forum.smartpls.com

First, for a formative construct in PLS, regardless of it being higher-order or not, you will always have an r-squared of 1 (or at least very close to one). This is, because the indicators define (compose) the construct and thus they should and will explain all the variance of the construct because ...

I have updated the link. It was moved to a different section of the website during some restructuring process.

In Becker et al. (2023) we have a section on higher-order constructs where we summarize the latest advice on this topic. In addition, Sarstedt et al. (2019) is also a quite good an comprehensive source on this topic. Becker, J.-M., Cheah, J.-H., Gholamzade, R., Ringle, C.M. and Sarstedt, M. (2023), ...

It was one of the first approaches that was proposed and it is also used in CB-SEM sometimes. However, latest PLS literature recommends the two-stage because it has been show in simulation studies that it is the overall best approach in a variety of models. To quote Becker et al. (2023) p.334: "...

Why do you want to use the product-indicator approach? According to the literature the two-stage approach is almost always preferable over the product-indicator approach in PLS. Anyways you can also create the product indicators yourself outside of SmartPLS and import a datafile if you need them. Ho...

If you specify the same model, you should get the same results. Otherwise, please contact our support (https://www.smartpls.com/support) and send the Amos and SmartPLS project files.

BTW: This is a really good paper on why dichotomization into significant and not significant is not a good idea: https://doi.org/10.1016/j.jbusres.2023.114086 Rigdon, E. E. (2023). How improper dichotomization and the misrepresentation of uncertainty undermine social science research. Journal of Bus...

Significance is a strange concept that is also often criticized because of its dichotomous nature. Yes/no In laymen terms (and not 100% correct): What you actually want to know is whether your effect is strong enough given the sampling uncertainty that you have to conclude that is is likely differen...

If your categorical / grouping variable is missing in the generated data file, you need to select it under "Other" to have it also available in the generated data file.

If the problem persist, please contact our technical support: https://www.smartpls.com/support

In general, the PLS algorithm assumes metric or quasi metric data. If you do not have such data for all indicators, it does not make much sense to combine them into a construct, because the constructs will also have a metric representation. If you think that your ordinal scale satisfies quasi metric...

Thank you for the suggestion. In the next release there will be an option "Generate results per sample" in Bootstrapping and Permutation. Without this option we will not save all the results, but only the overview statistics.

Single indicator construct have a fixed weight and loading of 1. As Christian has said previously: You can ignore the outer relationship of the single indicator construct. The latent variable and the indicator are identical. Thus, there is no variance and therefore also no t or p-values to calculate...

The groups do not need to be equal. However, it can cause problems if the groups are very imbalanced and one group is extremely small (<=10%).
In addition, if you do MGA, your smallest group should still fulfill the minimum sample size requirements.

Yes. Single indicator construct have a fixed weight and loading of 1. Thus, there is no variance and therefore also no t or p-values to calculate. The t and p-value are just not defined and therefore reporting them is also not meaningful. We have updated this in the recent version of SmartPLS to bet...

Yes, based on the literature, the BIC and the and GM (Geweke–Meese criterion) are best for model selection and all other criteria are not well suited. SmartPLS does not include the Geweke–Meese criterion (GM), which is based on the model-complexity adjusted mean square error (MSE) from a saturated (...