Search found 13 matches

by FHH
Mon Dec 15, 2014 3:43 pm
Forum: Method and application
Topic: Partial derivate plots of interactions
Replies: 8
Views: 8242

Re: Partial derivate plots of interactions

[Using Smart PLS 2.0M3] I tried to plot a simple slope graph for a one-way interaction (1 DV, 1 IV, 1 Moderator). Could anyone comment on the correctness of my approach? 1) I built the model in SmartPLS 2) I ran the PLS algorithm 3) I opened the report and copied the 'Latent Variable Scores (unstand...
by FHH
Sun Dec 14, 2014 6:42 pm
Forum: Method and application
Topic: Partial derivate plots of interactions
Replies: 8
Views: 8242

Re: Partial derivate plots of interactions

Is it possible to plot Simple Slopes in SmartPLS3?
by FHH
Fri Oct 03, 2014 3:37 pm
Forum: Method and application
Topic: R square values
Replies: 10
Views: 27070

Re: R square values

Hey Hengkov!
I am using SmartPLS 2. Should I use LV scores (Report > PLS > Calculation results) to test collinearity ALTHOUGH the LV scores are negative in my case (due to negative outer weights)?
by FHH
Fri Oct 03, 2014 3:34 pm
Forum: Method and application
Topic: convergent and divergent validity
Replies: 12
Views: 17034

Re: convergent and divergent validity

Thank you for sharing your thoughts!

* Could you give me an example about what effects of the nomological network cause such a change in weights?
* The negative weights result in negative LV scores. Can I still interpret the LV scores, or are negative LV scores of no use anymore?

Cheers!
by FHH
Fri Oct 03, 2014 3:19 pm
Forum: Method and application
Topic: Which data for VIF calculation
Replies: 8
Views: 7987

Re: Which data for VIF calculation

In the PLS report I get negative LV scores!

Can I still use these scores to test for collinearity among the LVs using VIF?
Or are all conclusions drawn from negative LVs scores wrong?
Is there a way to tranform/standardize the negative LV scores to make them interpretable?
by FHH
Fri Oct 03, 2014 2:11 pm
Forum: Method and application
Topic: R square values
Replies: 10
Views: 27070

Re: R square values

1. Check collinearity among the constructs
How exactly is this done? Do I use the LV scores? Standardized or standardized?
by FHH
Fri Oct 03, 2014 1:49 pm
Forum: Method and application
Topic: convergent and divergent validity
Replies: 12
Views: 17034

Re: convergent and divergent validity

Hey everyone! the following issue has been neglected in the forum discussions so far. * I am writing a scale development paper * For this paper I followed the known scale development process to simultaneously develop 7 formative constructs . Each construct represents a marketing tool. Each construct...
by FHH
Sun Sep 28, 2014 5:33 am
Forum: Method and application
Topic: weights higher than 1
Replies: 16
Views: 18866

Re: weights higher than 1

For clarification: * Should I use the unstandardised or standardized LV scores? * I do have negative weights in some of my formative constructs? Will this not effect the LV scores and lead to "false" results? * To avoid negative LV scores caused by negative outer weights, should I use Mode...
by FHH
Fri Sep 26, 2014 2:20 pm
Forum: Method and application
Topic: weights higher than 1
Replies: 16
Views: 18866

Re: weights higher than 1

Hey everyone! I could really need some help! What are the steps in SPSS to test for a possible suppression effect in the formative indicators ? "Negative weights are the result of the pattern of correlations among the formatively measured construct indicators. Suppression occurs when an indicat...
by FHH
Sat May 03, 2014 8:26 am
Forum: Method and application
Topic: Why do outer loadings change when using another DV?
Replies: 0
Views: 2925

Why do outer loadings change when using another DV?

Situation: a) I have one independent and one dependent variable. Both are reflective. All outer loadings >0.7, AVEs >0.7, Cronbach's a >0.9 http://www.imageurlhost.com/images/p8mt6snfj9u20ogn8uu_Situation-A.jpg b) Now, I change the dependent variable with another dependet variable. All quality crite...
by FHH
Sun Mar 23, 2014 6:54 pm
Forum: Method and application
Topic: Strange LV Index values
Replies: 1
Views: 2034

Strange LV Index values

Hi everyone! Can anyone give me advice on the following issue? Background information: I have several formative latent variable (LV) to measure the application of marketing tools. Each formative variable has three indicators on a 7-point Likert scale. I would like to analyze the index value as a fir...
by FHH
Sun Mar 23, 2014 4:16 pm
Forum: Method and application
Topic: PLS-MGA: How to dichotomize/split latent variable
Replies: 7
Views: 6136

Can anyone help?
by FHH
Thu Mar 20, 2014 3:00 pm
Forum: Method and application
Topic: PLS-MGA: How to dichotomize/split latent variable
Replies: 7
Views: 6136

PLS-MGA: How to dichotomize/split latent variable

Hi everyone! [*] I have a latent, formative construct ("technological complexity") with 4 indicators. [*] I would like to compare two groups of data; i.e. "high" techn. complexity vs. "low" techn. complexity. [*] I am thinking about a median split. a) Which steps are ne...