A2-JonathanChung
From CS294-10 Visualization Fa07
Contents |
[edit] Dataset
I am using three different data sets, the World Wine Statistics, Penn World Table, and custom data depicting the distances between countries (through capital city distances). The World Wine Statistics provides pertinent information on worldwide wine production and consumption. The Penn World Table provides purchasing power parity and national income accounts converted to international prices for 168 countries for some or all of the years 1950-2000.
[edit] Format
I used Tableau to create my visualizations.
To keep the data set simpler, I pruned the results down to 25 select countries from around the world. All data was put into an Excel spreadsheet and then imported into Tableau. I simplified the data further by restricting it to the latest year - 1999. I noticed that in general there wasn't a whole lot of change in wine consumption from year to year, so I left time out as a dimension.
I chose a scatter plot for my visualizations and overlaid a trend graph on them.
[edit] Questions about Wine Consumption
[edit] Per Capita Wine Consumption and GDP
Q1) Is there a relationship between the GDP of a country and its wine consumption?
From a sampling of 25 of the top 50 wine-consuming countries, there appears to be some correlation between GDP and wine consumption. In particular, countries with a higher GDP consume more wine per capita than those with a low GDP.
However, there are some exceptions. Japan is a country with a high GDP, but its inhabitants do not drink very much wine or alcohol at all. They drink sake at most, and that is consumed in small quantities. The US is another outlier here, which comes to me as a surprise given that California is one of the top wine-producing entities in the world. Perhaps it's because consumption of beer here is high, but the same cannot be said of wine.
Overall, I wasn't terribly satisfied with the results on this graph, so I refined the question a bit based on what I noticed in the data.
[edit] Per Capita Wine Consumption and Distance from Luxembourg
Q2) Is there a falloff in wine consumption as one gets farther from the countries of heavy consumption (France, Spain, Luxembourg)? I asked this question to see if there was a “halo” effect going on.
This was a more interesting one to plot out. I took the distances between the capital of Luxembourg and of the other country and plotted that against wine consumption. This generated interesting results, that if fitted, would best be shown as an exponential falloff. There are a few outliers here, namely a cadre of South American countries in which all are far from Luxembourg but have moderately high wine consumption. I noticed that a lot of South American countries are in the same position and believe that there is a halo effect going on there in which there is a central country responsible for this trend (Argentina) and follower countries nearby that do the same. On the flip side, there are a few relatively close countries that consume almost no wine. These are mainly countries in the Middle East and North Africa, where I imagine that wine consumption is prohibited by Islamic law. I do not know for sure, but this is my conjecture. For the rest however, wine consumption and distance from Luxembourg seems to fall off exponentially, which is an interesting find.
