adhdData: Children's reaction times (milliseconds) to stimuli of...

In welchADF: Welch-James Statistic for Robust Hypothesis Testing under Heterocedasticity and Non-Normality

Description

The data (Keselman et al., 2003) represent the reaction times in milliseconds of children with attention-deficit hyperactivity (ADHD) and normal children when they are presented four kinds of inputs: a target alone or an arrow stimuli incongruent, congruent and neutral to the target. According to the authors, the dataset was artificially generated from the summary measures given in the original study by Jonkman et al. (1999), in groups of 20 and 10 children to create an unbalanced design.

Usage

adhdData

Format

A data frame with 30 rows and 5 variables:

Group

whether the child has ADHD or is healty (normal)

.

TargetAlone

Reaction time (milliseconds) to a target alone.

Congruent

Reaction time (milliseconds) to a congruent stimulus.

Neutral

Reaction time (milliseconds) to a neutral stimulus.

Incongruent

Reaction time (milliseconds) to an incongruent stimulus.

Source

H. J. Keselman, R. R. Wilcox, and L. M. Lix. A generally robust approach to hypothesis testing in independent and correlated groups designs. Psychophyisiology, 40:586-596, 2003. (Data displayed in page 593).

References

L. Jonkman, C. Kemner, M. Verbaten, H. van Engeland, J. Kenemans, G. Camfferman, J. Buitelaar, and H. Koelega. Perceptual and response interference in children with attention-deficit hyperactivity disorder, and the effects of methylphenidate. 36(4):419-429, 1999.

Examples

# Assuming a multivariate response
omnibus_LSM_multi <- welchADF.test(adhdData, response = c("TargetAlone", "Congruent",
  "Neutral", "Incongruent"), between.s = "Group", within.s = "multivariate", contrast = "omnibus")
# The same using the S3 method for class formula
omnibus_LSM_multi_form <- welchADF.test(cbind(TargetAlone, Congruent, Neutral, Incongruent)
~ Group, data = adhdData)
  
# Pairwise comparisons of the implicit within-subjects effect present in the multivariate response
pairwise_LSM_multi <- update(omnibus_LSM_multi, contrast = "all.pairwise", effect = "multivariate")
summary(omnibus_LSM_multi)
summary(pairwise_LSM_multi)

Example output

Call:
   welchADF.test(formula = adhdData, response = c("TargetAlone", 
    "Congruent", "Neutral", "Incongruent"), between.s = "Group", 
    within.s = "multivariate", contrast = "omnibus")

                     WJ statistic Numerator DF Denominator DF  Pr(>WJ)   
Group                      0.2249            1          24.84 0.639482   
multivariate               5.6591            3          21.02 0.005282 **
Group : multivariate       0.5750            3          21.02 0.637759   
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Call:
   welchADF.test(formula = adhdData, response = c("TargetAlone", 
    "Congruent", "Neutral", "Incongruent"), between.s = "Group", 
    within.s = "multivariate", contrast = "all.pairwise", effect = "multivariate")

                        WJ statistic Numerator DF Denominator DF adj.pval   
Congruent:TargetAlone         3.8254            1          20.52 0.256928   
Incongruent:TargetAlone      14.2331            1          27.90 0.003868 **
Neutral:TargetAlone           0.6215            1          25.19 0.459884   
Congruent:Neutral             1.6498            1          16.87 0.459884   
Incongruent:Neutral          15.6572            1          27.99 0.002830 **
Congruent:Incongruent         0.5703            1          18.07 0.459884   
---
Signif. codes (Hochberg p-values):  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

welchADF documentation built on Sept. 8, 2019, 9:02 a.m.