Data Cleaning
The process of detecting and correcting errors, inconsistencies, duplicates, and impossible values in a dataset before analysis.
Data cleaning prepares raw data for analysis by finding and fixing problems: entry errors, impossible values such as an age of 200, inconsistent formats and codings, duplicate records, and responses that fail logic checks. It also involves standardizing labels, handling missing values according to a stated rule, and documenting every change made. Cleaning decisions are analytical decisions, so the raw data should be preserved and each transformation recorded, ideally in a script or log that can reproduce the cleaned dataset from the original.
In a thesis, cleaning is where credibility is quietly won or lost, because errors that survive it contaminate every result downstream. Describe your cleaning steps in the methodology chapter: what checks you ran, what you excluded, and why. A transparent, reproducible cleaning process reassures examiners that your findings rest on sound data rather than on unexamined files.
Writing the thesis this term belongs to?
CiteDash takes a thesis from research question to a compiled document, with AI that cites only real papers and verifies every claim against its source.