Details
| Original language | English |
|---|---|
| Pages (from-to) | 3332-3343 |
| Number of pages | 12 |
| Journal | Communications in Statistics - Theory and Methods |
| Volume | 45 |
| Issue number | 11 |
| Early online date | 28 Apr 2016 |
| Publication status | Published - 2 Jun 2016 |
Abstract
ABSTRACT: Decision-tree approaches are commonly used in the analysis of toxicological assays. This paper considers the widely-used and recommended approach of first applying the ANOVA F-test and, if it is significant, using the many-to-one comparison procedure of Dunnett (1955). Data configurations are provided for which the F-test is not significant, but the Dunnett-test is significant. Because a conditional F-test before the Dunnett-test may increase the false negative rate, it is recommended that the ANOVA F-test not be used at all as a global pre-test. In addition, related non-parametric tests are investigated.
Keywords
- ANOVA, decision-tree approach, Dunnett-test
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Communications in Statistics - Theory and Methods, Vol. 45, No. 11, 02.06.2016, p. 3332-3343.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - The two-step approach—a significant ANOVA F-test before Dunnett's comparisons against a control—is not recommended
AU - Hothorn, Ludwig A.
PY - 2016/6/2
Y1 - 2016/6/2
N2 - ABSTRACT: Decision-tree approaches are commonly used in the analysis of toxicological assays. This paper considers the widely-used and recommended approach of first applying the ANOVA F-test and, if it is significant, using the many-to-one comparison procedure of Dunnett (1955). Data configurations are provided for which the F-test is not significant, but the Dunnett-test is significant. Because a conditional F-test before the Dunnett-test may increase the false negative rate, it is recommended that the ANOVA F-test not be used at all as a global pre-test. In addition, related non-parametric tests are investigated.
AB - ABSTRACT: Decision-tree approaches are commonly used in the analysis of toxicological assays. This paper considers the widely-used and recommended approach of first applying the ANOVA F-test and, if it is significant, using the many-to-one comparison procedure of Dunnett (1955). Data configurations are provided for which the F-test is not significant, but the Dunnett-test is significant. Because a conditional F-test before the Dunnett-test may increase the false negative rate, it is recommended that the ANOVA F-test not be used at all as a global pre-test. In addition, related non-parametric tests are investigated.
KW - ANOVA
KW - decision-tree approach
KW - Dunnett-test
UR - http://www.scopus.com/inward/record.url?scp=84969932011&partnerID=8YFLogxK
U2 - 10.1080/03610926.2014.902225
DO - 10.1080/03610926.2014.902225
M3 - Article
AN - SCOPUS:84969932011
VL - 45
SP - 3332
EP - 3343
JO - Communications in Statistics - Theory and Methods
JF - Communications in Statistics - Theory and Methods
SN - 0361-0926
IS - 11
ER -