Some of the important factors that must be examined when analyzing the statistical results of a path analysis or an SEM are that the characteristics and phases of the structural equation models are exposed, as well as the construction stages, these being the specification, identification, parameter estimation, adjustment evaluation, model re-specification and interpretation of results (Garson, 2013). The exploratory and confirmatory factorial analysis is presented as part of the construction of a model. There are two exceptional cases in which the model does not contain both parts and that are used relatively frequently. First, the path analysis model’s confirmatory factorial contains only the measurement model and the relationships between Latent variables can only be correlated. Like the path analysis, structure equation model will confirm the factorial that contains the model of measurement (Garson, 2013). Alongside, it is also similar in terms of path analysis models not containing latent variables; instead, the variables observables are equated with latent variables; The difference is that consequently, only there is the model of structural relationships. In return, the errors of Measurement and prediction errors are confused into a single common term (Garson, 2013).

It is indeed correct to highlight the importance of examining causal or theoretical models in both path analysis and SEM. One clarification however would be in the use of computed or what we sometimes call latent variables. In SEM we examine causal models by using multiple measured variables to create latent variables. These latent variables are then tested for relationships on the basis of the theory we are testing. What more could you say about latent variables? Could you provide an example of how you might go about developing a latent variable that may be of interest to you?

  • Cite any sources in APA format.
  • Attach a Report.

"Get 15% discount on your first 3 orders with us"
Use the following coupon

Order Now