Search found 1018 matches

by jmbecker
Mon Jul 29, 2019 4:35 pm
Forum: FAQ (Methodology)
Topic: MGA Significance
Replies: 1
Views: 72

Re: MGA Significance

1) You may simply have insufficient power in both groups because you only have the half sample size. Especially if the effects are quite similar in both groups power is the most likely reason. 2) Again you may have power problems in one of your groups that results in insignificant effects. In additi...
by jmbecker
Mon Jul 29, 2019 4:23 pm
Forum: FAQ (Methodology)
Topic: Combining Continuous and Likert-Scale constructs
Replies: 1
Views: 60

Re: Combining Continuous and Likert-Scale constructs

Generally, combining different variables with different scales is not a problem in PLS because of the routine standardization of input data. However, PLS as does regression analysis or any other correlational method assumes metric data. Most people assume that their Likert data is at least quasi-met...
by jmbecker
Mon Jul 29, 2019 3:49 pm
Forum: FAQ (Methodology)
Topic: IPMA for model having Second-order construct
Replies: 1
Views: 56

Re: IPMA for model having Second-order construct

You have two options when you want to use the IPMA for a second-order construct. 1) You use the repeated-indicator approach. That should work just fine. 2) You could use the disjoint two-stage approach and take the necessary latent variable scores as unstandardized LV scores from an IPMA analysis on...
by jmbecker
Mon Jul 29, 2019 3:45 pm
Forum: FAQ (Methodology)
Topic: Set bootstrap confidence intervals in MGA for two tailed test and bonferroni correction
Replies: 1
Views: 47

Re: Set bootstrap confidence intervals in MGA for two tailed test and bonferroni correction

The Bonferroni correction should be the same regardless of whether you take 5% devided by 3 and then devide by 2 (for both ends) or devide 5% by 2 (for both ends) and then devide by 3. In any case it should be 0.008333333 for the lower end and 99.99166667 for the upper end. The PLS-MGA output is not...
by jmbecker
Mon Jul 29, 2019 3:32 pm
Forum: FAQ (Methodology)
Topic: Second order moderating model
Replies: 6
Views: 121

Re: Second order moderating model

I would actually prefer model 3 using a disjoint two-stage approach for the second-order model combined with a two-stage approach for the moderating effects. You should also read: https://www.researchgate.net/publication/333877607_How_to_specify_estimate_and_validate_higher-order_constructs_in_PLS-S...
by jmbecker
Mon Jul 29, 2019 3:27 pm
Forum: FAQ (Methodology)
Topic: Bootstrapping R2
Replies: 1
Views: 55

Re: Bootstrapping R2

It is rarely reported in PLS, but an F test on the R² is quite common in regression analysis. This is very similar owing to the fact that PLS is nonparameteric and therefore F test is not viable. And even low R² can be significant if it has small variability. Maybe even more interesting than the sig...
by jmbecker
Mon Jul 29, 2019 3:22 pm
Forum: FAQ (Methodology)
Topic: Selected indicators are not shown
Replies: 3
Views: 124

Re: Selected indicators are not shown

Most common problem is that people have selected a filter in the indicators list. There are three options: 1) Show all, 2) Show only used, 3) Show only unused indicators.

Sometimes variable names also contain special characters that prevent display in the indicators list or model.
by jmbecker
Mon Jul 29, 2019 3:17 pm
Forum: FAQ (Methodology)
Topic: Getting N/A error while doing a consistent PLS Bootstrapping for moderation
Replies: 1
Views: 49

Re: Getting N/A error while doing a consistent PLS Bootstrapping for moderation

Please search the forum there are plenty of answers for this problem.
by jmbecker
Mon Jul 15, 2019 9:55 am
Forum: FAQ (Methodology)
Topic: Hierarchical Component Model
Replies: 2
Views: 57

Re: Hierarchical Component Model

I think you should read the following new paper that clarifies how to specify and validate second-order constructs:
https://www.researchgate.net/publicatio ... in_PLS-SEM
by jmbecker
Fri Jul 12, 2019 1:15 pm
Forum: FAQ (Methodology)
Topic: Adding a control variable or moderating variable?
Replies: 1
Views: 63

Re: Adding a control variable or moderating variable?

Treating Ethnicity as a control: 1) Do not add all the dummies onto one construct but model them separately as single-item constructs. Adding them to a construct doesn't make much sense. The measurement model statistics are also not meaningful in such a case. Also be aware that if you create dummy v...
by jmbecker
Fri Jul 12, 2019 12:17 pm
Forum: FAQ (Methodology)
Topic: Konsistentes Bootstrapping - P & T-Werte 0,000
Replies: 2
Views: 69

Re: Konsistentes Bootstrapping - P & T-Werte 0,000

1) Please send your data and model to our support. They might look at the problem. https://support.smartpls.com

2) Please also read this post: viewtopic.php?f=5&t=25181&p=43515
by jmbecker
Fri Jul 12, 2019 12:11 pm
Forum: FAQ (SmartPLS 2.0)
Topic: Activation key not working?
Replies: 16
Views: 6835

Re: Activation key not working?

If you have a problem with activiation please contact our support: https://support.smartpls.com/
by jmbecker
Fri Jul 12, 2019 12:10 pm
Forum: SmartPLS Small Talk Corner
Topic: Singular Matrix Problem
Replies: 4
Views: 168

Re: Singular Matrix Problem

The "indicators" is another word for the items/manifest variables that you assign to a construct as measurement model. Thus, they are the variables in your dataset. A common problem with singular matrix in MGA is fact that people include the grouping variable in their model (e.g., gender). This vari...
by jmbecker
Fri Jul 12, 2019 12:06 pm
Forum: FAQ (Methodology)
Topic: Can I use the unstandardised latent variable scores as they are provided in the IPMA for hypothesis testing (t-test)?
Replies: 1
Views: 35

Re: Can I use the unstandardised latent variable scores as they are provided in the IPMA for hypothesis testing (t-test)

First, yes you should check for measurement invariance. Second, when you test for measurement invariance, in step 3 you get a permutation estimate of the mean differences which is similar to the t-test that you want to calculate. Hence, you might just use this (assuming that you have measurement inv...