Using Data Analytics to Better Understand the Impact of COVID-19 in Brazil

Introduction

The COVID-19 pandemic had a profound impact on communities around the world. In this article, we'll explore how data analytics can help us better understand the impact of the pandemic in Brazil. We'll use a real-world example of a project I recently completed, analyzing COVID-19 data in Brazil using Python and Google Looker Studio. The dataset used is updated daily and can be found at this link. Furthermore, the analysis covers the scenario from 2021-2023.

Steps

1. Understanding the Problem: Before diving into the data, it's important to define the problem we're trying to solve. In this case, we're trying to understand the impact of COVID-19 in Brazil, including the number of cases, deaths, tendencies, moving averages as well as information by geography (states and regions).
2. Gathering Data: The next step is to gather relevant data. As beforementioned, I used public data source from John Hopkins Univeristy to collect data on COVID-19 in Brazil.
3. Cleaning and Prepping the Data: Once the data is collected, it needs to be cleaned up and properly prepared for analysis. This step is critical for ensuring the accuracy and reliability of our results.
4. Analyzing the Data: Using Python and Google Looker Studio, I was able to analyze the data and generate visualizations that help us to interactively understand the impact of COVID-19 in Brazil. The link for both dashboard and data processing Python notebook can be found at the Conclusion section.

The dataset

The dataset provided by John Hopkins Univeristy nis a great resource to work on your Data Analytics projects. Indeed, it has several columns in which you can generate a few other columsns. The dictionary for the Covid-19 dataset can be seen below:

• Variable: descritption
• date: reference date;
• state: country state;
• country: country;
• population: estimated population;
• confirmed: accumulated number of infected people;
• confirmed_1d: daily number of infected people;
• confirmed_moving_avg_7d: 7-days moving average for infected people;
• confirmed_moving_avg_7d_rate_14d: 7-days moving average for infected people divided by the 7-days moving average of 14 days before;
• deaths: accumulated number of deaths;
• deaths_1d: daily number of deaths;
• deaths_moving_avg_7d: 7-days moving average for deaths;
• deaths_moving_avg_7d: 7-days moving average for deaths divided by the 7-days moving average of 14 days before;
• month: reference month;
• year: reference year.

What the Google Looker Studio dashboard can provide us

In brief, the Google Looker Studio dashboard can provide us Key Performance Indicators (KPIs) and Interactive Exploratory Data Analysis (EDA). The KIPs are the following:

• Cases and deaths within 24 hours;
• Moving average (7 days) of cases and deaths;
• Trend of cases and deaths.

Meanwhile, the interactive EDA displayed on the Covid-19 in Brazil dashboard are the following:

• Distribution of numbers of cases and deaths over time;
• Distribution of the moving average (7 days) of the number of cases and deaths over time;
• Geographic distribution of cases by state by day.

Of course, the same process could be done with several other tools and techniques. For example, one can use R instead of Python to process data, or even use Tableau or Power BI to create the dashboard. The tools are only the media to reach a Data Analytics Solution.

Conclusion

The data analysis shows that the COVID-19 pandemic has had a significant impact on Brazil, with high numbers of cases, hospitalizations, and deaths. The use of data analytics helps us better understand the situation and inform decision-making.

By this example, it is noticeable that Data Analytics is valuable to understand complex issues like the COVID-19 pandemic. By using a programming language and Business Intelligence (BI) tool, we can extract valuable insights and overlook the current scenario. The Google Looker Studio final dashboard can be found here and it also includes the link for the Python notebook used for data collection and processing.