Young, J. W. (1977). The function of theory in a dilemma of path analysis. Journal of Applied Psychology, 62(1), 108-110.
http://dx.doi.org/10.1037/0021-9010.62.1.108
Argues that path analysis allows increased confidence in making causal inference from correlational data by putting the focal relationships into more complete multivariate causal contexts. This necessarily increases the number of variables in the system. However, a mathematical consequence of increasing the number of variables is that the number of possible systems from which one must choose the right context rapidly increases and subsequently may result in decreased rather than increased confidence in causal inference. (For example, there are over 1,000,000 different 5-variable system configurations to choose from.) Good theory helps ameliorate this dilemma. A conservative approach to model specification is prescribed. (PsycINFO Database Record (c) 2016 APA, all rights reserved)