#SmartPLS4

[@lilac_venus]

Thank you for the vid! I have a question, my model has this "NaN" label beside each circle (unlike the model shown in the vid). Does that mean anything? Or can i just ignore it?

[@researchwithfawad]

How many items do you have in each construct.

[@fenglin7864]

Hi, Professor. In the video, arrow P3 represents the relationship between Y1 and Y3. My study does not hypothesize the direct relationship between Y1 and Y3 but through a mediator Y2. Therefore I would like to consult you on whether I still have to draw an arrow between Y1 and Y3 when I run SmartPLS. I also tried to run without/with the arrow, and I found both the results are different. Thank you so much.

[@researchwithfawad]

If you are not hypothesizing the relationship, no need to draw the arrow. You are proposing complete mediation.

[@fenglin7864]

​@@researchwithfawad Thank you so much for your kindness.

[@sadhanaadhikari5863]

@@researchwithfawad so we don't have to explain mediation analysis?

[@anggoran7398]

Hello, I have the same question. In my model, I assume that the mediating variable is complete mediation. But, since we have to test the direct effect (which any complete mediation doesn't have), then how? I wouldn't want to assume that there is no direct effect without testing it before. Do we must hypothesize that the mediating variable is partial mediation in the first place like in the video?

[@sanazehra6621]

I have the same question @@sadhanaadhikari5863

[@susanna5254]

Hey! Thanks so much for your videos. One stupid question: What is the best way to test the validity of my constructs with spss (or smartpls)? For reliability it's clear to me...

[@researchwithfawad]

For validity see video on validity on the channel/playlist on SmartPLS4. Please check for AVE and Discriminant Validity.

[@nehariaz1405]

9:10 one tail and two tail se koi difference ata h kia results and what should we use while testing hypothesis one tail or two tail..??please answer

[@researchwithfawad]

Yes, there are differences. Use One Tailed if you have proposed a positive or negative relationship. Use Two Tailed if you are not sure about the direction of relationship that whether it will be positive or negative.

[@mukhtar1319]

Asalmualikum sir… In my case p1 and p2 are negative and p3 is postive. Multiple of p1 and p2 is mathematical positve and in essence also these two paths are supposed to be negative as per theory and literature. With direct effect positive and indirect effect also positive, when actually the paths to and from mediators are actually negative, are significant at .05. How shall I interpret this result.

[@researchwithfawad]

If they are supposed to be negative and they are negative, interpret and support hypotheses as required. For mediation look for significance of indirect effect instead of signs.

[@mukhtar1319]

@@researchwithfawad Thankyou sir much sir…. That shall really help…

[@مسلمعبدالله-س7ن]

Great effort Dr. I have a question please: regarding consistent PLS-SEM algorithm, most of what I have read on the internet tells that it should be used with models containing reflective constructs only, as opposed to standard algorithm which you used in the video. In my case, consistent algorithm gives standardized path coefficients greater than one, while standard algorithm gives normal results. this only happens when I enter composite scores of factors of the independent variable in the model. Is it real that consistent algorithm is more accurate for reflective construct? and can I still use standard algorithm for reflective constructs? Thank you.

[@researchwithfawad]

Please use normal PLS Algorithm and Bootstrapping.

[@مسلمعبدالله-س7ن]

@@researchwithfawad Thanks alot Dr

[@Alhamzah_F_Abbas]

Easy to understand. Thank you

[@researchwithfawad]

Glad it was helpful!

[@AP_Pathak]

my diagram is not showing the beta coefficient along with significant level

[@researchwithfawad]

Check for the combo boxes above the diagram and select the required statistics.

[@dhanukaperera594]

Great Explanation. Thanks

[@researchwithfawad]

Glad you liked it

[@majdamuhammad3208]

can't thank you enough!

[@researchwithfawad]

I am glad you liked it.

[@susanna5254]

13:00 two of my betas are negative, one is positive. Multiplying it, of course i get a positive result. So i've got a complemantary partial mediation, right?

[@researchwithfawad]

Yes, but do not interpret it based on simple multiplication, please see if the beta were meant to be negative

[@susanna5254]

@@researchwithfawad Thank you! What do you mean by "meant to be negative"? The betas being negative makes sense.. i mean it makes sense, that the influences are negative. if thats what you mean?

[@mukhtar1319]

Please connect, I am having the same problem.. I want to know how to interpret it.. Thanks

[@susanna5254]

@@mukhtar1319 hey, sorry in the meanwhile I already submitted my thesis. In my thesis I just spoke about a partial mediation. Don't know if that helps you.

[@mukhtar1319]

Thankyou… And best of luck

[@shreyamukherjee9792]

Reference paper please

[@researchwithfawad]

You can quote the book by Hair et al. A Primer on Partial Least Squares Structural Equation Modelling

[@shreyamukherjee9792]

Thank you