#SmartPLS4

Higher-Order Constructs are increasingly used in research to perform complex modelling using #SmartPLS. There a number of different types of Higher-Order Constructs. The focus of this session on how to assess Reflective-Reflective (Ref-Ref) higher order construct in SmartPLS 4.
Previous Session (Understanding Formative Constructs)
• #SmartPLS4 Series 15 -...
Next Session (How to Report Reflective-Reflective Higher Order Construct)
• #SmartPLS4 Series 17 -...
Measurement Model with All Lower Order Constructs
• #SmartPLS4 Series 11 -...
Do you have any questions, tips, or ideas? Let me know in the comments section below!
About:
Research With Fawad is a Research initiative to help young and early career research learn the research tools and techniques. It is an initiative to share research knowledge and help researchers learn to research the easy way.
Learn more about Research With Fawad at www.researchwi....
#SmartPLS4 #SmartPLS #PLSSEM #SEM #HigherOrder #HierarchicalComponentAnalysis

[@KatibKhan-s6d]

hello! sir in case we dont follow R-R model and show the relationship of 4 variables of IM with bootstrapping in the same way you did while calculating LOC without adopting latent variable score...do measurement and structural model results changes or remains same as R-R model

[@researchwithfawad]

The remain the same.

[@KatibKhan-s6d]

@@researchwithfawad sir if the results remains same without adopting reflective-reflective higher order constructs. why there is need to calculate hoc of R-R rather we can show the relationship of IM 4 variables with each factor in the given model separately by forming the hypothesis for each 4 variables of IM and running simple LOC and structural model bootstrap.

[@KATIB-k9n]

HELLO SIR....as you calculated the reflective- reflective model in case of internal marketing All the 4 variables of IM are first calculated on LOC then moved towards HOC what happens if rather than removing 4 variables of IM we run BOOTSTRAPPING with similar relationship in that case does it reflects R-R model

[@researchwithfawad]

Same.

[@啾啾-h3y]

Dear dr.My measurement model 's Alpha is 0.697. If I Should report it ? If I report it ,is there some papers to cite?Thanks for your response.

[@researchwithfawad]

It is acceptable, you can report CR with it, that will be over 0.70.

[@rinkyyadav4654]

Sir... What to do if lower order construct shows reliability and validity of all lv but when we calculate reliability and validity for higher order construct after calculating lv score.... It is below the threshold value.... What to do in this case??

[@researchwithfawad]

You can perform similar procedure as in the LOC, may consider removing items.

[@AP_Pathak]

Sir after making HOC, the PLS says it cannot run, says the model is invalid check for the unconnected construct but my all construct are in blue what to do sir?

[@researchwithfawad]

Please double check by reducing the size of the canvas, there may be a construct that you cannot see.

[@Alhamzah_F_Abbas]

Great work. how about F-R and F-F?

[@researchwithfawad]

Hopefully soon.

[@pratikasingh0490]

In a reflective-reflective model, for one of the HOC my cronbach alpha is 0.489 but CR and AVE are fine. For another construct the cronbach alpha is 0.596, rho_a is 0.624, rho_c is 0.830 and AVE is 0.709. Are these values fine? Also, four of my HTMT values are coming more than 0.9. What can I do solve the HTMT problem?

[@researchwithfawad]

Please watch kzbin.info/www/bejne/o2fXhp99nq1sla8 The video may help

[@pratikasingh0490]

@@researchwithfawad Thank you Sir

[@chathurikajayasooriya5753]

Hi Dr, May I know how to deal with the HOC-dependent variable when there are more than one HOC independent variable?

[@researchwithfawad]

The process remains the same even if it is a dependent variable at Higher Order. There is a video on the channel if your higher order is formative.

[@researchwithfawad]

Please watch kzbin.info/www/bejne/haOleKqjqNKtbZo

[@hulktang7710]

Dear Dr., When I ran my model in the first stage, an item (BLOYAL_1) restricts the discriminant validity of its parent construct (Brand Loyalty) compared to another construct -- both are not HOC; the item also had a somewhat insufficient outer loading and if it were to be removed, the said discriminant validity would be established. I am unsure whether to [1] delete the item immediately in the first stage, then run the path analysis again to obtain the latent variable scores (i.e., after having deleted the BLOYAL_1) or [2] keep the item, acquire the latent variable scores (i.e., not deleting BLOYAL_1), use them in the second stage, then delete the item in this second stage. I am confused at the LVS are quite different in each scenario. I hope you can provide some suggestions or a definite answer. Thank you

[@researchwithfawad]

Delete the item at lower level.

[@hulktang7710]

@@researchwithfawad Much appreciated, sir

[@frednarh6866]

Asalamualaikum, Please after generating the latent variables, how do you turn the indicators from reflective to formative. thank you

[@researchwithfawad]

Please right click on the latent variable and click invert measurement model

[@shadeel9293]

Sir when I change my construct into Reflective- reflective and run the model my Cronbach alpha comes down; what should I need to do? Please guide me.

[@researchwithfawad]

What is your alpha and CR?

[@shadeel9293]

@@researchwithfawad alpha 0.355, CR 0.755, AVE 0.646.

[@researchwithfawad]

Dont worry about Alpha. You do not even need to report it. Just report CR and AVE.

[@shadeel9293]

@@researchwithfawad sir thank you so much for your speedy response