Now,I use SmartPls2.0 to solve the SEM problem . And I have some questions about Smartpls. If someone knew it,Please help me .The questions as follows:
1.How to explain the validity of model by the results of Smartpls' output. The standard of judge is what.
2.The biggest problem is how to calculate the ACSI scores by the result of output. That can the weights and the scores of ACSI Formula be getted from the output?
Thanks a million !!!
questions
-
Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi
1) The assessment of the measurement model:
1.1) Run the PLS algorithm:
Report / PLS / Quality criteria / Overview:
- AVE equal or greater than .5 (convergent validity)
- reliability equal or greater than .7
Report / PLS / Quality criteria / Latent Variable Correlations
- all correlations should be lower than the AVE squared root (discriminant validity)
Report / PLS / Quality criteria / Cross loadings
- the outer loading of each indicator should be greater in its LV and lower in the others (discriminant validity).
- the outer loading should be greater than .7 (convergent validity).
1.2) Run the bootstrap
Report / Bootstrapping / Bootstrapping / Outer loadings (Mean, standard deviation, T-values)
- all t values should be greater than 1.96 (5%, two tail)
2) The assessment of the structural model:
2.1) Run the PLS algorithm:
- in the model you will see the path coefficients and inside the blue balls the R2
Report / PLS / Calculation results / Path coefficients
- The same results that were showed in the model
Report / PLS / Quality criteria / Total effects
- you could assess the direct, indirect, and total effects.
2.2) Run the bootstrap
- in the model you will see the t-values of the path coefficients
Report / Bootstrapping / Bootstrapping / path coefficients (Mean, STDEV, T-values)
- all t values should be greater than 1.96 (5%, two tail)
The structural model assessment will give you a criterion validity or nomological validity.
About the ACSI scores:
In the PLS algorithm report / Index values / Results / Latent variable scores (unstandardized)
You will have the LV scores in the same unit of the indicators
Or
Report / PLS / Calculation results / Latent variable scores
You will have the standardized LV scores (mean = 0 and StdDev = 1), than you will must to convert them to the scale (probably 0 – 100) that you want.
Best regards,
Bido
1) The assessment of the measurement model:
1.1) Run the PLS algorithm:
Report / PLS / Quality criteria / Overview:
- AVE equal or greater than .5 (convergent validity)
- reliability equal or greater than .7
Report / PLS / Quality criteria / Latent Variable Correlations
- all correlations should be lower than the AVE squared root (discriminant validity)
Report / PLS / Quality criteria / Cross loadings
- the outer loading of each indicator should be greater in its LV and lower in the others (discriminant validity).
- the outer loading should be greater than .7 (convergent validity).
1.2) Run the bootstrap
Report / Bootstrapping / Bootstrapping / Outer loadings (Mean, standard deviation, T-values)
- all t values should be greater than 1.96 (5%, two tail)
2) The assessment of the structural model:
2.1) Run the PLS algorithm:
- in the model you will see the path coefficients and inside the blue balls the R2
Report / PLS / Calculation results / Path coefficients
- The same results that were showed in the model
Report / PLS / Quality criteria / Total effects
- you could assess the direct, indirect, and total effects.
2.2) Run the bootstrap
- in the model you will see the t-values of the path coefficients
Report / Bootstrapping / Bootstrapping / path coefficients (Mean, STDEV, T-values)
- all t values should be greater than 1.96 (5%, two tail)
The structural model assessment will give you a criterion validity or nomological validity.
About the ACSI scores:
In the PLS algorithm report / Index values / Results / Latent variable scores (unstandardized)
You will have the LV scores in the same unit of the indicators
Or
Report / PLS / Calculation results / Latent variable scores
You will have the standardized LV scores (mean = 0 and StdDev = 1), than you will must to convert them to the scale (probably 0 – 100) that you want.
Best regards,
Bido
-
HussainWaasly
- PLS Senior User
- Posts: 21
- Joined: Thu Mar 21, 2013 7:37 pm
- Real name and title:
Where can i find articles about the above numbers. i meant based on what you wrote "AVE equal or greater than .5 (convergent validity)Diogenes wrote:Hi
1) The assessment of the measurement model:
1.1) Run the PLS algorithm:
Report / PLS / Quality criteria / Overview:
- AVE equal or greater than .5 (convergent validity)
- reliability equal or greater than .7
Report / PLS / Quality criteria / Latent Variable Correlations
- all correlations should be lower than the AVE squared root (discriminant validity)
Report / PLS / Quality criteria / Cross loadings
- the outer loading of each indicator should be greater in its LV and lower in the others (discriminant validity).
