Using SAS® PROC CALIS for path analysis: An introduction

This article presents a straightforward approach for using PROC CALIS programs that will perform path analyses using observed variables. The step‐by‐step approach should help the beginner understand concepts in path analysis and how to use the SAS^® System's PROC CALIS procedure. The approach begins with the development of a program figure, a figure that graphically illustrates the researcher's theoretical model and identifies the parameters to be estimated in the model. A set of 14 rules guide this task. The article illustrates how to convert the program figure into a PROC CALIS program that will estimate parameters in the diagrammed model and how to review the output of PROC CALIS (e.g., residuals, chi‐square test, fit indices) to assess the fit of the model to the data.     

Designing simple library hierarchies using linear programming

The problem of designing a best three level library hierachy is considered from two different points of view. While the form of the optimization problems obtained is a little more complicated, it is shown that at most the methods of linear programming suffice to solve them. Certain design parameters may be specified by the problem solver, and it is shown that the best objective function values are convex in these, thus making possible the extraction of considerable additional, useful information. Finally, the methodology is extended to longer hierarchies and found to be feasible if the length is moderate.