6969]

How do I thank you enough... So straightforward and very clear. I'm grateful

[@researchwithfawad]

Thanks. I am glad you liked it. Please do share in your research circle.

[@sakshimalawaliya6451]

You simplied EFA for me! Thank you!

[@researchwithfawad]

Happy to help!

[@alamzubair321]

Excellent demonstration. No other video on KZbin explains EFA so well.

[@researchwithfawad]

Thanks. I am glad you liked it

[@priyamishra4706]

sir it's great but reporting mechanism kaha milega

[@sehreensara6402]

Absolutely

[@DFCinBE]

The best video out there explaining this concept. Congratulations and thanks so much!

[@researchwithfawad]

Glad it was helpful! Do Share in your research circle.

[@ehteshamali2893]

I will definitely mention your channel and your efforts in guiding me in my acknowledgements section of my thesis.

[@researchwithfawad]

Awesome, thank you!

[@YEJIFOOT]

Thank you very much. you are amazing instructor. I really appreciate you shared this valuable information for free. bless you so much

[@researchwithfawad]

You are very welcome

[@ranthony1556]

clear and concise from the start without unnecessary visaula or audio distraction

[@researchwithfawad]

Thanks. I am glad you liked it.

[@simonasusnea4836]

Thank you Fawad, your help is greatly appreciated.

[@researchwithfawad]

Glad it helped

[@missnerd4832]

Thank you so much. I'm writing my dissertation and have to use EFA. you've helped me a lot 😊

[@researchwithfawad]

Thanks for watching.

[@molisha4149]

thank you. After watching this many videos this helps me the most.

[@researchwithfawad]

Thanks for watching.

[@dr.vimalchandraverma443]

Amazing sir. Your way of teaching is superb

[@researchwithfawad]

So nice of you

[@polarbear2393]

Dear Fawad, thank you very much for all your videos here on KZbin. They have made a significant difference for me as I am a practitioner who has decided to embark on his journey of research without any prior background in science. Without your videos, I would have been lost. Thanks again and keep up the great work. Greetings from Switzerland!

[@researchwithfawad]

Pleasure. I am glad you liked the videos. I hope they are helpful in your research endeavor.

[@rejoiceihenetichidi8934]

Indeed, you are the best.

[@MsSaddal]

sir hats off to u...doing a great job.....all videos are worth watching as it has simple to understand material...thank you ..best wishes from punjab (amritsar)

[@researchwithfawad]

Thanks and welcome

[@sweet-girl6724]

Again all your videos are useful thank you a million 💕💕

[@researchwithfawad]

Glad you like them!

[@jacksonelias4790]

Excellent explanations and demonstration. It is really helpful. I will share the link that you provided with my students to get a better understanding.

[@researchwithfawad]

Thanks. I am glad you liked it. Thanks for sharing :)

[@musiclalalaaa]

Cant express how helpful this was thank you so much

[@researchwithfawad]

You're very welcome!

[@mdshahparan8029]

It helped a looooooooot, I became a fan of yours sir, respect, love and dua for you from my deep heart. Shukriya.

[@mdshahparan8029]

can you please help me with the interpretation format please, I couldn't find it on the comment section.

[@researchwithfawad]

Thanks. I am glad you liked it.

[@researchwithfawad]

Please visit researchwithfawad.com/index.php/lp-courses/data-analysis-using-spss/how-to-perform-exploratory-factor-analysis-using-spss/

[@saidgaryare4104]

Thank you, sir, for accepting my request and thank you for your kindness and generosity, may Allah bless you.

[@researchwithfawad]

Pleasure is all mine. I hope the session helps you understand EFA.

[@researchwithfawad]

You can also visit the link in description to download supporting documents

[@saidgaryare4104]

@@researchwithfawad Absolutely, sir, this is a big step of my data analysis. All praises are to you. Thank you

[@saidgaryare4104]

@@researchwithfawad yea I already downloaded sir.

[@deepakdass5286]

Great knowledge sharing Fawad, so much learning here. Thank you!

[@researchwithfawad]

Thanks. I am glad you liked it

[@sanfay5839]

You are an awesome teacher!

[@researchwithfawad]

Thanks. I am glad you liked it

[@parinayamjal5768]

Excellent explanation!! I understood most of it. Thank you

[@researchwithfawad]

Pleasure. I am glad you liked it

[@dirichiumunna9365]

This is so useful for understanding FA!

[@researchwithfawad]

Thanks. I am glad you liked it.

[@rfc8507]

It was amazing. Thank you so much. Allah bless you.

[@researchwithfawad]

Thanks. I am glad you liked it. Ameen

[@lalaliza315]

Thank you so much sir... It was best video on EFA I have ever seen on EFA... 🙏🙏🙏🙏

[@researchwithfawad]

Most welcome

[@Julietsilviya]

Clear and beautiful explanation 😊

[@researchwithfawad]

Thanks a lot 😊

[@benazirmahar01]

you have explained in a very easy way, so many thanks.

[@researchwithfawad]

Thanks for watching.

[@anitachopra9579]

Sirji, FA is making sense.thank you. Great work.

[@researchwithfawad]

Thanks and welcome

[@ggdelata9182]

Goodevening sir! I am a senior high school student from Benigno S. Aquino National High School, Tarlac, Philippines. May I request to have a permission to use your tool format/ your survey instrumentation format for our research study! We won't forget to cite your name sir... It would be a great help. Thank you!

[@researchwithfawad]

Sure you can use the format

[@nehagarg9379]

Very beautifully explained. Thank you

[@researchwithfawad]

Thanks. I am glad you liked it.

[@MaryanneNduka]

You are really a great teacher..

[@researchwithfawad]

Thank you! I am grateful for your appreciation.

[@cecyliaadamczak4301]

Very good video! I finally feel like I understand efa a bit :)

[@researchwithfawad]

Glad it helped!

[@ronaldmoirangcha638]

Very clearly understood from your lectures Sir. Can you also provide the reference textbook we can use for EFA and CFA 😇

[@researchwithfawad]

Thanks for watching.

[@EmmanuelMushi-l6u]

the lecture is great

[@researchwithfawad]

Thanks

[@abcdefghijkl534]

Out standing explanation sir....

[@researchwithfawad]

Thanks. I am glad you liked it.

[@AdmasuDemelash]

Ultrasmart explanation

[@researchwithfawad]

Thank You. I am glad you liked it.

[@greatmind3842]

Hello Dr. Fawad. Please can you include the article for Reporting EFA that you mentioned would be in the description? It's not there. Thanks so much

[@researchwithfawad]

Thanks for watching. Here is the reference Latif, K. F. (2018). The development and validation of stakeholder-based scale for measuring university social responsibility (USR). Social indicators research, 140(2), 511-547.

[@greatmind3842]

@@researchwithfawad Thanks so much, Dr. Fawad. I really appreciate your help. But, please I have another request, a question actually: Please after EFA, must I perform a CFA before using the factors in my SPSS analysis. In this case, no theories are involved. Again. thanks for your help.

[@cobbinahgloria2984]

Thank you very much Dr

[@researchwithfawad]

You are most welcome

[@tvjagdishrao4486]

very nice work

[@researchwithfawad]

Thanks. I am glad you liked it.

[@shimelisayele3704]

Thank you for your kindness

[@researchwithfawad]

Thanks for watching

[@Alhamzah_F_Abbas]

Great explanation thank you, prof. regarding CFA can we use it in PLS?

[@researchwithfawad]

Thanks. For CFA use AMOS

[@abdul-ganiyufuseini2920]

What a great presentation. This has been so helpful for my PhD. project. Thank you

[@researchwithfawad]

Thanks. I am glad you liked it

[@MUSICAHOLIC999]

Dr. in the total variance explained table... what is the component referring to? is component no. 1 the same as item no. 1?

[@researchwithfawad]

This refers to the factors that can be extracted. Normally we look at the Eigen Value to assess the no. of factors to be extracted.

