22. SPSS AMOS | Path Analysis as Structural Model

[@ハンウズマ]

now i m at this point to do SEM modeling. thanks for sharing it

[@researchwithfawad]

Thanks for watching

[@abdulrafay134]

excellent explanation

[@researchwithfawad]

Thanks. I am glad you liked it

[@hereagain-ot9ml]

Thank you very much sir, I am very grateful for this video. coming from STEM to Non-STEM, your video nailed it at doing it myself. Please I have a couple of questions to ask. At first my model did not achieve a good fit, then I followed step by step your vuideo and achieved model fit and hence all is pretty cool in my measurement model. However, I employed common method bias using Latent common method to ensure common Method bias is absent. How will I go about this in the path analysis? Should I add all indicator per constructs to obtain cpmposite and do same for the Latent common variable which was matched to all the indicators? Please I will be very grateful for your reply.

[@researchwithfawad]

Hi. Yes please add all indicators.

[@Alhamzah_F_Abbas]

New knowledge I just earned thank you, prof. I have two questions: 1- why don't we compute the path analysis along with the measurement model at once as long as both of them give us results without computing composite variables? and is it possible? 2- as long as we assess the measurement model (Model Fit, Reliability, and Validity). do we need to check the model fit when we run path analysis as well? I will be very grateful for your guidance and feedback.

[@researchwithfawad]

Thanks. I am glad you liked it. Full Structural model is preferred over path model. Model fit shall be assessed for both measurement and structural model.

[@heyDk007]

Sir, for my measurement model to be fit common method bias using Latent common method was adopted. And it showed a significant fit of my model. So while creating structural model how will I create composite variable of this common factor? Shall i take mean of all the variables used in the measuremnt model? Also, sir another doubt is that in structural model do i need to use all the variables or should I remove the variables that were not used in measurement model?

[@researchwithfawad]

You can take mean. Remove all that are not part of the measurement model.

[@heyDk007]

@@researchwithfawad Sir, Is path analysis a structural model? Can I stop my analysis after this?

[@researchwithfawad]

Do full structural model instead

[@TheCrn28]

Thank you very much for the video Dr. Fawad. I want to ask that what should we do if we have a multi-dimensional scale as "service quality". As you may know it has 5 sub-dimensions. Should we calculate the mean value of the each dimension and then calculate the mean value of those dimensions? I hope I could explain it :)

[@researchwithfawad]

If you would like to do a path model as done in this video, yes, simply take mean of all indicators (no need to do it by dimensions, results won't change). However, if you want to do full structural model, you will need to treat SQ as a Higher Order Construct with dimensions at lower order.

[@soundaryamsd1009]

Thank you for the detailed explanation. When to use path analysis and SEM analysis?

[@researchwithfawad]

Thanks for watching. Use it when the variables are observed not latent.

[@MeandYoutube516]

Dear Prof, thank you for explaining. I have one independent variable and 4 dependent variables. Can i still do a path analysis?

[@researchwithfawad]

Thanks for watching. Yes you can

[@muhammadaamirkhan8182]

If we don't achieve the required Model Fit values in path analysis then how we can/should do to improve them?

[@researchwithfawad]

Please refer to this video. kzbin.info/www/bejne/aYqalYKOfrOJaKc

[@profkhan6576]

Respected Sir, while running structural model, should we use measurement model,'s remaining items (after model fit ) or all items of questionnaire. Plz guid.

[@researchwithfawad]

Please use items remaining after measurement model assessment

[@novairahazeemat6538]

AssalamAlikum Thanks for this video. I have a question if there is no item for dependent variable then how can we check hypotheses in cfa ? Please reply. Thankyou!

[@researchwithfawad]

WAS. If it is not measured, then we cannot.

[@novairahazeemat6538]

Thankyou so much