A priori tests in repeated measures designs: Effects of nonsphericity

Boik, Robert J.

doi:10.1007/bf02293733

Cited by 127 publications

(66 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As soon as one considers specific, 1 df contrasts, as is often done after MANOVA is applied, the sphericity assumption is no longer in effect. Thus, a viable solution is to use the appropriate specific error term for each contrast (e.g., Boik, 1981) and avoid the sphericity assumption altogether. In our discussions thus far, we have made the usual assumptions (see Appendix A for a description of them).…”

mentioning

confidence: 99%

Using confidence intervals in within-subject designs

1994

View full text Add to dashboard Cite

mentioning

confidence: 99%

Using confidence intervals in within-subject designs

1994

View full text Add to dashboard Cite

“…Significant (P# their interaction with task-specific ability was confirmed by 0.05) results of ANOVAs were evaluated using hypothesisthe statistical analyses (see Table 2). The interaction driven linear contrasts calculated with specific error vari-EXPERIMENT3LOCATION was significant for all seven ances [3].…”

Section: Introductionmentioning

confidence: 86%

Restriction of task processing time affects cortical activity during processing of a cognitive task: an event-related slow cortical potential study

Lamm

Bauer

Vitouch

et al. 2001

Cognitive Brain Research

View full text Add to dashboard Cite

As is known from psychometrics, restriction of task processing time by the instruction to respond as quickly and accurately as possible leads to task-unspecific cognitive processing. Since this task processing mode is used in most functional neuroimaging studies of human cognition, this may evoke cortical activity that is functionally not essential for the particular task under investigation. Using topographic recordings of event-related slow cortical potentials, two experiments have been performed to investigate whether cortical activity during processing of a visuo-spatial imagery task is substantially influenced by the time provided to process the task. Furthermore, it was investigated whether this effect is additionally modulated by a subject's task-specific ability. The instruction to respond as quickly and accurately as possible led to increased negative slow cortical potential amplitudes over parietal and frontal regions and significantly interacted with task-specific ability. While cortical activity recorded over parietal and frontal regions was different between subjects with low and high spatial ability when processing time was unrestricted, no such differences were found between ability groups when subjects were instructed to answer both quickly and accurately. These results suggest that restricting processing time has considerable effects on the amount and the pattern of brain activity during cognitive processing and should be taken into account more explicitly in the experimental design and interpretation of neuroimaging studies of cognition.

show abstract

“…The interaction of the between-subjects (i.e., independent grouping or treatment variable) and the within-subjects (i.e., repeated measures) factors is of interest in many applications (Boik, 1993;Koch et al, 1980). In educational experiments, the interaction typically represents differential gains in achievement for a treatment group.…”

Section: Methodsmentioning

confidence: 99%

“…However, the multivariate approach would be expected to yield more precise confidence intervals than the univariate approach, especially in situations where the pooled covariance matrix is non-spherical (Boik, 1981).…”

Section: Multiple Comparison Procedures For Aligned Rank Proceduresmentioning

confidence: 99%

“…Thompson (2002) discusses a bootstrap methodology to compute confidence intervals for effect sizes from parametric analyses. For location parameters less sensitive to skewness, confidence intervals for medians have been proposed (Bonett & Price, 2002;Campbell & Gardner, 1988;Hodges & Lehmann, 1963).…”

Section: Defining Confidence Intervals For Interpretable Parametersmentioning

confidence: 99%

See 1 more Smart Citation

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

Beasley

Zumbo

2009

J. Mod. App. Stat. Meth.

View full text Add to dashboard Cite

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.

show abstract

A priori tests in repeated measures designs: Effects of nonsphericity

Abstract: repeated measures designs, sphericity criterion, multiple contrasts,

Cited by 127 publications

References 30 publications

Using confidence intervals in within-subject designs

Using confidence intervals in within-subject designs

Restriction of task processing time affects cortical activity during processing of a cognitive task: an event-related slow cortical potential study

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

Contact Info

Product

Resources

About