2015
DOI: 10.1080/00273171.2014.968836
|View full text |Cite
|
Sign up to set email alerts
|

Power and Type I Error Control for Univariate Comparisons in Multivariate Two-Group Designs

Abstract: Simulations were conducted to evaluate the statistical power and Type I error control provided by several multiple-comparisons procedures in two-group designs. Stepwise Bonferroni-based procedures, which are known to control the familywise Type I error rate, tended to be more powerful than other methods but did not control the per-family Type I error rate (PFER). It is proposed that more attention should be given to the PFER, particularly with regard to these procedures. Only two methods controlled the PFER: t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
24
1

Year Published

2015
2015
2024
2024

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 32 publications
(25 citation statements)
references
References 48 publications
0
24
1
Order By: Relevance
“…We also verified the individual anova s for each variable to diagnose which one differed between the groups. Although we found mild correlations among vocal features (maximum Pearson's r = 0.441), these relationships do not influence the effect of manova in any way (Frane ). To test whether acoustic variables could diagnose differences between populations, we used a discriminant function analysis (DFA).…”
Section: Methodscontrasting
confidence: 53%
“…We also verified the individual anova s for each variable to diagnose which one differed between the groups. Although we found mild correlations among vocal features (maximum Pearson's r = 0.441), these relationships do not influence the effect of manova in any way (Frane ). To test whether acoustic variables could diagnose differences between populations, we used a discriminant function analysis (DFA).…”
Section: Methodscontrasting
confidence: 53%
“…Similar significant differences were also evidenced by multivariate analysis of variance ( manova ) for which the Pillai–Bartlett statistic (Hand and Taylor ) among VCUs ( V = 0.13) and genotypes ( V = 0.10) was significant at the 1‰ probability level. The main difference between anova and manova is that while the former tests for the difference in means between groups, manova tests for the difference in appropriate vectors of means and can protect against type I errors (Frane ) that might occur if multiple anova s were conducted independently as in this work. VCU 2013 and VCU 2014 performed comparably for b * value and tuber grades, but VCU 2014 outperformed VCU 2013 for total yield, the number of tubers per plant, tuber dry matter content, and produced more stems per plant (Table ).…”
Section: Resultsmentioning
confidence: 99%
“…No of employees (coded as the categories: less than 10, from 10 to 100, from 100 to 250 and more than 250 employees), turnover (coded as: less than 1 million, from 1 to 10 millions, from 10 to 50 millions and more than 50 millions) and geographical location (India / the UK). In addition, multivariate analysis of variance (MANOVA) is utilized in the case where the dependent variable is a combination of a set of correlated observed items (Frane, 2015). For the MANOVA models, the cronbach"s alpha test for reliability has been previously applied for checking of the reliability of the analysis conducted.…”
Section: Discussionmentioning
confidence: 99%