Das Tor News

Countries Are What They Eat: A Statistical Adventure

By Aaron Rockwell, Staff Writer

The gods have blessed me with a ‘gutter palate.’ If they could make a tasteless, perfectly nutritional milkshake, I would drink it for every meal.

So when I came across how the United Nations statistics team creates a dataset of ‘what countries are eating,’ I bit into it. Beyond the UN’s data, I grabbed data from all sorts of other sets to see correlations to different country metrics. So buckle up–let’s ride.


Correlation of Food to IQs
Data Courtesy: https://iq-research.info & http://faostat3.fao.org

Eggs! They had no correlation to a country’s GDP per capita but were strongly correlated to IQ. Maybe they really are a super food. Here’re the top 5 egg consuming countries, quite a geographically diverse team:

Courtesy: http://cooking.my.panasonic.com
Courtesy: http://cooking.my.panasonic.com


Here are some of the things countries with a higher percentage of folks with BMIs above 30 consume:

Data Courtesy: http://apps.who.int & http://faostat3.fao.org
Data Courtesy: http://apps.who.int & http://faostat3.fao.org

I was shocked that chicken was the number one correlator. The only explanation I could think of was that the ways people often eat chicken (chicken wings, fried chicken, etc.) might contribute to obesity.

Main Take-Aways:

This little look into the data world gives way to saying:

  • Eat more eggs, apples, potatoes, and milk (pigmeat was so closely related to GDP per capita, I’d be leery of saying it correlates to improved IQ)
  • Add more soy, freshwater fish, rice, and sweet potatoes if trying to lose weight
  • I have a hard time suggesting avoiding chicken, but for sure avoid sugar. Seems to be news to me, but also avoid those lemons and limes!


Most of the data would hit around 160 compared data points giving a ~7.5% margin of error.

Also, I compared GDP per capita to ensure that correlations were not just applicable because the countries were rich, e.g., ability to have increased medical expenditures.

Correlations were created with the Pearson product-moment correlation coefficient to measure the strength of a linear association between two variables and is denoted by r.


Courtesy of: https://statistics.laerd.com
Courtesy of: https://statistics.laerd.com

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.