Search found 1284 matches
- Tue Feb 17, 2015 9:49 pm
- Forum: Method and application
- Topic: Partial derivate plots of interactions
- Replies: 8
- Views: 8286
Re: Partial derivate plots of interactions
Simple Slope Interaction Plots for the standardized coefficients are now part of SmartPLS. It is a new feature of version 3.2.0.
- Tue Feb 17, 2015 1:50 pm
- Forum: Method and application
- Topic: PLS and Genetic algorithm Segmentation (PLS-GAS)
- Replies: 11
- Views: 10078
Re: PLS and Genetic algorithm Segmentation (PLS-GAS)
In our MISQ paper, we describe a process how to use and how to identify the best number of segments in PLS-POS. Becker, J.-M./ Rai, A/, Ringle, C.M./ Völckner, F.: Discovering Unobserved Heterogeneity in Structural Equation Models to Avert Validity Threats, in: MIS Quarterly (MISQ), Volume 37 (2013)...
- Tue Feb 17, 2015 1:49 pm
- Forum: SmartPLS 3 - Small Talk Corner
- Topic: Heterotrait-monotrait (HTMT) ratio of Correlations
- Replies: 3
- Views: 6916
Re: Heterotrait-monotrait (HTMT) ratio of Correlations
Hi,
there is a paper in JAMS about this new criterion. You will find Information about cutsoffs and its usage there:
Henseler, J., Sarstedt, M., Ringle, C. M. 2014 "A New Criterion for Assessing Discriminant Validity in Variance-Based Structural Equation Modeling".
there is a paper in JAMS about this new criterion. You will find Information about cutsoffs and its usage there:
Henseler, J., Sarstedt, M., Ringle, C. M. 2014 "A New Criterion for Assessing Discriminant Validity in Variance-Based Structural Equation Modeling".
- Mon Feb 02, 2015 10:23 pm
- Forum: Method and application
- Topic: Significance of indirect effect in multiple mediator models
- Replies: 10
- Views: 13492
Re: Significance of indirect effect in multiple mediator mod
Actually I have quickly illustrated what happens if you use the individual sign change option in a case, where the true population coefficient is zero (so that there is actually no relation between the two variables). The original model coefficient (in this case -0.028) is only a result of sampling ...
- Mon Feb 02, 2015 9:52 pm
- Forum: Method and application
- Topic: Significance of indirect effect in multiple mediator models
- Replies: 10
- Views: 13492
Re: Significance of indirect effect in multiple mediator mod
The problem is that you are using the individual sign-change option, right? Hence, your direct effect estimates are all set to the sign of the original model even if they are in opposite direction. Unfortunately, this is not done for the indirect effects. I will look into this issue. However, given ...
- Mon Feb 02, 2015 4:19 pm
- Forum: Method and application
- Topic: Significance of indirect effect in multiple mediator models
- Replies: 10
- Views: 13492
Re: Significance of indirect effect in multiple mediator mod
Hi Sam, At the moment you can only calculate these specific indirect effects by hand. Therefore, you have to use the data that is provided under "Path Coefficients > Samples". There you find for each bootstrap sample the path coefficients of your model. You can then calculate the indirect ...
- Thu Jan 29, 2015 4:05 pm
- Forum: Method and application
- Topic: Interaction Effect Term and Regression Handling
- Replies: 7
- Views: 8300
Re: Interaction Effect Term and Regression Handling
Hi Michael, Thank you for thoroughly investigating this issue. These options have never really been the focus of academic papers and hence there is some room for improvement. I have looked into your issue and indeed there is a small calculation bug for unstandardized interaction coefficients. The al...
- Tue Jan 27, 2015 8:08 pm
- Forum: Method and application
- Topic: help with 2nd order construct
- Replies: 5
- Views: 8336
Re: help with 2nd order construct
How do you implement the two-stage approach? Did you use the latent variable scores of all constructs from the first stage in the second stage or only for the HOC? If you have collinearity in your formative measures then the weights are usually quite unstable in the bootstrapping: collinearity cause...
- Tue Jan 27, 2015 10:55 am
- Forum: SmartPLS 3 - Requested Features
- Topic: Collinearity Statistic
- Replies: 7
- Views: 26585
Re: Collinearity Statistic
VIF statistics are important for the inner model and formative outer (measurement) models. If you have only reflective measures, then using only the inner model VIFs is ok. If you also have formative measures, then you should assess the VIFs of the indicators of the formative measures as well. Gener...
- Mon Jan 26, 2015 8:17 am
- Forum: Method and application
- Topic: Retaining insignificant indicator - reflective model?
- Replies: 3
- Views: 3034
Re: Retaining insignificant indicator - reflective model?
If the loadings of the indicator range from -1 to +1, then your indicator simply seems to be not a reliable measurement instrument for your construct.
- Sat Jan 24, 2015 3:44 pm
- Forum: Method and application
- Topic: Retaining insignificant indicator - reflective model?
- Replies: 3
- Views: 3034
Re: Retaining insignificant indicator - reflective model?
Retaining an insignificant indicator has pros and cons. Yet, I would first explore what the problem could be by looking at all the samples in the bootstrapping procedure to get more insights on the problem. In general, it sounds strange that such a loading (0.675) is insignificant. Are the loadings ...
- Wed Jan 21, 2015 12:19 pm
- Forum: Method and application
- Topic: Interaction Effect Term and Regression Handling
- Replies: 7
- Views: 8300
Re: Interaction Effect Term and Regression Handling
First, I would write “not recommended” also in the rightmost column of your table. Second, the choice of product-term handling is not well explored in the literature on PLS. The basic problem is that simulated data is usually already standardized. Hence, it is not much known about whether mean-cente...
- Tue Jan 20, 2015 5:53 pm
- Forum: Method and application
- Topic: Interaction Effect Term and Regression Handling
- Replies: 7
- Views: 8300
Re: Interaction Effect Term and Regression Handling
Hi Michael, we have recently discussed this implementation and might change it in the near future to avoid misuse. In some rather specific cases all combinations make sense to some degree. That’s why we offered all possible combinations. Ultimately, the interpretation is different between the approa...
- Mon Jan 19, 2015 10:07 am
- Forum: Method and application
- Topic: BiasCorrected Bootstrapping, p-values & ConfidenceIntervals
- Replies: 2
- Views: 2955
Re: BiasCorrected Bootstrapping, p-values & ConfidenceInterv
There are different assumptions underlying P-Values and Confidence Intervals. They are not the same and must not give the same result, especially in the case of biascorrected CI.
You have to decide which one is more appropriate for you and then choose the correct one.
You have to decide which one is more appropriate for you and then choose the correct one.
- Wed Jan 07, 2015 9:45 pm
- Forum: SmartPLS 3 - Requested Features
- Topic: weighted data
- Replies: 16
- Views: 19559
Re: weighted data
Yes, it is one of the new features of 3.1.7.
There is no published paper yet, but we are working on it and there should be one in the near future (at least a conference paper).
There is no published paper yet, but we are working on it and there should be one in the near future (at least a conference paper).