I personally prefer SPSS AMOS for SEM as compared to SmartPLS. If you want to calculate real time results, better use AMOS because it does not automatically covariate every latent variable whereas SmartPLS has by default setting to covariate the latent variables. This create differences in results of both tools.
“The philosophical distinction between CB‑SEM and PLS‑SEM is straightforward. If the research objective is theory testing and confirmation, then the appropriate method is CB‑SEM. In contrast, if the research objective is prediction and theory development, then the appropriate method is PLS‑SEM. Conceptually and practically, PLS‑SEM is similar to using multiple regression analysis. The primary objective is to maximize explained variance in the dependent constructs but additionally to evaluate the data quality on the basis of measurement model characteristics.” (Hair et al., 2011, p.140).
I personally prefer AMOS as it is more reliable but for your convenience recommend Mat Roni, Saiyidi (2014) following points:
Covariance based SEM (CB-SEM): AMOS, LISREL
Partial least square SEM (PLS-SEM): SmartPLS, WarpPLS
1. Think about your research objective:
If you want to predict outcomes using your model -> use PLS-SEM
If you want to confirm your data fits your model -> use CB-SEM
2. Look at your data distribution:
If it's normally distributed --> use CB-SEM
non-normal, and you don't want to transform --> use PLS-SEM
3. Look at number of items per construct:
If all constructs have minimum of 3 items (questions) each --> Both CB-SEM and PLS-SEM are OK.
If any construct has less than 3 items --> use PLS-SEM
4. Look at sample size:
Small sample --> use PLS-SEM
large sample --> both approaches are OK.
Advisable sample size, N > 50 + 8m, where N is sample size, m is predictor variables (Tabachnick and Fidell, 2007, p.123).
Reference
Dated. 18-07-2012 Mat Roni, Saiyidi. (2014). Re: How to decide which package is best for my research "Amos" or "Smart PLS" ?. Retrieved from: https://www.researchgate.net/post/How_to_decide_which_package_is_best_for_my_research_Amos_or_Smart_PLS/545c6a21d3df3ebc128b468d/citation/download.
I personaly prefer to use AMOS for SEM. Its very robust, easy to understand and fairly easy to use and interpret the output. However, i would highly reccomend to check your data for any missing values, because if you dont and you have missing values, you will end up with loads of errors in your analysis and output.
The philosophical distinction between CB‑SEM and PLS‑SEM is straightforward. If the research objective is theory testing and confirmation, then the appropriate method is CB‑SEM. In contrast, if the research objective is prediction and theory development, then the appropriate method is PLS‑SEM. Conceptually and practically, PLS‑SEM is similar to using multiple regression analysis. The primary objective is to maximize explained variance in the dependent constructs but additionally to evaluate the data quality on the basis of measurement model characteristics.” (Hair et al., 2011, p.140).