[@malikasharma1390]

Hello sir. First of all i would like to appreciate ur video. This video is one of the informative one. Sir, can u plz tell me that if KMO value is 0.705, is it consider appropriate ? As u mentioned in ur report that MSA values above 0.800 are consider appropriate for factor analysis. Help me with this sir. Thanks regards.

[@researchwithfawad]

Thanks. I am glad you liked it. 0.70 is also fine.

[@myrarao4148]

Amazing video! Thank you for this. I had an urgent doubt, if I have 5 factors but only 4 are significant i.e more than 1, is my data wrong?

[@researchwithfawad]

Thanks. I am glad you liked it. No it is not wrong, when you say significant, do you mean eigenvalue over 1. Are there any items loading onto factors that they should not? Check and remove the items, also if you are using an already validated scale, no need for EFA.

[@myrarao4148]

@@researchwithfawad I am using validated scales for each construct, but do I need to check again since its all put together? and yes I meant eigenvalue

[@tenzindawa3338]

your video is very helpful..... thank u sir

[@researchwithfawad]

Glad to hear that

[@SorayaBintiIshak]

hi dr.fawad, im doing presentation in my classroom. this video helped me a lot. However, may i know, how can i get the sample data to run in the spss?

[@researchwithfawad]

Thanks. I am glad you liked it. I am afraid this data is part of an ongoing research.

[@AsadKhan-lu8kx]

Great video, just need a bit more clarity on the sample size though. Given that you have 19 items/observed variables in this demo , I would presume (based on the rule of thumb described here), that you had anywhere around 19*4 or 5 sample size for this video, which is quite large for a pilot study or questionnaire validation. Would greatly appreciate if you could clarify this bit, thanks!

[@researchwithfawad]

Thanks. I am glad you liked it. Thats the minimum one should have, the higher the sample soze, the better.

[@nibeditashyam5614]

Just wanted to ask that in EFA is it compulsory for all the variables after extraction to represent one latent factor?

[@researchwithfawad]

Thanks for watching. Use EFA only of the scale is self developed. If you have adapted or adopted the scale, there is no need for EFA.

[@evansaddison7332]

Iam well educated about the factor analysis

[@researchwithfawad]

Thanks for your interest.

[@PreetiPreet-v9d]

Nice...very well explained

[@researchwithfawad]

Thanks for watching.

[@imanejai6953]

Thank you for your clear explanation, but how to calculate CR and AVE?

[@researchwithfawad]

Thanks for watching. That will have to be done manually. You can put in the loadings values into the calculate available on the website (Under Excel Sheet). researchwithfawad.com/index.php/download-research-resources/

[@Varunbhalla999]

Thank you very much sir.

[@researchwithfawad]

Most welcome

[@ggtharinduyg]

Thank you Dr. Fawad for the wonderful video which was very much helpful. Appreciate it if you could clarify whether it is ok to try running EFA with different fix factor numbers when no meaningful results could be obtained running EFA without specifying any predefined factor number.

[@researchwithfawad]

Thanks. I am glad you liked it. If you are using an existing scale, no need for EFA. Instead of fixing the factors, first try to do EFA on individual factors this will help clean the factors.

[@kwasiaduboamah7867]

Very insightful

[@researchwithfawad]

Thanks for watching.

[@adnansabzi]

It ll be good to share slides of the lecture with viewer's

[@researchwithfawad]

Thanks. Please visit the link in the description. Thanks.

[@hosseinazadmanesh9089]

Amazing

[@researchwithfawad]

Thank you! Cheers! I am glad you liked it.

