highly significantP < 0.0001
Linear regression of seasonal curve shape against methodological score was highly significant ( r 2 = 0.3, F = 16.9, P < 0.0001, test for normality = 0.2, P < 0.0001) confirming that the tendency to demonstrate seasonal dynamics between the different retained studies depends largely on the degree of statistical or mathematical technical Sophistication of the study.