visStatistics (version 0.1.3)

vis_anova_assumptions: Visualisation of the normality distribution of the standardised residuals of the ANOVA

Description

vis_anova_assumptions checks for normality of the standardised residuals of the ANOVA. Both the Shapiro-Wilk test shapiro.test() and the Anderson-Darling test ad.test() check the null that the standardised residuals are normally distributed. It generates a scatter plot of the standardised residuals versus the fitted mean values of the linear models for each level of fact. Furthermore a normal QQ plot of the standardised residuals is generated. The null of homogeneity of variances of each factor level is tested with the bartlett.test().

Usage

vis_anova_assumptions(
  samples,
  fact,
  conf.level = 0.95,
  samplename = "",
  factorname = "",
  cex = 1
)

Value

list containing the test statistics of the anova, the p values generated by the Shapiro-Wilk test shapiro.test(), the Anderson-Darling test ad.test() and the bartlett.test().

Arguments

samples

vector containing dependent variable, datatype numeric

fact

vector containing independent variable, datatype factor

conf.level

confidence level, 0.95=default

samplename

name of sample used in graphical output, dataype character , ''=default

factorname

name of sample used in graphical output, dataype character, ''=default

cex

number indicating the amount by which plotting text and symbols should be scaled relative to the default. 1=default, 1.5 is 50% larger, 0.5 is 50% smaller, etc.

Examples

ToothGrowth$dose <- as.factor(ToothGrowth$dose)
vis_anova_assumptions(ToothGrowth$len, ToothGrowth$dose)

vis_anova_assumptions(ToothGrowth$len, ToothGrowth$supp)
vis_anova_assumptions(iris$Petal.Width, iris$Species)