Abstract
We dispute the claim made by Dubbelman and colleagues [J. Opt. Soc. Am. A 22, 1216 (2005); this issue] that “incorrect statistical methods” were used to compare the MRI and Scheimpflug data. We clearly stated that we “analyzed the covariance of regression lines” in the results section. We believe the analysis of covariance of regression lines, as shown by Snedecor and Cochran [Statistical Methods (Iowa State U. Press, Ames, Iowa, 1989)] is as “correct” as the “straightforward statistical comparison” of confidence intervals employed by Dubbelman and colleagues; however; our statistical method has the benefit of being more precise and stringent than a simple comparison of confidence intervals.
© 2005 Optical Society of America
Full Article | PDF ArticleMore Like This
Michiel Dubbelman, Rob G.L. van der Heijde, and Henk A. Weeber
J. Opt. Soc. Am. A 22(6) 1216-1218 (2005)
Jane F. Koretz, Susan A. Strenk, Lawrence M. Strenk, and John L. Semmlow
J. Opt. Soc. Am. A 21(3) 346-354 (2004)
Peter Ott
J. Opt. Soc. Am. A 22(6) 1077-1085 (2005)