Skip to main content

Table 2 Comparison of linear and non-linear regression analysis of changes in swimming speed for female and male freestyle swimmers at national level across years to determine which model is the best

From: Changes in breaststroke swimming performances in national and international athletes competing between 1994 and 2011 –a comparison with freestyle swimming performances

  Kind of regression Sum of Squares DOF AICC Best regression AIC-Test Best regression F-Test Delta Probability Likelihood
50 m freestyle polynomial 0.0067 0 −108.11 linear undetermined 29.06 4.87 e-07 100%
male Swiss swimmers linear 0.0077 16 −137.18
100 m freestyle polynomial 0.0002 2 −45.86 linear linear 21.05 2.67 e-05 99.99%
male Swiss swimmers linear 0.0013 6 −66.92
200 m freestyle polynomial 0.0015 0 −134.90 linear linear 32.22 1.005 e-07 100%
male Swiss swimmers linear 0.0014 16 −167.13
50 m freestyle polynomial 0.0016 0 −132.88 linear undetermined 27.20 1.23 e-06 99.99%
female Swiss swimmers linear 0.0021 16 −160.08
100 m freestyle polynomial 0.0027 0 −124.34 linear undetermined 28.91 5.26 e-07 99.99%
female Swiss swimmers linear 0.0031 16 −153.25
200 m freestyle polynomial 0.0006 0 −152.27 linear undetermined 14.68 0.00064 99.93%
female Swiss swimmers linear 0.0014 16 −166.95
  1. For all distances, the changes were linear.