- the outer loading should be greater than .7 (convergent validity).
1.2) Run the bootstrap
Report / Bootstrapping / Bootstrapping / Outer loadings (Mean, standard deviation, T-values)
- all t values should be greater than 1.96 (5%, two tail)
2) The assessment of the structural model:
2.1) Run the PLS algorithm:
- in the model you will see the path coefficients and inside the blue balls the R2
Report / PLS / Calculation results / Path coefficients
- The same results that were showed in the model
Report / PLS / Quality criteria / Total effects
- you could assess the direct, indirect, and total effects.
2.2) Run the bootstrap
- in the model you will see the t-values of the path coefficients
Report / Bootstrapping / Bootstrapping / path coefficients (Mean, STDEV, T-values)
- all t values should be greater than 1.96 (5%, two tail)
The structural model assessment will give you a criterion validity or nomological validity.
Best regards,
Bido
- reliability equal or greater than .7" ?
I want to add references to all provided numbers by you in my research .
Regards,
-
Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi Hussain,
1) The Classical about AVE and CC:
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 39–50.
2) Newer:
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: indeed a silver bullet. The Journal of Marketing Theory and Practice, 19(2), 139–152. doi:10.2753/MTP1069-6679190202
Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares path modeling in international marketing. Advances in International Marketing, 20, 277–319. doi:10.1108/S1474-7979(2009)0000020014
Peng, D. X., & Lai, F. (2012). Using partial least squares in operations management research: A practical guideline and summary of past research. Journal of Operations Management, 30(6), 467–480. doi:10.1016/j.jom.2012.06.002
3) More here: http://www.mendeley.com/groups/1116711/pls-pm/papers/
Best regards,
Bido
1) The Classical about AVE and CC:
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 39–50.
2) Newer:
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: indeed a silver bullet. The Journal of Marketing Theory and Practice, 19(2), 139–152. doi:10.2753/MTP1069-6679190202
Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares path modeling in international marketing. Advances in International Marketing, 20, 277–319. doi:10.1108/S1474-7979(2009)0000020014
Peng, D. X., & Lai, F. (2012). Using partial least squares in operations management research: A practical guideline and summary of past research. Journal of Operations Management, 30(6), 467–480. doi:10.1016/j.jom.2012.06.002
3) More here: http://www.mendeley.com/groups/1116711/pls-pm/papers/
Best regards,
Bido
Re:
Diogenes wrote:Hi
1) The assessment of the measurement model:
1.1) Run the PLS algorithm:
Report / PLS / Quality criteria / Overview:
- AVE equal or greater than .5 (convergent validity)
- reliability equal or greater than .7
Report / PLS / Quality criteria / Latent Variable Correlations
- all correlations should be lower than the AVE squared root (discriminant validity)
Report / PLS / Quality criteria / Cross loadings
- the outer loading of each indicator should be greater in its LV and lower in the others (discriminant validity).
- the outer loading should be greater than .7 (convergent validity).
1.2) Run the bootstrap
Report / Bootstrapping / Bootstrapping / Outer loadings (Mean, standard deviation, T-values)
- all t values should be greater than 1.96 (5%, two tail)
2) The assessment of the structural model:
2.1) Run the PLS algorithm:
- in the model you will see the path coefficients and inside the blue balls the R2
Report / PLS / Calculation results / Path coefficients
- The same results that were showed in the model
Report / PLS / Quality criteria / Total effects
- you could assess the direct, indirect, and total effects.
2.2) Run the bootstrap
- in the model you will see the t-values of the path coefficients
Report / Bootstrapping / Bootstrapping / path coefficients (Mean, STDEV, T-values)
- all t values should be greater than 1.96 (5%, two tail)
The structural model assessment will give you a criterion validity or nomological validity.
About the ACSI scores:
In the PLS algorithm report / Index values / Results / Latent variable scores (unstandardized)
You will have the LV scores in the same unit of the indicators
Or
Report / PLS / Calculation results / Latent variable scores
You will have the standardized LV scores (mean = 0 and StdDev = 1), than you will must to convert them to the scale (probably 0 – 100) that you want.
Best regards,
Bido
Dear Bido,
Does the comparison of t-values of the respective path coefficients of two variables sufficient to claim that one has a greater impact than the other? Since I can not do the Chi -square difference test in Smartpls, what would you suggest me?
Looking forward to hearing form you
Best Regards
Yesim