**This project is for students in the Online Statistics & Data Science programme (M.Sc) only. No other student will be accepted for this project.*
Data quality is defined as ‘the capability of data to satisfy stated and implied needs when used under specific conditions’ (International Organisation for Standardisation). Data quality issues can include missing data, misspellings and typos, unit changes, schema changes, truncated values et cetera. Data users need to first detect the data issues and then handle the data issues appropriately. For example, in the case of missing data, data users may choose to implement listwise deletion, pairwise deletion, mean imputation, random forest imputation et cetera. The choice will generally depend on the percentage of data missing and the type of data missing (missing at random, missing not at random, or missing not completely at random). The choice of how to handle the missing data could impact the results of later analysis. Hence, a sensitivity analysis might be conducted to explore the results of analysis under different missing data assumptions. In this project, the student will examine when data quality issues begin to impact the results of data analysis and explore methods for detecting and solving these issues.