Search found 248 matches

by christian.nitzl
Wed May 08, 2013 10:18 am
Forum: Method and application
Topic: Mediation: Direct path gets stronger through mediation?
Replies: 3
Views: 2854

…when using reflective measurements you control the measurement error. In contrast, for formative measurements you don’t control this. This could be one reason for the differences. There are many articles who discussing the differences between reflective and formative measurements. You can find also...
by christian.nitzl
Wed May 08, 2013 10:07 am
Forum: Method and application
Topic: Two groups with different numbers of indicators
Replies: 6
Views: 4064

Nevertheless, it is important to pay attention to measurement invariance. That means, that at least you have to measure your latent constructs with the same indicators for the different groups. Please check this forum for the topic of measurement invariance and you will find some important hints. Gr...
by christian.nitzl
Fri Apr 26, 2013 9:26 am
Forum: Method and application
Topic: Mediation: Direct path gets stronger through mediation?
Replies: 3
Views: 2854

…seems to me that it is a correlation problem between your latent constructs. In a regression model a coefficient can be also become higher when you introduce a new variable. This is because a high correlation between the variables. Therefore we test VIF (variance inflation factor). I ask myself how...
by christian.nitzl
Tue Apr 23, 2013 12:39 pm
Forum: Method and application
Topic: Mediation-R square Reduced
Replies: 4
Views: 3647

Hi Frederic, Let me explain this on an example: Your direct effect is: x->y (with path a) Your mediation effect is: x->m->y (with path b and c) You check first the direct effect separately (a). Afterwards you integrate the both other paths b and c. Now you can check how the path c is changing by thi...
by christian.nitzl
Thu Apr 18, 2013 2:26 pm
Forum: Method and application
Topic: Interpretation of moderating effects
Replies: 1
Views: 1657

Hi, No, you do not interpret the direct effects after the integration of the interaction variable. See for that my paper on page 435 (sorry only in German) ;-) Nitzl, C./Schloderer, M. (2011): Die Prüfung von Interaktionseffekten in Partial Least Squares (PLS)-Modellen – Eine Fallstudie zur Nutzung ...
by christian.nitzl
Thu Apr 18, 2013 2:19 pm
Forum: Method and application
Topic: Mediation-R square Reduced
Replies: 4
Views: 3647

Hi, This is because of the algorithms of PLS which optimize the variance of dependent variable in your model at the same time. In contrast covariance based method optimizes the correlation matrix. Therefore, when you connect the mediation variables, PLS optimize the R^2 now of the mediator variable ...
by christian.nitzl
Sat Mar 02, 2013 11:53 am
Forum: Method and application
Topic: Moderation effect size
Replies: 2
Views: 2425

Hi,

When you calculate the effect size for a moderating effect you just delete the effect of the IV x Moderator path.

Greetings,

Christian
by christian.nitzl
Thu Jan 24, 2013 8:56 pm
Forum: Method and application
Topic: Interaction effects
Replies: 89
Views: 84009

Sure! Please send me a message to my mail address christian.nitzl@unibw.de. Then I can send you the Excel template.

Best
Christian
by christian.nitzl
Sun Nov 11, 2012 4:44 pm
Forum: Method and application
Topic: Bootstrap sample size in Henseler's PLS-MGA
Replies: 7
Views: 6725

You can get the excel sheet directly from Jörg Henseler when you send him an email to:

joerg.henseler@pls-school.com

Cheers,
Christian
by christian.nitzl
Sat Oct 06, 2012 3:47 pm
Forum: SmartPLS 2 - FAQ
Topic: second order
Replies: 24
Views: 17582

Hi Reinhard, This is really an interesting discussion. Because of the multidimensional measurement in my case it is indirectly a formative measured construct. In the first order I use reflektiv measurements. It is also a mediation variable in my model. As far as I know the main problem is that the w...
by christian.nitzl
Mon Oct 01, 2012 9:23 pm
Forum: Method and application
Topic: Zhao et al (2010) mediation approach in SmartPLS
Replies: 1
Views: 2654

Hi Dennis! Ad 1) In my opinion it is the best approach to calculate the mediation effect in SmartPLS actually. Ad 2) In the case of a partly mediation of an effect there have to be an effect on R2 normally through the integration of an additional predictor. I could imagine there are some cases witho...
by christian.nitzl
Fri Sep 14, 2012 9:21 am
Forum: Method and application
Topic: Calculating T-Value for indirect effects: a*b/sd(ai*bi) ?
Replies: 18
Views: 19415

…that 43% of the effect from an independent variable on a depended variable are mediated by the mediator variable.
Greetings,
Christian
by christian.nitzl
Tue Sep 04, 2012 12:51 pm
Forum: Method and application
Topic: Using a binary variable (a single-item construct) as DV
Replies: 4
Views: 8803

Hi, Hair et al. (2012) wrote about this on page 421: “Several researchers stress that PLS-SEM generally works with nominal, ordinal, interval, and ratio scaled variables (e.g., Fornell and Bookstein 1982; Haenlein and Kaplan 2004; Reinartz et al. 2009). It is therefore not surprising that researcher...
by christian.nitzl
Sat Sep 01, 2012 9:59 am
Forum: SmartPLS 2 - FAQ
Topic: FIMIX and bootstrapping
Replies: 6
Views: 5713

Hi Tatjana, Did this message really appear when you run the bootstrap procedure or did the message also appear when you run the normal PLS algorithms? Did you try different algorithms’? Normally there is no technical problem to use single items in PLS. I am already calculating a model in PLS with on...
by christian.nitzl
Sat Aug 18, 2012 11:41 am
Forum: Method and application
Topic: Preditive relevance q-squared
Replies: 11
Views: 14069

The suggested procedure is correct!

When you can read German, please check my working paper for more information about Q^2.

Greetings,

Christian