Using this series for my undergrad thesis study Seems really apt, probably better than online paid courses too (though I haven't gone through them rigorously) Love your approach and steps!! Thank you!!
Thanks, sir for great effort For which Likert scale we can delete these cases with SD less than 0.25. Likert scale may be on 3, 4,5,6,7,9 points. Greetings
Respected Sir, thanks for this very informative video; can you please tell us should we delete cases with SD less than 0.25 even if we have a 5-point likert scale? Bcs, there is not information about what to do if we have 5 point likert scale, or 9 or 3 etc. in Collier's book. Is there a formula to calculate SD cut off point for the purpose of cleaning cases for 5 point likert scale? Thanks a lot🙏@@researchwithfawad
Dear Dr. Fawad, great job, you are doing fantastic job that will last forever. Kindly clarify one question of mine based on your expertiese: There are four Latent constructs, each containing 6 items/questions. Where in my upcoming study, I want to take all those 24 items (6 questions each) and test one higher order construct as an independant Variable...kindly confirm in this case, if any of my item is deleted because of poor loading out all those 24 items, will it cause any trouble? In simple words there were four different constructs representing one higher element, but I want to treat all these items under one higher order construct. Thanks
Thanks. I am glad you liked it. No deleting n item from LOC will not cause ajy trouble. LoC will have to be validated. After LOCs are validated, validate HOCs as well.
Hi Dr Fawad, thank you for these lectures, I've been using them to teach myself SEM for my PhD. For my work I need to use the references that found up to 20-30% of missing data can be remedied with imputation. You cite Hair et al., 2009, and Eekhout et al., 2013, however I'm unable to find the full references anywhere. Would you be able to send me these as full references please? Thank you
Thanks. I am glad you liked them. Please find below the full references Eekhout, Iris, Henrica De Vet, Jos Twisk, Jaap P.L. Brand, Michiel de Boer, and Martijn W. Heymans. (2013), “Missing Data in Multi-Item Instrument Were Best Handled by Multiple Imputation at the Item Score Level”, Journal of Clinical Epidemiology, 67 (3), 40-55. Hair, Joseph F., William C. Black, Barry J. Babin, and Rolph E. Anderson. (2009), Multivariate Data Analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.
Dear Sir, I have seen all your videos about PLS and also AMOS, so helpful for me. May I ask your a questions about the comments I have received from a reviewer. In the model of my study, there are two formative constructs which made me to run PLS. I've conducted EFA for exploring the dimensionality. The reviewer suggested me to do also CFA after EFA, and then run the model with Smart-PLS. I'm confused about if it will work like that ..... Thank you very much for your help since now. Best regards
If your items are based on established scale you do not need to perform EFA. As for CFA, Smart-pls only deals with part of it and can assess reliability and validity. Whereas Model fit indices are not yet established for PLS. The indicators for reflective constructs ar required to be correlated, whereas the indicators for formative constructs are not required to be correlated. As such, formative constructs should not be expected to properly factor in an EFA
Extremely helpful video on data screening. Appreciate your effort. It has helped me understand the process of data screening completely. Thank you!
Thanks. I am glad you liked it.
Using this series for my undergrad thesis study
Seems really apt, probably better than online paid courses too (though I haven't gone through them rigorously)
Love your approach and steps!!
Thank you!!
Thank you. I am glad you liked it.
Thanks, sir for great effort
For which Likert scale we can delete these cases with SD less than 0.25. Likert scale may be on 3, 4,5,6,7,9 points.
Greetings
Thanks for Watching. Yes Please.
Respected Sir, thanks for this very informative video; can you please tell us should we delete cases with SD less than 0.25 even if we have a 5-point likert scale? Bcs, there is not information about what to do if we have 5 point likert scale, or 9 or 3 etc. in Collier's book.
Is there a formula to calculate SD cut off point for the purpose of cleaning cases for 5 point likert scale? Thanks a lot🙏@@researchwithfawad
Great knowledge
Keep watching
Where can I find a dataset please to practice your explained sessions? Please help.
Please visit www.researchwithfawad.com
Dr Fawad, do you have any reference that states that we have to delete the response whose STDEVP is less than.025, any paper or book that states this?
Please refer to
Collier, J. E. (2020). Applied structural equation modeling using AMOS: Basic to advanced techniques. Routledge.
Thank you, prof. can we use the same process when we use smartPLS?
Pleasure. No.
@@researchwithfawad in this case ,how we gonna deal with missing data in PLS?
There are mean replacement options that you can choose from when running the analysis
Really useful
Glad to hear that
Dear Dr. Fawad, great job, you are doing fantastic job that will last forever.
Kindly clarify one question of mine based on your expertiese:
There are four Latent constructs, each containing 6 items/questions. Where in my upcoming study, I want to take all those 24 items (6 questions each) and test one higher order construct as an independant Variable...kindly confirm in this case, if any of my item is deleted because of poor loading out all those 24 items, will it cause any trouble? In simple words there were four different constructs representing one higher element, but I want to treat all these items under one higher order construct.
Thanks
Thanks. I am glad you liked it.
No deleting n item from LOC will not cause ajy trouble. LoC will have to be validated. After LOCs are validated, validate HOCs as well.
If the data distribution is not normal, are we allowed to manipulate the data? and how? (all data are in likert spectrum) Thank you in advance. 😇🥰
Thanks for watching. No, we are not allowed to manipulate the data.
Hi Dr Fawad, thank you for these lectures, I've been using them to teach myself SEM for my PhD. For my work I need to use the references that found up to 20-30% of missing data can be remedied with imputation. You cite Hair et al., 2009, and Eekhout et al., 2013, however I'm unable to find the full references anywhere. Would you be able to send me these as full references please?
Thank you
Thanks. I am glad you liked them. Please find below the full references
Eekhout, Iris, Henrica De Vet, Jos Twisk, Jaap P.L. Brand, Michiel de Boer, and Martijn W. Heymans. (2013), “Missing Data in Multi-Item Instrument Were Best Handled by Multiple Imputation at the Item Score Level”, Journal of Clinical Epidemiology, 67 (3), 40-55.
Hair, Joseph F., William C. Black, Barry J. Babin, and Rolph E. Anderson. (2009), Multivariate Data Analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.
@@researchwithfawad Thank you so much
Dear Sir, I have seen all your videos about PLS and also AMOS, so helpful for me. May I ask your a questions about the comments I have received from a reviewer. In the model of my study, there are two formative constructs which made me to run PLS. I've conducted EFA for exploring the dimensionality. The reviewer suggested me to do also CFA after EFA, and then run the model with Smart-PLS. I'm confused about if it will work like that ..... Thank you very much for your help since now.
Best regards
If your items are based on established scale you do not need to perform EFA. As for CFA, Smart-pls only deals with part of it and can assess reliability and validity. Whereas Model fit indices are not yet established for PLS.
The indicators for reflective constructs ar required to be correlated, whereas the indicators for formative constructs are not required to be correlated. As such, formative constructs should not be expected to properly factor in an EFA