James, L. R. (1980). The unmeasured variables problem in path analysis. Journal of Applied Psychology, 65(4), 415-421.
http://dx.doi.org/10.1037/0021-9010.65.4.415
The ramifications of inadequate attention to the problem of unmeasured variables in path analysis are discussed in terms of correlations between causal variables, the errors of causal equations, and the resulting bias in solutions of path coefficients. The discussion recognizes that obviation of the unmeasured variables problem is an unrealistic objective. Consequently, logic is provided in the form of decision steps to help investigators determine whether the influence of unmeasured variables that can be expected in any particular analysis is of sufficient seriousness to preclude the use of path analysis. (10 ref) (PsycINFO Database Record (c) 2016 APA, all rights reserved)