Where does the data originate?

The underlying data is sourced from the Environment, Social, and Governance Data dataset available in the World Bank data catalog. Our analysis centers on the indicator measuring the volume of published scientific and technical journal articles across different countries and years. For additional details on this indicator, please refer to the aforementioned link.

Understanding the Data

Upon examining the data, you will notice that it includes values for each country across various years.

The decimal values may appear unusual at first glance. This occurs because when multiple authors contribute to an article, each author receives a fractional count. For instance, if an article has two authors from France and three from Germany, France would receive 0.4 points, while Germany would get 0.6.

Now, let's examine some visual data representations.

What Countries Excel in Scientific Publications?

To address the fundamental question of which nations produce the most scientific articles, we start by analyzing the data.

The data clearly indicates that China leads the pack, surpassing the United States. However, this trend hasn't always been the case. The following line chart illustrates how China has outperformed the U.S. in scientific publications over the years.

Starting from nearly zero in 1996, China's output has consistently grown, overtaking the United States in 2016. This remarkable growth underscores the nation's commitment to advancing science.

Per Capita Scientific Production Insights

To determine which country generates the most scientific articles relative to its population, we calculated the number of articles per 100,000 individuals by multiplying the original figures by 100,000 and dividing by the total population for each nation and year.

Here are the findings:

The United States closely follows with 137 publications per 100,000 residents, while China stands at 47. Predominantly, European nations, particularly the Nordic countries, dominate the top rankings.

Creating Stunning Charts with Seaborn

As highlighted earlier, you can find all the code and data in this GitHub repository. Here are some helpful tips for creating visually appealing charts using Seaborn.

Selecting a vibrant color palette is essential. For this analysis, I utilized a palette from Canva called Rosettes and Cream, which I modified using www.coolors.co. The outcome is visually striking.

Moreover, I implemented a function to add padding around the charts generated by Seaborn:

def add_padding_to_chart(chart, left, top, right, bottom, background):

width = chart.size[0] + left + right

height = chart.size[1] + top + bottom

image = Image.new("RGB", (width, height), background)

image.paste(chart, (left, top))

return image

Before applying this function, it's necessary to convert the Seaborn figure to a PIL image using the following code snippet:

fig.canvas.draw()

data = np.frombuffer(fig.canvas.tostring_rgb(), dtype=np.uint8)

data = data.reshape((fig.canvas.get_width_height()[::-1]) + (3,))

plt.close()

chart_image = Image.fromarray(data)

chart_image = add_padding_to_chart(chart_image, 30, 20, 30, 30, background_color)

For more details, please refer to the Notebook.

Final Thoughts

Thank you for taking the time to read this analysis. If you found the information valuable, consider subscribing to my channel and following oscarl3o on Twitter. I look forward to your next visit!