Hi Reader,
Starting next week, Iβll be offering a Black Friday sale on ALL of my courses.
Iβll send you the details tomorrow! π¨
βVisual Vocabulary (PDF)
Not sure which type of visualization to use? This beautiful poster from the Financial Times will help direct you to a suitable visualization based on the type of data you have and the story you're trying to tell.
Also available in Spanish, French, Chinese, and Japanese.
βTwo weeks ago, I showed you three visualizations that I created from the World Happiness Report using Datawrapper.
βLast week, I walked through the steps needed to clean the data in order to create the first visualization, namely a world map.
Today, I'll show you how I transformed the raw data into the line chart data using Python's pandas library. Here are the specific steps:
If you want to follow along with the code, you can run it online using Google Colab.
Hereβs the end result:
To start, I read in the dataset from a URL using the read_excel function and selected which columns to keep.
Then, I standardized the column names and rounded the happiness column to 2 digits using the round method.
I wanted the line chart to span multiple years so that we could examine happiness trends over time. However, not every country has happiness data for every year, so I needed to decide which years to include.
To inform this decision, I selected the year column, used the value_counts method to count the number of times each year appears, and used the sort_index method to sort the resulting Series by year.
I knew the line chart would end with 2022, so I decided to start with 2011 since that year had a lot of data and was long enough ago to create a interesting visualization.
First, I used the range and set functions to create a set of integers from 2011 through 2022, which I called years.
Then, I used the isin method to create a boolean Series marking which year values were within the years set, and I used that Series to filter the DataFrame.
As you can see, the only years left in the dataset are 2011 through 2022.
I decided to limit the visualization to only include countries that had complete data for the entire 12-year span. Thus, I needed to eliminate countries that were missing any years between 2011 and 2022.
There are many ways to accomplish this (as I learned from my Stack Overflow question), but the simplest method is to group the data by country, and then for each country, check whether the set of that country's years is equal to the years set.
You can see that there are only 83 countries left in the dataset, and each country has data for all 12 years.
In order to create this particular visualization, Datawrapper needs the year as the index, the country as the columns, and the happiness as the values within the table. In other words, the data needs to be in "wide" format rather than "long" format.
To reshape the data, I used the pivot method, specifying which columns should be used as the index, the columns, and the values.
Finally, I wrote the contents of the DataFrame to a file using the to_csv method. I included the index in the CSV (as the first column) since it contains the year values.
I uploaded the CSV into Datawrapper, customized the chart's appearance (by setting United States to red, for example), and published it online.
Hereβs the end result, which you can click on and interact with:
If you enjoyed this weekβs tip, please forward it to a friend! Takes only a few seconds, and it really helps me reach more people! π
Also, stay tuned for an email tomorrow about my Black Friday sale! π€
- Kevin
P.S. Frog riding a unicycleβ
Did someone awesome forward you this email? Sign up here to receive Data Science tips every week!
Join 25,000+ intelligent readers and receive AI tips every Tuesday!
Hi Reader, This week, I've got a short tip about AI agents, followed by some Data School news... π Tip #56: What are AI agents? Google is calling 2025 "the agentic era," DeepLearning.AI says "the agentic era is upon us," and NVIDIA's founder says "one of the most important things happening in the world of enterprise is agentic AI." Clearly AI agents are a big deal, but what exactly are they? Simply put, an AI agent is an application that uses a Large Language Model (LLM) to control its...
Hi Reader, Last week, I launched a brand new course: Build an AI chatbot with Python. 120+ people enrolled, and a few have already completed the course! π Want to join us for $9? π Tip #55: Should you still learn to code in 2025? Youβve probably heard that Large Language Models (LLMs) are excellent at writing code: They are competitive with the best human coders. They can create a full web application from a single prompt. LLM-powered tools like Cursor and Copilot can autocomplete or even...
Hi Reader, The Python 14-Day Challenge starts tomorrow! Hope to see you there π€ π Tuesday Tip: My top 5 sources for keeping up with AI I'll state the obvious: AI is moving incredibly FAST π¨ Here are the best sources I follow to keep up with the most important developments in Artificial Intelligence: The Neuron (daily newsletter) My top recommendation for a general audience. Itβs fun, informative, and well-written. It includes links to the latest AI news and tools, but the real goldmine is...