Skip to content

MANE 6313

Week 4, Module C

Student Learning Outcome

Analyze simple comparative experiments and experiments with a single factor.

Module Learning Outcome

Justify model assumptions incorporated in One-Way Analysis of Variance.


Model Adequacy Testing

  • Normality

    • Most often checked with normal probability plots of residuals

    • Will detect non-normality and outliers

  • Residual Analysis

    • Plot of residuals in time sequence

      • Will detect correlation between residuals and also process drift
    • Plot of residuals versus fitted values

      • Will detect non-constant variance
    • Plot of residuals versus factors

      • Will detect non-constant variance

Standardized Residuals

  • Interpretation is easier if the residuals are standardized

AOV Structure

R AOV Structure


Graphical Normality Check in R

Normalilty Graphs in R


Shapiro-Wilks Normality Test

Shapiro Wilks Test for Normaility


Simple Residual Plot

Simple Residual Plot


Cluttered Residual Plots

Cluttered Residual Plots


Single Residual Plot

Residuals versus Fitted


Standardized Residual Plot

Standardized Fitted versus Residual Plot


One More Residual Plot

Additive versus Residuals


R Demonstration