Experiments with a Single Factor

This chapter considers the simplest class of experiments, that with a single factor. It presents modeling and analysis techniques like one-way fixed effects models and random effects models, analysis of variance (ANOVA), multiple comparisons, and residual analysis. The chapter derives expected mean squares in ANOVA tables for sample size determination. It demonstrates, by using a pulp experiment, that the corrected total sum of squares can be split up into two components: treatment sum of squares and residual sum of squares. The corresponding mean squares are obtained by dividing these sum of squares by their respective degrees of freedom. The chapter also provides a discussion on the assessment of model assumption.