A3-DavidZats

From CS 294-10 Visualization Sp10

Jump to: navigation, search

In this assignment, my initial question was:

How do countries compare in terms of life expecatancy?

In order to answer this question, I went to the UNdata website (http://data.un.org/) and obtained the Human Development Index table. This table lists the life expectancy at birth for a large set of countries. It also lists the GDP per capita for each country. So, I used Tableau to plot a bar graph with the countries sorted by life expectancy set the colors of the bars to be a function of the per capita GDP. The results for the top 30 countries in terms of life expectancy can be seen in the following figure:

Clearly from this figure, there isn't a strong correlation between the per capita GDP of a nation and its life expectancy. The lack of a correlation caused my question to change to:

What is the correlation between life expectancy and adult obesity?

To answer this question, I went to the OECD website and downloaded an excel file containing a lot of data (http://www.oecd.org/document/16/0,2340,en_2649_34631_2085200_1_1_1_1,00.html). From this file, I obtained the most current obesity values for each country and added them as a new sheet to the excel file containing the Human Development Index Table. Then I used Tableau to perform an inner join, so that life expectancy and obesity could be placed on the same chart. The resulting chart can be seen in the following figure:

Although there does seem to be some correlation between obesity and life expectancy, from this bar chart, it did not seem to be very strong. Given the lack of a strong result, I searched the UNdata website and my question became:

Do the rates of death due to non-communicable diseases, cancer, injury, and cardio-vascular issues provide a strong indication of life expectancy?

This question was answered with data from the UNdata website (http://data.un.org/). I found a table that contained the death rates per 100000 people for each country. The following bar chart shows results for the top 30:

Caption: Mortality rates per 100000 people due to injury, cardiovascular issues, cancer, and non-communicable disease. Countries are sorted by life expectancy, with Japan having the longest.

In this bar chart, the countries have been sorted from top to bottom in terms of life expectancy, with Japan having the longest. As we can see from this figure, these four causes of mortality provide a strong indicator of the overall life expectancy in the country. Specifically, countries with higher total mortality rates tend to have decreased life expectancies. Additionally, the sum of these mortality rates seems to provide a much better indication of life expectancy than could be found if any one was used alone. However, other mortality factors should probably be included into this graph as well. For example, Cyprus has a much longer life expectancy rate than would be indicated by the sum of these four mortality rates. Based on these four rates, people living in Cyprus should have a shorter life expectancy than all of the other countries presented in this graph.



[add comment]
Personal tools