Graphical and Statistical Analysis

The simplest way to analyse the effects is, perhaps, by means of proper graphics, among which the Pareto chart and the main effects or interactions effects plots are used widely. In a Pareto chart we represent the different effects ordered by magnitude (absolute value, on the vertical axis) and the magnitude 

Experimental design does not explain the nature of any interaction, as this is a chemical task, but it provides helpful suggestions to explain it. Fairly often, 

Fitting the data to an empirical surface response can also be used to estimate the response within the range of experimental conditions. The simplest model that we can consider is the following  

Source Sum of squares Mean square F-ratio p-Value 

We can remove from this model all terms related to a non-significant factor or interaction. Therefore, in our example, the model is simplified to 

---

The proper combination of graphical and statistical analysis will usually provide definitive answers for most problems in model discrimination. Severd specific examples are given below. 

---