This chapter introduces the theoretical rationale and practical application of variable transformation techniques to improve model fit, satisfy assumptions, and enhance interpretability. It begins with transformations for continuous variables, including standardization, log transformations, power transformations (e.g., Box-Cox), and winsorization. These methods are presented in the context of their effects on skewness, variance stabilization, and coefficient interpretation. The chapter also addresses transformation strategies for categorical variables, including dummy coding, effects coding, and treatment contrasts. Readers will learn when transformations are appropriate, when they might obscure interpretation, and how they interact with model specification. The emphasis throughout is on practical implementation and the analytical reasoning behind transformation decisions.