Search found 116 matches

by janschreier
Tue Sep 01, 2015 8:20 pm
Forum: Requested Features
Topic: auto-color coefficients with negative weight in red
Replies: 0
Views: 3994

auto-color coefficients with negative weight in red

see title, would be nice to get an even fast overview on the model's results :)
by janschreier
Mon Aug 31, 2015 8:43 am
Forum: FAQ (Methodology)
Topic: Mode A for formative indicators
Replies: 2
Views: 2299

Re: Mode A for formative indicators

Thank you, Jan-Michael!
by janschreier
Sun Aug 30, 2015 9:56 am
Forum: SmartPLS Small Talk Corner
Topic: Set "notify me when a reply is posted" to yes by default
Replies: 0
Views: 3814

Set "notify me when a reply is posted" to yes by default

I would suggest putting this setting to yes by default. ATM is "no". I think this results in some comments initially being answered and once there is an additional questions this gets overlooked, which is frustrating for the person who asked and also for anybody who used the search engine (and I use...
by janschreier
Sat Aug 29, 2015 9:02 pm
Forum: FAQ (Methodology)
Topic: Bootstrap formative weight analysis
Replies: 2
Views: 2287

Re: Bootstrap formative weight analysis

Bootstrapping is used to get results about the variance, not the mean. You can have a high mean and still the data has high variance. You can see that also in your second image. the variances generated from bootstrapping are rather high. either the variance in your data is high or your sample size i...
by janschreier
Fri Aug 28, 2015 7:00 pm
Forum: FAQ (Methodology)
Topic: Mode A for formative indicators
Replies: 2
Views: 2299

Mode A for formative indicators

In Rigdon, E. E. (2012): „Rethinking Partial Least Squares Path Modeling: In Praise of Simple Methods“. In: Long Range Planning. 45 (5-6), S. 341–358. Rigdon's proposal is to calculate formative constructs with Mode A (correlations instead of regressions). My questions are now: 1) what do other expe...
by janschreier
Fri Aug 28, 2015 10:17 am
Forum: FAQ (Methodology)
Topic: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)
Replies: 20
Views: 15312

Re: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)

Hi Clemen, I would do it a bit different in 1.: run it for each segment 10 times and take that run which looks best in terms of R²s and model. This is a bit of looking for a needle in the haystack but you should run it more than once to reduce the risk of getting stuck in a local optimum (to reduce ...
by janschreier
Fri Aug 28, 2015 9:55 am
Forum: FAQ (Methodology)
Topic: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)
Replies: 20
Views: 15312

Re: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)

I think segments and groups are synonyms. What you refer to groups are group comparisons. Another paper worth reading is this one that describes an omnibus test to check for group differences and comparing multiple groups at once. Sarstedt, M.; Henseler, J.; Ringle, C. (2011): „Multi-group analysis ...
by janschreier
Thu Aug 27, 2015 1:30 pm
Forum: FAQ (Methodology)
Topic: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)
Replies: 20
Views: 15312

Re: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)

ooops, lol - that's why i sounded somewhat familiar. Time for a coffee break ;)
by janschreier
Thu Aug 27, 2015 1:25 pm
Forum: FAQ (Methodology)
Topic: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)
Replies: 20
Views: 15312

Re: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)

could you add the source for your citation?
by janschreier
Thu Aug 27, 2015 12:22 pm
Forum: FAQ (Methodology)
Topic: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)
Replies: 20
Views: 15312

Re: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)

Does each group fulfill all model criteria itself?

Apart from that I think what you do sounds reasonable. I'm not one of the experts only another student that tries to learn by helping others :)
by janschreier
Thu Aug 27, 2015 11:20 am
Forum: FAQ (Methodology)
Topic: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)
Replies: 20
Views: 15312

Re: Prediction-Oriented Segmentation in PLS-SEM (PLS-POS)

I found the process described here quite okay: Becker, J.-M.; Rai, A.; Ringle, C. M.; et al. (2013): „Discovering Unobserved Heterogeneity in Structural Equation Models to Avert Validity Threats“. In: MIS Quarterly. 37 , S. 665–694. start reading on 684 at the bottom. I don't think that there is a m...
by janschreier
Tue Aug 25, 2015 12:14 pm
Forum: Requested Features
Topic: q² effect strenght
Replies: 6
Views: 8895

Re: q² effect strenght

yes
by janschreier
Mon Aug 24, 2015 8:27 am
Forum: FAQ (Methodology)
Topic: How to generate predictive relevance q square values?
Replies: 3
Views: 2391

Re: How to generate predictive relevance q square values?

I asked for this feature here: viewtopic.php?f=3&t=3558

and here is some additional information here: viewtopic.php?f=12&t=1913&p=5633
by janschreier
Sat Aug 22, 2015 5:23 pm
Forum: FAQ (Methodology)
Topic: How to generate predictive relevance q square values?
Replies: 3
Views: 2391

Re: How to generate predictive relevance q square values?

unfortunately not. it's a tedious manual process