Yilin Ying, Byte 1 for Big Data Pipeline

Jan 23, 2017

OECD's PISA Report --- World Education Ranking for 2010

link of data source: here


The Program for International Student Assessment (PISA) is a triennial international survey aiming to evaluate worldwide education systems by testing skills and knowledge of 15-year-old students.

In 2009, around 470,000 15-year-olds kids across the world took a numeracy, literacy and science test. The results were presented in the 2010 study by the Paris-based Organization for Economic Co-operation and Development (OECD). PISA data is a reflection of the education development of different countries and provides insight for countries to better develop their future education.

Interested Questions

In this project, we are interested in the following 3 questions and have manipulated some data exploration in Google Fusion Tables:

These questions are meaningful for each individual country to re-design their future education systems and curricula structures; for international organizations, such as World Bank, the data can provide insights on promoting education equality around the world.

1. What are the general education quality around the world in respect to the PISA test?

We use average test scores to roughly describe education quality and make comparisons between different countries. In the 3 graphs below, I put the average scores by subjects on intensity maps. The darker the color, the higher the scores.

(1) Average Reading Score

(2) Average Science Score

(3) Average Maths Score

2. Who Perform the Best?

I put the score rankings by subjects on bar charts. China (Shanghai), Finland, Hong Kong, Singapore, South Korea, Japan & Canada are on the rank for all of the subjects. We may want to conduct further case study on these countries to explore the reasons behind.

(1) Top in Reading Score

(2) Top in Science Score

(3) Top in Maths Score

3. Is there a correlation between different subject scores for these countries?

I made a correlation scattered plot for Maths, Science and Reading scores. According to the graphs below, there is a strong positive correlation between each pair of the scores among the countries, which indicates that if a country's students are good at one subject, it is more likely that they are also performing well in another subject. It is more likely that a country's sutdents perform good in all of the subjects or none. This implies that the development of education is not only the matter of each knowledge subject, but also should be considered from a systematic perspective.

(1) Average Reading Score by Average Science Score

(2) Average Science Score by Average Maths Score

(3) Average Reading Score by Average Maths Score


First, due to time limitation, I just did cross sectional data analysis and didn't make analyze on time series data to explore a country's education quality changing over time. Second, PISA is not the only or the best way to evaluate or compare different countries' education quality. We should also consider each country's economical, social and cultural development. Third, the data is not complete, We are missing instances from African countries. Thus, the understanding of worldwide education quality is partial in this exploration.