Statistical Methods

P > 0.05 0.01 < P < 0.05 0.001 < P < 0.01 * * * multiple regression analyses were tested by t tests. Regression lines were compared by means of analysis of covariance in which differences in slopes and in distances were tested. In analysis of variance F-tests were used to test main effects and in linearcontrast-analyses for pairwise comparisons ( 1 2 ) . The degree of probability was denoted as: