Key points are not available for this paper at this time.
Zelditch (1971) and Schoenberg (1972) have called attention to the need to compare structural parameters in different populations.1 Such comparisons can yield information about collective characteristics that influence the operation of the modeled system in different settings. Blalock (1964) proposes such comparisons and points out that comparison of standardized regression coefficients (path coefficients) is confounded by variance
Specht et al. (Thu,) studied this question.