[@army-cx7nk]

Many thanks Fawad. -1- What is your hypothesize factor in this your video, as I see some use 0.3, and 0.4(Spence et al 2001, The structure of anxiety symptoms among preschoolers, Behav Res Ther. 2001; 39(11):1293-316) -2- In this Spence et al, on Page 1298, what does this statement mean '' Of the 28 items, 26 had a loading in excess of 0.40 on their hypothesize d factor''. -3- What is the ideal hypothesis factor 0.3, or 0.4, or 0.5? -4- In this same paper, on Table 2, what is the meaning of the number in bracket of the factor loading means & how do I locate such in this your video? -4- In reporting EFA, are we allowed to write out the Initial Eigenvalue % of Variance together with each factor Eigenvalue, Example Factor 1: (Eigenvalue 5.94; 24.718%)

[@researchwithfawad]

The statement "Of the 28 items, 26 had a loading in excess of 0.40 on their hypothesize d factor" means that the factor loading for the items was over the minimum 0.40 on the expected factor. Preferred minimum loading is 0.50 Just report factor and their eigenvalue after removing the items that do not load well.

[@benazirmahar01]

superb , i am grateful

[@researchwithfawad]

Thanks for watching.

[@TarannumSiddique-kd2jc]

link for the notes in description is not working ...please do something

[@researchwithfawad]

Thanks for letting me know. Please check this link researchwithfawad.com/index.php/lp-courses/data-analysis-using-spss/how-to-perform-exploratory-factor-analysis-using-spss/

[@subhabratamitra4910]

God bless you !

[@researchwithfawad]

Thanks for watching

[@vincenarh8317]

That is very helpful. Thanks!

[@researchwithfawad]

Thanks. I am glad you liked it

[@nursufyanaulfachemustaffa6554]

Greetings Dr. I have adopted questionnaire items from well-established studies and I have translated the questionnaire to Malay Language through the use of certified translators. My question is, should I conduct EFA? Or can I just proceed with the analysis in PLS-SEM?

[@researchwithfawad]

Thanks for watching. It is not required.

[@nursufyanaulfachemustaffa6554]

@@researchwithfawad Many thanks to you, Dr.

[@ezeugwundubuisistan8904]

So helpful. Thanks a lot

[@researchwithfawad]

Glad it was helpful!

[@ma.erenitabahian874]

Very good, I learned a lot!

[@researchwithfawad]

Glad it was helpful!

[@shayistamajeed9889]

Sir if we are adapting a scale for a study in a non western culture like India that was originally developed in a western culture, do we need to go for CFA only ?

[@researchwithfawad]

Yes, go directly for CFA and report Reliability and Validity

[@anasaslam9798]

Thank you sir...

[@researchwithfawad]

Most welcome

[@nadiahfazil6501]

Thank you Dr. Fawad for this helpful video. Is it possible we retain items that load onto another factor as the said factor's item instead of deleting them? Perhaps are there any articles that I can read regarding this? Would be grateful for your help.

[@researchwithfawad]

Pleasure. I am glad you liked it. No, those items shall be removed.

[@kh.abdulrehman]

Why did we remove RD2 instead of including it in the first subscale? RD2 is loading well with the first subscale questions.

[@researchwithfawad]

The reason is that the item doesn't belong that construct.

[@bizmasters8499]

sir. my research is based on confirmation of past theories regarding Working capital management practices. which factor analysis method shall i use. is it CFA ? if CFA , it it same way of Cheking EFA.

[@researchwithfawad]

CFA is different from EFA. Use CFA

[@ehteshamali2893]

Asalamualikum sir, I'm really grateful for your videos. You really helped me with moderation analysis using smart pls. Sir I wanted to ask regarding adapted questionnaire and the use of factor analysis. Sir, can we set the number of factors based on the adapted questionnaire? When we run factor analysis through SPSS, it gives two options: 1) Based on eigen values 2) fixed number of factors. The papers from where I have adapted my questionnaire has total of 7 items/ dimensions/ themes in total. So can I use the option number 2 - fixed number of factors = 7? I have adapted my questionnaire from 3 different papers. Should I proceed with option 1- based on eigen values or option 2 - fixed number of factors. Thank you so much sir for your help. :) I have learned a lot from you.

[@researchwithfawad]

Thanks for watching. No need for EFA is scale is adapted from a validated study.

