Multivariate Statistical Analysis of Decathlon Performance Results in Olympic Athletes (1988-2008)

The performance results of the athletes competed in the 1988-2008 Olympic Games were analyzed (n = 166). The data were obtained from the IAAF official protocols. In the principal component analysis, the first three principal components explained 70% of the total variance. In the 1st principal component (with 43.1% of total variance explained) the largest factor loadings were for 100m (0.89), 400m (0.81), 110m hurdle run (0.76), and long jump (–0.72). This factor can be interpreted as the 'sprinting performance'. The loadings on the 2nd factor (15.3% of the total variance) presented a counter-intuitive throwing-jumping combination: the highest loadings were for throwing events (javelin throwing 0.76; shot put 0.74; and discus throwing 0.73) and also for jumping events (high jump 0.62; pole vaulting 0.58). On the 3rd factor (11.6% of total variance), the largest loading was for 1500 m running (0.88); all other loadings were below 0.4.

Authors:

• Jaebum Park
• Vladimir M. Zatsiorsky

Keywords:

• Decathlon
• principal component analysis
• Olympic Games
• multivariate statistical analysis.

References:

[1] T.F. Cox TF, and R.T. Dunn, "An analysis of decathlon data," The
Statistician, vol. 51, no. 2, pp. 179-187, 2002.
[2] International Amateur Athletic Federation official website. Available at
www.iaaf.org.
[3] I.A. Kenny, D. Sprevak, C. Sharp, and C. Boreham C, "Determinants of
success in the Olympic decathlon: Some statistical evidence", Journal of
Quantitative Analysis in Sports, vol. 1, no. 1, Article 5, 2005.
[4] R. Levins (1968) Evolution in Changing Environments. Princeton
University Press. New Jersey
[5] R. Van Damme, R.S. Wilson RS, B. Vanhoydonck, and P. Aerts,
"Performance constraints in decathletes", Nature, vol. 415, pp. 755-756,
2002.
[6] V.M. Zatsiorsky, and M.A. Godik, "Mathematics and decathlon", Track
and Field, vol. 10, pp. 28-30, 1962 (In Russian).