Exploratory Data Analysis and Data Cleaning

Introduction

Before building a dashboard or model, analysts need to understand the dataset they have. Exploratory data analysis (EDA) uses summaries, visualisations, and checks to learn:

• How the data is structured
• Where it may be unreliable
• Which patterns deserve closer investigation

This module covers practical techniques for inspecting data, treating missing values, identifying duplicates and outliers, checking consistency, and documenting changes. The focus is on careful, explainable choices, not automatic cleaning without understanding.

What you will learn in this module

• How to profile a dataset before analysis
• How to investigate missing data and duplicates
• How to identify potential outliers without deleting valid records
• How to standardise formats and validate assumptions
• How to document cleaning steps so results can be reproduced