4.3.6. Data Analysis

Data of EC50 from the two repetitions of each experiment were combined after checking for homogeneity of the experimental error variances by the F test (*P* ≥ 0.05)p. Subsequently, data were tested for normality, homogeneity of variances, and residual patterns. For *Experiments I* and *II*, a one-way ANOVA was conducted, with the EC50 as dependent variable and 'cultivar' or 'treatment' as independent variables. For *Experiment III*, a factorial ANOVA was conducted with EC50 as dependent variable, and 'cultivar', 'treatment', and their interaction as independent variables. Since interaction was significant (*P* = 0.0001), one-way ANOVAs were conducted for each olive cultivar. Treatment means were compared according to Fisher's protected LSD test or Tukey´s HSD test (both at *P* = 0.05) for the *Experiment I* (*n* = 3) and *III* (*n* = 6), or for *Experiment II* (*n* = 10), respectively [37]. Study data were analyzed using Statistix 10.0 software [38].
