|
|
|
|
|
|
PATH ANALYSIS
In statistics, path analysis is a type of multiple regression analysis. The term path analysis has been used to refer to the analysis of causal models when single indicators are empoyed for each of the variables in the model. Other terms used to refer to this model are causal modeling, analysis of covariance structures, latent variable models, structural modeling, and structural equation modeling. In the hypothetical model below, the two exogenous variables are taken as correlated and are shown to have direct as well as indirect effects (through En1) on En2.
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 to Ex1 to En2 would be deleted. It may be that the endogenous variables (dependent variables) are also affected by variables other than the identified exogenous variables (independent variables) and thus not in the model. The effects of such extraneous variables are depicted by the various e’s in the figure.
In Internet website analytics, path analysis is process of determining a sequence of pages visited in a visitor session prior to some desired event, such as the visitor purchasing an item or requesting a newsletter. The precise order of pages visited may or may not be important and may or may not be specified. In practice, this analysis is done in aggregate, ranking the paths (sequences of pages) visited prior to the desired event, by descending frequency of use. The idea is to determine what features of the website encourage the desired result. "Fallout analysis," a subset of path analysis, looks at "black holes" on the site, or paths that lead to a dead end most frequently, paths or features that confuse or lose potential customers.
- For the project management sense of path analysis, see critical path and PERT.
|
|
|
|
|
|
|