Analysis of Variance II: Factorial Designs; Does Marijuana Slow You Down? and Do Slimming Clinics Work?

2003 ◽  
pp. 162-181
Keyword(s):  
Analysis Of Variance ◽  
Factorial Designs

Note on the Denominator of ω2 in Factorial Designs

1973 ◽  
Vol 37 (3) ◽  
pp. 825-826
Author(s):  
Joseph Levin
Keyword(s):  
Factorial Design ◽  
Analysis Of Variance ◽  
Factorial Designs ◽  
Relevant Factor ◽  
Psychological Variables ◽  
Normally Distributed

The denominator of ω2 in a factorial design is inflated owing to rectangularity of the distribution of the factors in analysis of variance, while many quantitative psychological variables are normally distributed. The magnitude of the bias depends on the correlation between the dependent variable and the relevant factor and is therefore relative to each experiment. Corrections of ω2 for range are discussed.

Teaching Moderated Multiple Regression for the Analysis of Mixed Factorial Designs

1995 ◽  
Vol 22 (3) ◽  
pp. 197-198 ◽  
Author(s):  
Robin M. Kowalski
Keyword(s):  
Multiple Regression ◽  
Analysis Of Variance ◽  
Continuous Variable ◽  
Factorial Analysis ◽  
Factorial Designs ◽  
Moderated Multiple Regression

Traditionally, researchers have analyzed mixed factorial designs by dichotomizing the continuous variable and performing a factorial analysis of variance. Several problems surround the use of this approach, prompting many statisticians to recommend using moderated multiple regression as an alternative. This article presents the case for moderated multiple regression and provides on overview of the procedure.

scholarly journals The application of orthogonal contrasts to determine homogeneous groups

2021 ◽  
Vol 82 ◽  
pp. 31-44
Author(s):  
Zbigniew Laudański ◽  
Dariusz Mańkowski ◽  
Leszek Sieczko ◽  
Monika Janaszek-Mańkowska
Keyword(s):  
Analysis Of Variance ◽  
Multiple Comparison ◽  
Factorial Designs ◽  
Homogeneous Groups ◽  
Orthogonal Contrasts ◽  
Multiple Comparison Procedure ◽  
Effective Use

The paper presents a modified approach to analysis of data obtained from experiments carried out according to classical factorial designs. Four examples were discussed in order to present details of proposed method. Modification of the analysis of variance presented here enables more effective use of information on how studied factors affect the means of dependent variable. The specificity of this approach is based on alternative multiple comparison procedure incorporating orthogonal contrasts to determine homogeneous groups.

Sign in / Sign up

Export Citation Format

Share Document