CBDA Study Guide 2026

An overfit model has learned the noise in the training data rather than the underlying pattern, causing poor generalization.

Small multiples repeat the same chart structure for different subgroups, enabling side-by-side comparison.

Standardization promotes consistency across the organization, simplifies maintenance, and enables better collaboration between team members.

Cross-validation is a robust technique used to validate the performance of a predictive model and assess how well it generalizes to an independent dataset. It involves partitioning the data into multiple subsets, training the model on a portion, and testing it on the remaining portion, repeating this process multiple times. This method helps to reduce overfitting and provides a more reliable estimate of the model's predictive accuracy.

Effective data storytelling begins with the key insight, then uses evidence to support and contextualize it.

A data disposal policy ensures that data no longer needed for business, legal, or compliance purposes is destroyed securely and in a documented manner to prevent unauthorized access.