Sample Answer
Exploratory Data Analysis and Statistical Interpretation of a COVID-19 Vaccination Dataset
Introduction
Data analysis is a key part of understanding real-world problems, especially in public health, where decisions rely heavily on patterns found in large datasets. For this assignment, I selected a dataset related to COVID-19 vaccination trends. This topic is both relevant and widely studied, and it provides meaningful insights into how vaccination rollout progressed across different countries.
The aim of this report is to describe the dataset, explain the methods used for analysis, and present clear conclusions based on visual and statistical interpretation. The focus is on understanding vaccination distribution patterns and identifying general trends over time.
Description of the Dataset
The dataset used for this analysis is the “COVID-19 World Vaccination Progress” dataset available on Our World in Data. It contains global vaccination records collected from official government reports and health organisations.
The dataset includes information such as:
- Country names
- Date of reporting
- Total vaccinations
- People vaccinated with at least one dose
- People fully vaccinated
- Daily vaccination rates
- Vaccination per hundred population
This dataset is updated regularly and covers a wide range of countries, making it suitable for comparative and trend-based analysis.
For this report, I focused on a subset of the data covering three countries: United Kingdom, United States, and India. These countries were selected because they represent different population sizes, healthcare systems, and vaccination rollout strategies.
The data was collected from the publicly available platform Our World in Data, which compiles reliable information from official sources such as health ministries and the World Health Organization.
Description of Methods
The analysis was conducted using basic exploratory data analysis techniques supported by visualisation and simple statistical comparisons.
Data Cleaning and Preparation
Before analysis, missing values were identified and handled by removing incomplete entries for key variables such as total vaccinations and people fully vaccinated. Dates were converted into a time-series format to allow trend analysis over time.
Visualisation Methods
Several graphical methods were used to explore the dataset:
Line graphs were used to show vaccination progress over time for each country. This helped identify the speed and consistency of vaccine rollout.
Bar charts were used to compare total vaccination numbers across countries at specific time points.
Scatter plots were used to examine relationships between total vaccinations and people fully vaccinated.
These visual tools helped highlight differences in vaccination strategies and progress between countries.
Statistical Analysis
To support the visual findings, descriptive statistics were used, including mean, median, and growth rate comparisons.
A simple comparative analysis was also conducted to assess differences in vaccination speed between countries. For example, average daily vaccination rates were compared across the selected countries.
Although formal hypothesis testing such as t-tests could be applied, the focus of this extra credit analysis was mainly on descriptive statistics and pattern recognition rather than inferential modelling.
