Page 1 of 1

IPMA Performace of constructor

Posted: Tue May 03, 2016 8:00 pm
by juliosolano
Mr.
When I use IPMA, How performance is calculated from one Construct? , if I have 2 or more items or indices?

Thanks

Julio

Re: IPMA Performace of constructor

Posted: Tue May 03, 2016 8:14 pm
by jmbecker
The Performance is the mean of the rescaled latent variable score values. Rescaling is done by tranforming the original scale of each indicator into a 0 to 100 scale. Unstandardized weights are then used to calculate the rescaled latent variable scores as weighted averages of the rescaled indicators.

Re: IPMA Performace of constructor

Posted: Tue May 03, 2016 8:26 pm
by juliosolano
Ok thanks, but I can not understand how the value is calculated complete. That is, once I have solved model (Bootstrapping) then I have several values, like p value, t student, coeficient, etc. Then what values are involved in the calculation of performance?.

I am sorry my ignorance, but can send me a simple complete example where I can see this calculation. Thank you.

Julio

Re: IPMA Performace of constructor

Posted: Wed May 04, 2016 8:25 am
by jmbecker
Don't use bootstrapping. I think we are talking about IPMA.
In the IPMA you will find under "Outer Weights" the "Rescaled" outer weights. Use these to calculate linear combinations of the indicators of each construct.
For this purpose use the "Indicator Data (Rescaled)", which will give you the rescaled indicator data or you could also use the original unstandardized data "Indicator Data (Original)".
Depending on which data you use you will either get the unstandardized or rescaled latent variable scores (you will find both of these in the IPMA under "latent variables").
The mean of these latent variable scores is your index value (original unstandardized data) or your performance value (rescaled data).

Thus, you will have all necessary information in your IPMA results.

If you click on "Reade more" when starting the algoritm, we also provide some useful literatur on this topic which explains the steps in more detail:
Höck, C., Ringle, C. M., and Sarstedt, M. 2010. Management of Multi-Purpose Stadiums: Importance and Performance Measurement of Service Interfaces. International Journal of Services Technology and Management, 14(2/3): 188-207.
Rigdon, E. E., Ringle, C. M., Sarstedt, M., and Gudergan, S. P. 2011. Assessing Heterogeneity in Customer Satisfaction Studies: Across Industry Similarities and Within Industry Differences. Advances in International Marketing, 22: 169-194.
Schloderer, M. P., Sarstedt, M., and Ringle, C. M. 2014. The Relevance of Reputation in the Nonprofit Sector: The Moderating Effect of Socio-Demographic Characteristics. International Journal of Nonprofit and Voluntary Sector Marketing, 19(2): 110-126.
Völckner, F., Sattler, H., Hennig-Thurau, T., and Ringle, C. M. 2010. The Role of Parent Brand Quality for Service Brand Extension Success. Journal of Service Research, 13(4): 359-361. Link to Article


I hope this helps.

Re: IPMA Performace of constructor

Posted: Wed May 04, 2016 3:14 pm
by juliosolano
Dr. Jan-Michael Becker

Thank you very much, the truth is I clearer from where the calculation of performance. The artiuclos, especially the second is very clear.

Regards

Julio

Re: IPMA Performace of constructor

Posted: Wed May 04, 2016 6:42 pm
by juliosolano
Mr.

What is the diference of Indicator Data Standardized vs Indicator Data Original ?

Julio

Re: IPMA Performace of constructor

Posted: Fri May 06, 2016 8:06 am
by jmbecker
"Original" is the data in the original scale as in your data set.
"Standardized" is the original data standardized.

Re: IPMA Performace of constructor

Posted: Wed Jul 13, 2016 9:26 am
by JGylden
Dear users,

I have a related question.

I'm writing my master's thesis about loyalty creation in the news media business. One of my research objectives is to measure the impact of several news brandpersonality dimensions (Sincerity, Ruggedness, Trustworthiness etc.) on reader loyalty by measuring their congruity with the news media along those dimensions.

Intuitively I think it makes sense to define Brandpersonality as a Second-order construct in a reflective-formative model, where the underlying dimensions are measured reflectively. But the most valid way of setting up that kind of model in SmartPLS is by use of the repeated indicator approach (Mode B as described in Becker et al, 2012).

My suspicion is, whether the weights from each of the lower-order constructs on the higher-order construct (Brandpersonality) becomes skewed in some kind of way.

Thus, my question is: How do I measure the total effect of each Lower-order construct on Loyalty when using the repeated indicator approach?

Is it as simple as calculating indirect effects? Eg. just multiplying the path coefficient from the lower-order construct on the higher-order construct with the path coefficient from the higher-order construct to loyalty, in the manner
Rugged -> Brandpersonality -> Loyalty

Hope you can help!

Best,

Jacob

Re: IPMA Performace of constructor

Posted: Wed Feb 06, 2019 12:58 pm
by MuktaS
Dear All,

I have a similar query. I am using reflective-reflective HCM using product indicator approach. When I am running IPMA, I am only getting total effects for second-order constructs, but not for the first order constructs.

Kindly tell me the process of getting the impact (total effects) for the first-order constructs in IPMA.

Regards
Mukta

Re: IPMA Performace of constructor

Posted: Mon Feb 11, 2019 2:33 pm
by jmbecker
1) What type of second-order construct?
2) What type of total effect are you missing?