forum.smartpls.com

Hi Bido,

Thank you for your comment.

It seems to me as though whether the coefficients should be termed small, moderate or large then to a large (!) extent comes down to my material and the context in which my results should be interpreted.

Regards,
Kristian

Hi, Chin (1998) Issues and Opinion on Structural Equation Modeling notes that "Standardized paths should be at least 0.20 and ideally above 0.30 in order to be considered meaningful", while indicating that coefficients of 0.10 are neither theoretically nor pratically interesting. Are there...

Hi needpls,

You find communality, AVE and R^2 in the report after running the PLS algorithm.

SQRT is square root.

Q2 is not Goodness of Fit.

According to Vinzi, Trinchera, Amato (2010), pp. 58-60, GoF then is:

GoF=√(Com×(R^2 ))=√(AVE×(R^2 ))

Com = average communality, R^2 is average R^2.

Yes, perhaps I just misunderstood what was meant with "path diagram" after performing a bootstrap. At least t-values are not the same as beta-coefficients.

-Kristian

Hengky,

Are the bootstrap coefficients really the beta-values? Are you sure it's not the results from the PLS algorithm that are the beta-coefficients?

-Kristian

Total effects are the sum of direct and indirect effects from one LV to another in the inner model (Albers 2009; Henseler, Ringle, Sinkovics 2009: 304). Using total effects you avoid the issue of direct effects, that are often underestimated and increasing the probability of a type 1 error - especia...

Hi Hengky, Thank you for your reply. I must have expressed myself somewhat unclear, however. What I find strange is this: Final Partition expresses each case's probability for belonging to the different segments. Each case's row-sum is thus 1.000. Each case should therefore belong to the segment to ...

Torben, Stone-Geisser’s ��2=1−������/������. SSE is the sum of squares of prediction errors, and SSO is sum of squares of observations. - &#55...

Whan you import a dataset/indicators, and there are missing values in the dataset, you tick the box that says the dataset contains missing values, and you specify the value that represents missing values. If you already have imported the dataset, doubleclick mobi_250.txt in Smart PLS, tick the box t...

If you use Case Wise Replacement the cases that have missing values will not be used in the analysis. You therefore lose a lot of information. The option could/should be called case wise deletion. Mean replacement might also produce some erroneous results. You can use som other software for dealing ...

I am testing the FIMIX application on a dataset with 7777 respondents. Runnig the FIMIX with k=3, Segment 1 is 19,8%, Segment 2 28,7% and Segment 3 51,5%. Assuming that Final Partition indicates each subject's probability for belonging to each of these segments, I copied the contents to Excel in ord...

Hi Alka, Casewise replacement is really ceswise deletion. If you choose this option, and one of your cases has a missing value, that case will be left out of the calculation. Mean replacement simply replaces missing values with the variable mean. If you want to deal with missing values in more sophi...

Hi,

Perhaps you will get some ideas from this post: viewtopic.php?t=812

Kristian

Hi, You have probably figured this out already, but in my experience - I also experienced issues with the downloaded user manual - if you watch the user manual from the smartpls.de web page it should be fine. Just click the different chapters of the manual on the bottom of the page, here, without do...

Imran, You should test your model with and without the moderator. Cohen's effect size - f2 - is calculated as follows: f2 = (R2_included - R2_excluded) / (1-R2_included) As Henseler, Ringle and Sinkovics (2009) writes, Cohen's effect sixe is calculated as the increase in R2 relative to the proportio...