Path analysis (statistics)

From Wikipedia, the free encyclopedia

In statistics, path analysis is a type of multiple regression analysis. Path analysis was invented by geneticist Sewall Wright around 1918 and he wrote about it more extensively in the 1920s. In addition to being thought of as a form of multiple regression focussing on causality, path analysis can be viewed as a special case of structural equation modeling - one in which only single indicators are employed for each of the variables in the causal model. Put another way, path analysis is SEM with a structural model, but no measurement model. Other terms used to refer to path analysis include causal modeling, analysis of covariance structures, and latent variable models.

In the model below, the two exogenous variables (Ex₁ and Ex₂)are modeled as being correlated and as having both direct and indirect effects (through En₁) on En₂ (the two dependent or 'endogenous' variables. In most real models, the endogenous are also affected by factors outside the model (including measurement error). The effects of such extraneous variables are depicted by the 'e' or 'error' terms in the model.

Using the same variables, alternative models are conceivable. For example, it may be hypothesized that Ex₁ has only an indirect effect on En₂, thus the arrow from Ex₁ to En₂ would be deleted, and the likelihood or 'fit' of these two models can be compared statistically.

[edit] See also

• Causality
• Latent variable model
• Structural equation model