Sample size determination for one-and two-sample trimmed mean tests

Wei Ming Luh, Stephen Olejnik, Jiin Huarng Guo

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)

Abstract

Formulas to determine the necessary sample sizes for parametric tests of group comparisons are available from several sources and appropriate when population distributions are normal. However, in the context of nonnormal population distributions, researchers recommend Yuen's trimmed mean test, but formulas to determine sample sizes have not been developed. The authors address this omission and provide statistically rigorous formulas of sample size determination for 1- and 2-sample Yuen's trimmed mean tests. Computer simulations for a wide range of nonnormal distribution shapes evaluated the adequacy of the proposed formulas. The results show that the proposed formulas control Type I error satisfactorily and achieve desired power, whereas the traditional formulas resulted in oversampling units for nonnormal distributions.

Original languageEnglish
Pages (from-to)167-184
Number of pages18
JournalJournal of Experimental Education
Volume77
Issue number2
DOIs
Publication statusPublished - 2008 Dec 1

All Science Journal Classification (ASJC) codes

  • Education
  • Developmental and Educational Psychology

Fingerprint Dive into the research topics of 'Sample size determination for one-and two-sample trimmed mean tests'. Together they form a unique fingerprint.

Cite this