Problem? Significant but small t-values in outer model

[bsmithx]

I have an independent LV with four indicators - each is a difference scores. The path coefficients and t-values for the inner model look very good. Bootstrap also shows the t-values for this particular LV are significant but they are lower than 8.00. Is this a problem?

[christian.nitzl]

Your t-value with 8.00 is much above the often cited critical value of 1.65 (10%) or 1.96 (5%).

Best regards,

Christian

[woltersfrank]

Hello christian,

Is the "T Statistics (|O/STERR|)" in the bootstrapping report the so called t-value? Or is this something else?

Thanks for your reply.

cheers Frank

[christian.nitzl]

Hey Frank,

You right, the reported "T Statisics (|O/STERR|)" is your t-value.

Regards,

Chrisitan

[JELASSI]

Hello,

I am looking for the t-values of my formative indicators?

Is it the "T Statistics (|O/STERR|)" in the bootstrapping report under “outer weights (Mean, STDEV, T-values)"?

Thanks for your reply.

Best regards.

Kaouther

Hey Frank,

You right, the reported "T Statisics (|O/STERR|)" is your t-value.

Regards,

Chrisitan

[christian.nitzl]

Hey Kaouther,

That is also right!

Best regards,

Christian

[JELASSI]

Hi Christian,

Thanks for your reply.
I would like to ask you another question.
I want to calculate the variance inflation factor (VIF) for my formative indicators to test the multicollinearity but I don't know how? Can we do it in smartpls?

Thanks a lot.

Best regards

Kaouther

[christian.nitzl]

Hey Kaouther,

SmartPLS does not calculate the VIF in the current release.

But you can calculate the VIF with SPSS regression analysis. Take the manifest scores of the formative measurements as your independent variables and the latent scores of the focal variable as dependent variable. Both of them you can find in your SmartPLS output.

Best regards,

Christian

[JELASSI]

Hey Christian!

I followed your advice and I did it on spss. The results are very good!

Thank you very much.

Best regards.

Kaouther

[Bjerg54]

Hi Christian,

Regarding the t-test on the outer weights in a formative model, can these be used to test whether the indicator provides significanse support to its LV? I seem to have read in much litterature that the only way to test this in a formative model, is by a vanishing tetrad.

BR

Hello,

I am looking for the t-values of my formative indicators?

Is it the "T Statistics (|O/STERR|)" in the bootstrapping report under “outer weights (Mean, STDEV, T-values)"?

Thanks for your reply.

Best regards.

Kaouther

Hey Frank,

You right, the reported "T Statisics (|O/STERR|)" is your t-value.

Regards,

Chrisitan