Before administering a test, we often ask: **Does it work equally well across different groups, or is it biased?** For example, some items might favor students based on **gender**, **ethnicity**, or **language background**. This tutorial extends a basic Classical Test Theory (CTT) item analysis by introducing Differential Item Functioning (DIF) analysis using the Mantel–Haenszel (MH) method. The goal is to detect items that function differently across groups after conditioning on total test score. Item 5 will be artificially manipulated to demonstrate DIF. We are moving from classical item analysis to DIF detection because some items might favor one group over another even if overall ability is the same. MH DIF is a widely used method to detect this bias.

This presentation simulates a realistic multiple-choice test, conduct a Classical Test Theory (CTT) item analysis, and interpret item-level and test-level statistics using R. The workflow mirrors what is typically done in operational assessment and evaluation contexts.