Abstract: | Attempts to clarify some issues that have stemmed from J. E. Overall and D. K. Spiegel's (see record 1970-01534-001) article on least squares analysis, as well as from a number of subsequent articles. C. E. Werts and R. L. Linn (1971) provided a logical basis for selecting Method 3 from the Overall and Spiegel article. The present article outlines statistical as well as substantive arguments for the choice of Method 3. A variation of Method 3 is also suggested. (PsycINFO Database Record (c) 2010 APA, all rights reserved) |