In the last Southeast Asian Games held in the Philippines, a Southeast Asian blogger suggested that there’s a correlation between the number of medals won by a country with the country’s population size and wealth. That sounds reasonable to me. With respect to the ongoing Asian Games at Doha, let’s test it.
Let’s touch on the data first. I use 2005 GDP at PPP (IMF) and population size of Asiad country-participants as listed at Wikipedia. The GDP at PPP is used as a proxy variable to wealth. Data on medals collected by countries as of 0400 Greenwich time is obtained from the official site of the 2006 Asiad. In order to differentiate between gold, silver and bronze, I assign three points to gold, two to silver and one to bronze. I have the all the data in one file and you may have it if you’d like to play around with it.
I got MS Excel to run the necessary regression. I know, it’s a bad choice but I don’t have access to other statistical software. I did download some free, legit softwares off the internet but that was too much hassle.
So, on MS Excel, I regressed medal points — number of medals multiplied by point assigned — on population per thousand and GDP at PPP per million.
Before I reveal the result, let’s talk about my initial hunch. I’d think population size and wealth have positive relationship to medals won by countries. To generalize it further, if we take medals won as a proxy to strength in sports, population size and wealth would contribute positively to countries’ strength in sports. What do you think about that?
Now, the result supports that wealth increases the number of medal won. Specifically, each billion of GDP at PPP leads to a 0.0008 increase in medal point, with all else constant of course.
The surprise comes from the correlation between population size and number of medal won. Each thousand leads to 0.0002 decrease in medal point; an inverse relationship, with the typical caveat, ceteris paribus.
The output:
Honestly, I’m kind of skeptical of my own regression.
Regardless, on a different set of regression – medal points on GDP at PPP per capita – reveals that a dollar increase in GDP (PPP) per capita increases medal point at about 0.0018, ceteris paribus. The regression result if you’re interested in it:
How significant are the figures?
Well…
