Accounting for Achievement in Athens: a Count Data Analysis of National Olympic Performance

Author Contact: Glen Roberts c/o David E. Giles, Dept. of Economics, University of Victoria, P.O. Box 1700, STN CSC, Victoria, B.C., V8W 2Y2, Canada; email: [email protected] (cc: [email protected]); FAX: (250) 721-6214 Abstract We model summer Olympic medal counts using count data analysis. The advantage of this methodology is its explicit recognition of the discrete non-negative form of the dependent variable; i.e. the total number of medals won by a nation in a summer Olympiad. Using data from the most recent 2004 Summer Games in Athens, Poisson and negative binomial count data regression models are constructed. The chosen model is negative binomial and attaches statistical significance to Gross Domestic Product (GDP) per capita, the age dependency ratio, and a relatively cold winter climate. In contrast to previous studies, population, health expenditure per capita, and the effect of being a host or neighbour nation are all insignificant in explaining medal counts. We also find no “cricket effect” or “rugby effect.”