Dear Al academicians,
I hope you are all fine and good in health.
I want to ask that which software should I use to Apply Exploratory and confirmatory analysis and SEM in my In research Paper based on primary data. For example, i use SPSS to filter the data only but other studies show AMOS is a better option than PLS while applying SEM. i am confused now. can someone practically explain how this software is different from others and which software is proven to be the best in terms of the accuracy of the outcome?
Hi
The difference between the software mentioned by you can be summarized in three categories:
1- The difference in the graphical interface, the user interface, AMOS software seems to have been more successful in these two fields.
2- Differences in reporting and using better and more accurate reporting of results: In this case, we can consider LISREL and AMOS as the leaders.
3- Differences in computational algorithms and computational methods: In this case, I consider LISREL to be the strictest and most accurate.
@Sarfaraz you need to decide whether you look to go for covariance based SEM or PLS based SEM. As covariance based SEM requires the assumptions of normality and also need large sample size whereas PLS based SEM doesn't require normality assumption and also works well with small sample. If you have borrowed validated scales to measure your constructs, I suggest you to go for Smart-PLS.
Dr Sarfaraz Javed for EFA you use the SPSS, for CFA you have two options SmartPLS and AMOS. Your study types and methodology will lead you to choose the software for your study.
Syed Naseeb Ullah, for example, I want an exploratory study in this case what should choose? the methodology will be EFA and CFA.
Dr Sarfaraz Javed we choose EFA and CFA on the basis of the questionnaire.
This means you adapted, adopted, or develop the questionnaire for your study.
If you use an already developed questionnaire, in such cases there is no need for the EFA, only run CFA. If you develop the questionnaire by your self then you need to run both EFA and CFA
No one yet has mentioned lavaan (www.jstatsoft.org/article/view/v048i02). It is free, fairly easy to use, and has many tutorials available (e.g., https://lavaan.ugent.be/).
Good evening my friend, if you have built your questionnaire on a solid theoretical basis, with well-defined constructs, I suggest that you use the CFA directly and then the SEM on AMOS. So you don't need to use EFA. As for accuracy, I think AMOS is more rigorous than PLS.
Respected Dr Sarfaraz Javed ,
Its depends of your reserach objectives whether variances based (PLS-SEM) technique is better or co-variance based (CB-SEM). If your objective is to test the theory than CB-SEM is better while if your objective is to predict than you can go for PLS-SEM.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling