2.8. Statistics
In the microcosm experiments, we compared the change in the number of A. aphidimyza and
C. carnea before and after exposure to H. axyridis using a 2-sample t-test. A repeated-measures mixed model analyses of variance(ANOVA) was used to compare differences in aphid abun-dance between microcosm treatments.
Block was included as a random effects factor and treatment, time, and their interaction as fixed effects factors.
Differences in aphid and predator abundance between treatments were assessed by comparing least squares means.
Aphid counts were square-root transformed prior to analysis to meet the assumptions of the model.
In the field cage experiment, the impact of excluding either large coccinellid predators (H. axyridis and C. sep-tempunctata) or all predators on aphid populations was
assessed using a repeated measures ANOVA random coefficients model with time included as a covariate and random effects for subject and subject-specific slopes on time.
Based on residual analysis aphid data were square-root transformed to meet the assumptions of the model.
Variation in the abundance of predators was also assessed using a repeated measures ANOVA random coefficients model with time included as a covariate and random effects for
subject and subject-specific slopes on time. A square-root
transformation was applied to predator means prior to
analysis to meet the assumptions the model.
Differences in aphid and predator abundance between treatments were assessed by comparing least squares means. Both the micro-cosm and field cage analyses were conducted using the PROC MIXED procedure of SAS version 9.1 (SAS Insti-tute, 1999).