Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Notes and Comments: Comments on “Contrast Coding in Least Squares Regression Analysis”:

Notes and Comments: Comments on “Contrast Coding in Least Squares Regression... American Educational Research Journal Winter 1978, Vol 15, No. 1, Pp. 149-151 Notes and Comments Comments on "Contrast Coding in Least Squares Regression Analysis" R. CLIFFOR D BLAIR J.J . HIGGINS University of South Florida Kaufman and Sweet (1974) performed a service for educational researchers by bringing together in one article much of the more useful information pertaining to the regression analysis of unbalanced factorial designs. However, several errors in the presentation are potentially misleading. Th e purpose of this pape r is to delineate and correct these difficulties. Th e first error is conceptual in nature. Kaufman and Sweet seem to feel that the criterion variance in an unbalanced factorial design can be decomposed and interprete d in the manner described by Pugh (1968). (Compare Pugh's Figure 1 with Kaufman and Sweet's Figures 1 an d 2.) Thu s Kaufman and Sweet state (p. 356) that "Figure 2 shows the amount of criterion variation which may be attributed to a particular source of variation independently of others in the model at each step in this analysis" (italics added). Indeed, this statement is even stronger than any made by Pugh, who carefully refrained from using terms such as "independent" http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Educational Research Journal SAGE

Notes and Comments: Comments on “Contrast Coding in Least Squares Regression Analysis”:

Download PDF
3 pages

Loading next page...
 
/lp/sage/notes-and-comments-comments-on-contrast-coding-in-least-squares-Gjs0ZmvQzP

References (9)

Publisher
SAGE
Copyright
Copyright © 2019 by American Educational Research Association
ISSN
0002-8312
eISSN
1935-1011
DOI
10.3102/00028312015001149
Publisher site
See Article on Publisher Site

Abstract

American Educational Research Journal Winter 1978, Vol 15, No. 1, Pp. 149-151 Notes and Comments Comments on "Contrast Coding in Least Squares Regression Analysis" R. CLIFFOR D BLAIR J.J . HIGGINS University of South Florida Kaufman and Sweet (1974) performed a service for educational researchers by bringing together in one article much of the more useful information pertaining to the regression analysis of unbalanced factorial designs. However, several errors in the presentation are potentially misleading. Th e purpose of this pape r is to delineate and correct these difficulties. Th e first error is conceptual in nature. Kaufman and Sweet seem to feel that the criterion variance in an unbalanced factorial design can be decomposed and interprete d in the manner described by Pugh (1968). (Compare Pugh's Figure 1 with Kaufman and Sweet's Figures 1 an d 2.) Thu s Kaufman and Sweet state (p. 356) that "Figure 2 shows the amount of criterion variation which may be attributed to a particular source of variation independently of others in the model at each step in this analysis" (italics added). Indeed, this statement is even stronger than any made by Pugh, who carefully refrained from using terms such as "independent"

Journal

American Educational Research Journal – SAGE

Published: Nov 23, 2016

There are no references for this article.