Prior to statistical analyses, skewed variables were transformed
to approach normality and to reduce the influence of extreme values,
using either angular or logarithmic transformations (Table 1).
All variables were then standardised to zero mean and unit variance,
to enhance comparability of effect sizes (e.g. Schielzeth,
2010). Collinearity was investigated using pairwise correlations,
and one variable from each pair correlated at >0.7 was eliminated
from further analyses (Dormann et al., 2013). All statistical analyses
were performed using the R statistical package (R Project for
Statistical Computing release 2.15.3 ).
Prior to statistical analyses, skewed variables were transformedto approach normality and to reduce the influence of extreme values,using either angular or logarithmic transformations (Table 1).All variables were then standardised to zero mean and unit variance,to enhance comparability of effect sizes (e.g. Schielzeth,2010). Collinearity was investigated using pairwise correlations,and one variable from each pair correlated at >0.7 was eliminatedfrom further analyses (Dormann et al., 2013). All statistical analyseswere performed using the R statistical package (R Project forStatistical Computing release 2.15.3 ).
การแปล กรุณารอสักครู่..
