Figure 3.

Schematic of a factorial design to study the effect of compound X on the histological score with sex as a second independent variable of interest. Compared with the completely randomised design (figure 2), the analysis now has two independent variables of interest (sex and treatment). The allocation also differs in that the assignment to treatment group happens for each level of sex independently. In this design, the experimental unit and observation unit is an individual Wistar rat and the biological unit is the Wistar rat which encompasses both male and female animals and represents the increased external validity of the study by the inclusion of both sexes.