ABSTRACT
Path analysis is not magic; it does not prove causality. It does not make a silk purse out of a sow’s ear; it cannot turn poor data into valid causal conclusions. Like multiple regression, there are assumptions underlying path analysis and the use of multiple regression to estimate paths. Like multiple regression, path analysis is open to abuse. This chapter will discuss these assumptions and the dangers of path analysis; it will also discuss how to avoid the dangers of the method.
