Good analysis begins with reliable data.

Before researchers can interpret findings, they need to examine whether their data is accurate, consistent, appropriately structured and suitable for analysis.

The SRS Data Cleaning Checklist provides a systematic workflow for preparing research datasets while preserving transparency and research integrity.

What you'll learn

• How to preserve original data before making changes.
• How to identify duplicate records and inconsistent categories.
• How to detect invalid entries and out-of-range values.
• How to examine missing observations and decide how to handle them.
• How to investigate unusual values without automatically deleting them.
• How to document every important cleaning decision.
• How to assess whether a dataset is ready for analysis.

Worked examples

The guide uses a fictional student survey dataset containing deliberately introduced errors.

Examples demonstrate how to standardise inconsistent categories, verify suspicious values, investigate duplicates, examine missing data and document justified corrections.

What's included

• A practical data-cleaning checklist.
• Before-and-after cleaning examples.
• Guidance on missing data and unusual observations.
• A completed cleaning decision log.
• A reusable blank decision log for your own research.

How to use this guide

1. Preserve a copy of your original dataset.
2. Review the checklist before beginning analysis.
3. Compare the worked examples with problems in your own data.
4. Investigate each issue before making changes.
5. Record corrections, exclusions and unresolved uncertainties.
6. Complete the final validation checks before proceeding to analysis.

Important: Data cleaning should improve data quality without inventing evidence or concealing inconvenient observations. Every substantive decision should be justified and documented.

Suitable for: Undergraduate students, postgraduate students, PhD scholars, early-career researchers and anyone preparing survey or observational data for analysis.