2009
DOI: 10.22237/jmasm/1241136180
|View full text |Cite
|
Sign up to set email alerts
|

Aligned Rank Tests for Interactions in Split-Plot Designs: Distributional Assumptions and Stochastic Heterogeneity

Abstract: Three aligned rank methods for transforming data from multiple group repeated measures (split-plot) designs are reviewed. Univariate and multivariate statistics for testing the interaction in split-plot designs are elaborated. Computational examples are presented to provide a context for performing these ranking procedures and statistical tests. SAS/IML and SPSS syntax code to perform the procedures is included in the Appendix.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
14
0
1

Year Published

2016
2016
2024
2024

Publication Types

Select...
9
1

Relationship

0
10

Authors

Journals

citations
Cited by 18 publications
(15 citation statements)
references
References 81 publications
0
14
0
1
Order By: Relevance
“…In particular, to explore the main effects, we used an Aligned Rank Transformation (ART) as described by Wobbrock, Findlater, Gergle and Higgins 39 with the r package ARTool 40 . To test for interactions, we used a non-parametric ART test for interactions in designs with repeated measures as described by Beasley & Zumbo 41 with the r package “npIntFactRep” 42 . However, while this last package permits to test effects for mixed models with one ‘ within’ and one ‘ between’ factors, it does not allow for testing interactions with more than one ‘ within’ factors, hence we only ran interaction tests for the Threshold elevation data.…”
Section: Methodsmentioning
confidence: 99%
“…In particular, to explore the main effects, we used an Aligned Rank Transformation (ART) as described by Wobbrock, Findlater, Gergle and Higgins 39 with the r package ARTool 40 . To test for interactions, we used a non-parametric ART test for interactions in designs with repeated measures as described by Beasley & Zumbo 41 with the r package “npIntFactRep” 42 . However, while this last package permits to test effects for mixed models with one ‘ within’ and one ‘ between’ factors, it does not allow for testing interactions with more than one ‘ within’ factors, hence we only ran interaction tests for the Threshold elevation data.…”
Section: Methodsmentioning
confidence: 99%
“…Since the bubble scores were calculated on an ordinal scale and given the size of analyzed sample, we used nonparametric tests for comparisons. The Higgins & Tashtoush formula for repeated measures (at 0, 30, 60 and 90 minutes after decompression) was used for the interaction alignments with the aligned Koch rank score (based on ranking the K² pairwise differences among the 4 levels of the repeated measures, regardless of treatment membership ranks), and the resulting p-value was reported [ 13 , 14 ] Between groups differences per time point and for the cumulative bubble score were analyzed using paired Wilcoxon tests adjusted for multiple comparisons by the Hochberg method [ 15 ]. Differences in biometric data were tested by Student’s T-test, Mann Whitney U tests or Chi-square tests, as appropriate.…”
Section: Methodsmentioning
confidence: 99%
“…The Beasley and Zumbo (2009) procedure yielded all significant p-values, 0.0427, 0.0054, and 0.0339, respectively, for the aligned regular, Friedman, and Koch rank tests. The ARTool resulted in a non-significant value p = 0.0544.…”
Section: Nonparametric Testsmentioning
confidence: 92%