When modeling data structures that contain repeating groups (multiple values of the same type), proper database design principles require creating additional classes or tables to represent these groups. This follows normalization rules and maintains data integrity by avoiding multi-valued attributes within a single record.