This Article Full Text (PDF) References Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Bonett, D. G. Articles by Seier, E. Search for Related Content Social Bookmarking                         What's this?

Statistical Inference for a Ratio of Dispersions Using Paired Samples

Iowa State University

Edith Seier

East Tennessee State University

Wilcox (1990) examined the Type I and Type II error rates for several robust tests of H₀: ₁²/₂² = 1 in paired-data designs and concluded that a satisfactory solution does not yet exist. A confidence interval for a ratio of correlated mean absolute deviations is derived and performs well in small sample sizes across realistically nonnormal distributions. When used to test a hypothesis, the proposed confidence interval is almost as powerful as the most powerful test examined by Wilcox.

Key Words: Box-Scheffé test • directional two-sided hypothesis • finite-interval hypothesis • matched-pairs designs • mean absolute deviation from median • Morgan-Pitman test • repeated measures designs • variability • within-subjects designs

Journal of Educational and Behavioral Statistics, Vol. 28, No. 1, 21-30 (2003)
DOI: 10.3102/10769986028001021


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?