[@ehteshamali2893]

@@researchwithfawad Thank you so much sir :)

[@Vivek-np9vm]

Sir, is varimax good for EFA or should we go with Promax? I was thinking principal axis and promax. Kindly let me know.

[@researchwithfawad]

use Varimax.

[@Vivek-np9vm]

@@researchwithfawad ok Sir

[@naseermuhammad1056]

MashAllah well explained.

[@researchwithfawad]

الحمد للہ I am glad you liked it

[@MohdZaid-f3y]

SIR, I am conducting RESEARCH ON Topic SOCIAL IMPACT ON CONSUMER Behaviour and i am using TAM Theory factors i.e, PU & PE under the Heading PERCEPTION TOWARDS SOCIAL MEDIA PLATFORMS...should i have to conduct CFA and EFA both or only CFA And plz tell cfa is carried out from spss only or from Smart PLS too..

[@researchwithfawad]

If you are using a questionnaire that is from existing research, there is no need for EFA.

[@rukhsanamehmood6138]

Jazakallah khair Sir

[@researchwithfawad]

Thanks. I am glad you liked it.

[@digitalbuddypk]

Salute you sir. You are great

[@researchwithfawad]

Thanks. I am glad you liked it

[@syedarabianaqvi5336]

Can you apply factor analysis on an already established scale?

[@researchwithfawad]

You can, but it is not required

[@JohnRBaker-gw6oy]

Very good explanation. Thank you.

[@researchwithfawad]

Thanks. I am glad you liked it.

[@dasminedsouza9208]

This really helped. Thank u for the simple descriptive explanation. Just had one query. How did you come up with the value of 60.798 as the percentage of variance? Because when I calculated the total variance of the 3 factors from the given example, I got a value of around 56. It would be great if you could clarify.

[@researchwithfawad]

Thanks for your message. Actually, i entered the final variance value after removing the items that were cross-loading, not loading at all, or low loading.

[@dasminedsouza9208]

@@researchwithfawad Alright. Thank you for the clarification!

[@dasminedsouza9208]

@@researchwithfawad Alright. Thank you for the clarification!

[@miltontasi1714]

what an excellent demonstration beyond any other explanation

[@researchwithfawad]

Thanks l. I am glad you liked it.

[@asupriya808]

Sir Sample size should be in 50 Is there any possibility to for 40 respondents If no mean wt kind of tools were taken. ?

[@researchwithfawad]

Even 50 is low. if this is pilot testing, you do not need EFA. Simple Reliability test is enough.

[@yaseminrao8706]

thank you sooooo much

[@researchwithfawad]

Thanks for watching

[@kossonouprunelle7576]

Thank you for the presentation, i have a questionnaire with 5 likert scales variables, 7 nominal variables. Can i only use the 5 likert scales variables to do the exploratory factor analysis? or choose 2 or 3 between them. Also, is exploratory factor analysis the same as convergent abd discriminant validity? I am a little bit confused. Thanks in advance for your reply.

[@researchwithfawad]

Yes, you cam run EFA on your likert scale items

[@kossonouprunelle7576]

@@researchwithfawad Thanks for your reply. After doing the EFA, in should do again convergent and discriminant validity in the data analysis validity section?

[@researchwithfawad]

Why did assess validity before, you shall do it after doing CFA, if you are doing CFA.

[@kossonouprunelle7576]

@@researchwithfawad Thanks for your reply. I am using SPSS software. Can i first doing EFA using dimension reduction command on spss, and then do discriminant and convergent validity for the validity test? Thanks

[@researchwithfawad]

Yes, you can. But why not use SEM based tools

[@aliceyalenga7267]

This is so helpful but how can I contact you for further information and questions?

[@researchwithfawad]

You can email me on [e-mail]

[@tenzindawa3338]

sir one question plz if i adopt a scale from pretested one but I change one word of it ie. instead of apparel I wrote clothes in my questionnaire. so do I need to apply EFA

[@researchwithfawad]

No, apply EFA if you develop a new scale. Else, not required.

