These questions explore the distinctions between SmartPLS and AMOS in terms of their statistical methods, modeling capabilities, treatment of data, and user experience. It is essential to understand these differences to make informed decisions about which software tool is best suited for specific research projects and objectives. Researchers can use these questions as a starting point to delve deeper into the comparative analysis of SmartPLS and AMOS for their individual research needs.
Dear Neda Ravankhah
your response was incredibly insight, and I'm truly thankful for the time and effort you put into helping me.with utmost gratitude,
Dawit Negussie
Dawit Negussie : You may find this article useful: Article Estimation issues with PLS and CBSEM: Where the bias lies!
The new SmartPLS 4 release (https://www.smartpls.com/) offers PLS-SEM but also CB-SEM (like IBM SPSS Amos) - and many other methods such as regression analysis, path analysis and PROCESS, necessary condition analysis, logistic regression, and confirmatory factor analysis (via CB-SEM projects).
Dear Dawait, good question! SmartPLS and AMOS are both software tools commonly used in structural equation modelling (SEM), but they have some key methodological differences in terms of their underlying modelling approaches. SmartPLS is primarily associated with partial least squares structural equation modelling (PLS-SEM). PLS-SEM is a variance-based approach that focuses on predicting the variance in the observed variables. It is often used for exploratory research when the sample size is small or the data does not meet the assumptions of covariance-based SEM. AMOS, however, is a software package that primarily supports covariance-based SEM (CB-SEM). CB-SEM is a model-driven approach that emphasizes the relationships among latent variables and their ability to reproduce the observed covariance matrix. In SmartPLS, the measurement model is estimated separately from the structural model. It places more emphasis on the reliability and validity of the measurement model and allows for the use of reflective and formative indicators. In AMOS, the measurement and structural models are estimated simultaneously. It typically assumes that the measurement model is reflective, where latent variables cause the observed variables. SmartPLS is often recommended for smaller sample sizes, as it is considered more robust with limited data. AMOS is typically used with larger sample sizes, as it relies on the estimation of covariances and may require larger samples to produce stable parameter estimates. PLS-SEM in SmartPLS is less sensitive to distributional assumptions and is robust to non-normal data. It can handle both formative and reflective measurement models. CB-SEM in AMOS assumes multivariate normality and is more sensitive to violations of this assumption. It is primarily used with reflective measurement models. SmartPLS uses a bootstrapping procedure for estimation, which is a resampling technique. Bootstrapping provides confidence intervals for parameter estimates and is particularly useful when sample sizes are small. AMOS relies on maximum likelihood estimation (MLE) for parameter estimation. MLE is based on the likelihood function and assumes multivariate normality in the data. SmartPLS often uses goodness-of-fit measures such as the average path coefficient, R-squared values, and the Q-squared statistic to assess model fit. AMOS provides various fit indices like chi-square, Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), and Root Mean Square Error of Approximation (RMSEA) to assess the fit of the model to the data. In summary, the choice between SmartPLS and AMOS depends on your research objectives, the nature of your data, and the specific modelling approach you want to employ. SmartPLS is favoured when dealing with smaller samples, non-normal data, or when formative constructs are part of the model. AMOS is preferred for larger samples and when covariance-based SEM is the method of choice with reflective constructs. Researchers often choose the tool that best suits their research context and data characteristics.
It's essential to understand that Structural Equation Modeling (SEM) comprises two primary approaches: covariance-based SEM (CB-SEM) and partial least squares SEM (PLS-SEM). The good news is that SmartPLS has the capability to handle both CB-SEM and PLS-SEM. In contrast, Amos exclusively deals with CB-SEM.
Before utilizing any of these software tools, it's important to grasp when to employ each of these two approaches. You can find valuable guidance in the following references.
Sarstedt, Marko and Hair, Joseph F. and Ringle, Christian M., “PLS-SEM: Indeed a Silver Bullet” – Retrospective Observations and Recent Advances (April 22, 2022). Journal of Marketing Theory and Practice, 2022, Available at SSRN: https://ssrn.com/abstract=4090346
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing theory and Practice, 19(2), 139-152.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression