Dataset (Raw Example) – Customer Sales Data

Tasks & Sample Answers

1. Identify different types of errors

Missing values:

Duplicates:

Outliers:

Mary Adams (Purchase_Amount = -100 → negative sales not possible).
Peter Obi (Purchase_Amount = 15000 → extremely high compared to others, potential entry error).

Inconsistencies:

2. Clean the dataset

Remove duplicates: Keep only one John Smith.

Impute missing values:

Age for Jane Brown → take from row 6 (36).
Purchase_Amount for Sam Johnson → replace with average sales (≈ 374 after correction).
Location for David Lee → assume “Unknown” or infer from company records.

Handle outliers:

Cleaned Dataset

3. Reflection

Improvement in data quality:

Risks of poor data (if not cleaned):

M3:S4 - Practical Exercise

Data Cleaning Basics