Bad design of a database structure may lead to difficulties with implementing operations modifying (update, insert, delete) the database content.
Let's consider the following examples assuming that a student should always belong to exactly one group and that a student may be registered to any number (0,1,2...) of courses.
Problem: Same information is kept in multiple rows.
Example: Changing a Group of a student may require updates in multiple rows. When such change is only partially successful data become inconsistent.
Problem: Unrelated information is kept in different columns. Proper adding of a complete row is impossible.
Example: It is impossible to add a student Dany004 to the Python course until the student is assigned to a Group.
Problem: Unrelated information is kept in different columns. Deleting intended partial information in impossible.
Example: It is impossible to remove the student Cam003 from the Python course without removing the student from the assigned First Group.
Splitting unrelated information to separate tables resolves anomalies.
Database normalisation is a formal description of this process.