Search found 1284 matches
- Fri Oct 13, 2023 7:11 am
- Forum: Method and application
- Topic: p-value adjustment: 0.05 to 0.0169524 when executing the permutation test
- Replies: 5
- Views: 1881
Re: p-value adjustment: 0.05 to 0.0169524 when executing the permutation test
To be honest, I don't understand what you mean with "the significance level automatically changes from 0.017 to 0.02." If you run a (preferably permutation-based MGA), you will get a p-value for the difference in parameter between groups. You usually conclude that the difference is signifi...
- Fri Oct 13, 2023 6:56 am
- Forum: Method and application
- Topic: MGA for more than two groups
- Replies: 4
- Views: 1817
Re: MGA for more than two groups
I think this would be the perfect reference for you: https://www.sciencedirect.com/science/article/pii/S0148296322010049 Cheah, J. H., Amaro, S., & Roldán, J. L. (2023). Multigroup analysis of more than two groups in PLS-SEM: A review, illustration, and recommendations. Journal of Business Resea...
- Tue Oct 10, 2023 7:50 am
- Forum: Method and application
- Topic: Difference between control variables and moderating variables
- Replies: 1
- Views: 1499
Re: Difference between control variables and moderating variables
A moderator variable Z influence the strength of a relationship between two variables X and Y. The estimated effect of X on Y is conditional on Z (being equal to zero). For example, income may be a moderator of the relationship between satisfaction and loyalty: with higher (lower) income the relatio...
- Tue Oct 10, 2023 7:36 am
- Forum: Method and application
- Topic: Difference between MICOM and MGA results
- Replies: 1
- Views: 1471
Re: Difference between MICOM and MGA results
As the name says, full measurement invariance in MICOM refers to the measurement model; not the structural model.
You have invariance in construct composition (basically weights), construct means and construct variances. This does not imply that you have equal path relations.
You have invariance in construct composition (basically weights), construct means and construct variances. This does not imply that you have equal path relations.
- Mon Sep 18, 2023 7:26 am
- Forum: Method and application
- Topic: number of Indicator of latent variable (Reflective model)
- Replies: 2
- Views: 1534
Re: number of Indicator of latent variable (Reflective model)
The minimum number is 1. This would be a single indicator constructs, which means that the construct will equal the indicator. In general, you often read that you need 3 to 4 (reflective) indicators, but this is not a technical requirement for PLS. It is more a recommendation in terms of psychometri...
- Thu Sep 14, 2023 11:57 am
- Forum: SmartPLS 4 - Questions and answers
- Topic: Sample Selection Biase and Gaussian Copulas
- Replies: 1
- Views: 1849
Re: Sample Selection Biase and Gaussian Copulas
1) Selection bias is nothing that can be checked with a PLS analysis or similar. The only thing that you can do is comparing your sample characteristics to some other (outside) information about the population to show that you have an adequate sample. 2) If you use Gaussian copulas you should ad the...
- Tue Sep 12, 2023 10:33 am
- Forum: Method and application
- Topic: How to Deal With MICOM Result — Stage 2: compositional invariance not established
- Replies: 5
- Views: 1908
Re: How to Deal With MICOM Result — Stage 2: compositional invariance not established
What you refer to is more configural invariance, that you have identical indicators (e.g., the same questions), construct orientation (e.g., reflective vs. formative), data treatment, etc. This is a prerequisite for all later stages of invariance. Compositional invariance should establish that there...
- Mon Sep 11, 2023 6:41 am
- Forum: Method and application
- Topic: How to Deal With MICOM Result — Stage 2: compositional invariance not established
- Replies: 5
- Views: 1908
Re: How to Deal With MICOM Result — Stage 2: compositional invariance not established
If compositional invarience not established, you cannot compare the groups (the group-specific estimates such as path coefficients), because it is not assured that the construct are measured in the same way. Therefore, differences in the path coefficients, for example, could be the reason of measuri...
- Fri Sep 08, 2023 9:41 am
- Forum: SmartPLS 4 - Questions and answers
- Topic: Latent Growth Curve Model
- Replies: 2
- Views: 1820
Re: Latent Growth Curve Model
The latest release SmartPLS 4.0.9.6 allows you to estimate covariance structural equation models including the mean structure and not only the covariance structure. With this it should be able to estimate most latent growth models that other SEM software can also handle. You can for example use the ...
- Sat Sep 02, 2023 7:54 pm
- Forum: SmartPLS 4 - Questions and answers
- Topic: formative third order construct
- Replies: 3
- Views: 1877
Re: formative third order construct
If you want to do the disjoint approach you need to relate the lower-order constructs to your antecedents and outcomes of the higher-order construct as is described in the papers that you mention.
- Mon Aug 28, 2023 2:58 pm
- Forum: Method and application
- Topic: categorical IV and q-squared results
- Replies: 2
- Views: 1573
Re: categorical IV and q-squared results
Yes, if you have a binary IV the predictive results are not correct for standardized results. We have fixed the issue and it will be correct with the next release. The unstandardized results are correct.
- Thu Aug 24, 2023 9:25 am
- Forum: Method and application
- Topic: Sample size: minimum R2 method
- Replies: 6
- Views: 15962
Re: Sample size: minimum R2 method
I think the difference lies exactly in the difference between "Linear multiple regression: Fixed Model, R2 deviation from zero" and "Linear multiple regression: Fixed Model, single regression coefficient." The first is the power of a regression models' F-test to find that the R² ...
- Wed Aug 23, 2023 6:48 pm
- Forum: Method and application
- Topic: Missing values
- Replies: 1
- Views: 1470
Re: Missing values
You can also include them and use pairwise deletion as the missing value treatment. This way every relationship that has only users is estimated with the data of users only. If you have data for both the relationship is estimated for both.
- Mon Aug 14, 2023 9:22 am
- Forum: SmartPLS 4 - Questions and answers
- Topic: Boostrapping for Cronbach's Alpha and ...
- Replies: 1
- Views: 1886
Re: Boostrapping for Cronbach's Alpha and ...
1) You also get the confidence intervals with your bootstrapping results. If you choose "Amount of results: Complete" you will also get bootstrapping results for Cronbachs alpha etc. 2) If the reliability statistics such as Cronbachs alpha are very high, it could (but must not) imply that ...
- Sat Aug 12, 2023 9:13 am
- Forum: Method and application
- Topic: Higher-order model: explained variance
- Replies: 3
- Views: 1626
Re: Higher-order model: explained variance
1) You need to have the same sign for coefficient and correlation, so that after multiplication everything is always positive. 2) Can I figure out how much all the predictors together contribute to forming my third-order construct? All predictors together, i.e., the sum of products (path coefficient...