Objective
Structural Equation Modeling is used to inspect causal relationships between variables. The objective is to investigate hypothetical structures of correlations and influences. As with regression analysis, the analysis looks for the amount and direction of influence of one or more so-called exogenous variables on one or more endogenous variables.
As with factor analysis, the assumption is made that the attribute of real interest cannot be observed directly, but rather that it “latently” represents the basis for the observed behavior, opinion or attitude expressed by the respondent. These latter observations serve as indicators for the level of the underlying basic attributes.
Fig 1: Example of a causal model – Explanation of a shop rating and the resulting level of product interest
Path analysis is initially based around a hypothesis stating which attributes are influenced by which variables. In this manner, complex covariance structures can be set up, whereby attributes can simultaneously be independent (influencing) and dependent (influenced). The path analysis checks this hypothesis by measuring the degree of the influences and determining the degree of information conveyed by such a model.
Prerequisites
Path analysis cannot detect an improperly assumed direction of influence. The direction of the influence, or in other words, the question of whether feature an influences feature B or vice versa, must be established through preliminary theoretical consideration.
In order to be able to interpret the coefficients properly, the influencing variables of an endogenous variable should be as mutually independent of one another as possible.
RALV
With RALV (Relationships Among Latent Variables), IfaD provides a tool for the realization of path analysis, allowing a stable and unbiased estimation of strengths of influence. A special feature of RALV is a convenient and flexible option to avoid multicolinearity (orthogonalization of variables) in order to achieve comparable coefficients.
