Assess whether a dataset is fit for analysis by checking missing values, duplicates, inconsistent categories, invalid ranges and data types, then document the cleaning decisions you make.
Applied work for this week
Audit a small dataset, record the quality issues you find, apply justified cleaning steps and produce a concise data-quality note explaining what changed and why.
- 01Profiling a dataset before cleaning
- 02Missing values: diagnose before deciding
- 03Duplicates and entity-level consistency
- 04Incorrect data types and formatting problems
- 05Outliers, unusual values and data errors
- 06Validation rules and reasonableness checks
- 07Cleaning categorical values and labels
- 08Dates, time periods and temporal consistency
- 09Joining and combining tables safely
- 10Transformations, derived variables and units
- 11Documenting every cleaning decision
- 12Mini-case: produce an analysis-ready dataset