[@kirangupta221]

Hello Sir...thank you so much for such a clear explanation. I had a doubt about interpreting the rotated component matrix. In that table, you mentioned that RD factors were supposed to get loaded on component 2. But all factors are loading on component 3. What could be the possible explanation for this? Should it mean RD factors are measuring Philanthropic responsibilities?

[@researchwithfawad]

Thanks. I am glad you liked it. In simple terms, if an indicator doesn't load on its respective factor it shall be removed from further analysis. I hope this helps.

[@kirangupta221]

@@researchwithfawad Thank you so much Sir !

[@sohaibkhan6707]

Sir, it shows this after analysis. Please help me out. There are fewer than two cases, at least one of the variables has zero variance, there is only one variable in the analysis, or correlation coefficients could not be computed for all pairs of variables. No further statistics will be computed.

[@researchwithfawad]

The issue is with respect to missing values in your data. Refer to this link www.ibm.com/support/pages/factor-analysis-returns-error-there-are-fewer-2-cases-variable-0-variance-or-only-1-variable

[@shalinireddynaini7197]

good morning sir, do we have any option for conducting the factor analysis for only subgroup of participants, for example i want to conduct factor analysis seperately for male participants and seperately for female participants and compare the factors obtained by both of them. so is there any option in factor analysis for selecting particular group? pls explain

[@researchwithfawad]

Yes, you can do it. Please split the file and analyze cases by groups.

[@shalinireddynaini7197]

@@researchwithfawad split the file manually? do we have any option to select only male or female in factor anlaysis? i see selecting variable option down the independent variable option in factor analysis tab? does that serve the purpose of selecting and analysing only a particular group ? or everything should be done manually like creating two files for male and female and conducting factor analysis seperately?

[@shwetasingh6912]

What to do if factor loading shows a negative sign? Should we remove the variables?

[@researchwithfawad]

The negative sign shows that the item has a negative correlation. You may need to reverse code that item.

[@ParulManchanda]

I have a question, relating to EFA; Is normality of data an essential thing to be checked before proceeding for EFA

[@researchwithfawad]

You can check it using Skewness and Kurtosis.

[@2mtk]

Thank you so much! Great video. Is it possible for you to share the PowerPoint presentation either as pptx or pdf?

[@researchwithfawad]

Thanks. I am glad you liked it. Here is the link to the tutorial researchwithfawad.com/index.php/lp-courses/data-analysis-using-spss/how-to-perform-exploratory-factor-analysis-using-spss/

[@googledude5649]

Hi Fawad ! - Since there is a difference in PCA and FA (eg. Axis) the commonalities is slight different. In PCA we use >0.5 as a threshold (extraction) - but what to do in Principle Axis analysis hence the initial values is not 1. Some say an extraction value >0.3. Have you considered this or do you think it should be >0.5 as well ? Regards !!

[@researchwithfawad]

For Principal Axis communality above. 40 is acceptable. But sometimes you have to take these as subjective recommendations, you should also keep an eye open for loadings.

[@googledude5649]

@@researchwithfawad Thank you for your quick reply, Fawad. Best of luck ! Regards

[@mayoravlogs]

Thank you 🙏

[@researchwithfawad]

Thanks. I am glad you liked it.

[@mayoravlogs]

It's not about liking, but it's all about concept clearance. For last three months I have gone through many KZbin channel regarding factor analysis, even I had already been through your channel before. As I got this video, I got clear idea about it. I will definitely recommend others to go through it for concept clearance.

[@seemiTariq]

its great sir

[@researchwithfawad]

Thanks. I am glad you liked it.

[@hassanbashir8668]

Hi sir, I need a little assistance..actually in my research I found one item in original scale that is culturally inappropriate..can I remove this item by simply taking permission from author because there is only one factor in that that scale or I have to do factor analysis again? Thank you.

[@researchwithfawad]

Yes as long as you can provide the justification.

[@hassanbashir8668]

@@researchwithfawad ok sir thank you so much..

[@optripathi616]

Reporting is very useful

[@researchwithfawad]

Thanks. I am glad you liked it.