NotesFAQContact Us
Search Tips
ERIC Number: ED438322
Record Type: Non-Journal
Publication Date: 2000-Jan-29
Pages: 20
Abstractor: N/A
Reference Count: N/A
A Review of Experimentwise Type I Error: Implications for Univariate Post Hoc and for Multivariate Testing.
Altman, Dan
Although many researchers are familiar with "testwise" alpha, "experimentwise" Type I error rates are also important and must be considered in many research situations. Experimentwise error rates can inflate rapidly when researchers use multiple univariate tests. Both analysis of variance (ANOVA) post hoc and multi-variate methods incorporate a correction for experimentwise error. Researchers ought to understand experimentwise error if they are to understand what post hoc tests really do and an important rationale for multi-variate methods. (Contains 4 tables and 13 references.) (Author/SLD)
Publication Type: Information Analyses; Reports - Descriptive; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A