I'm designing a database so I can enter my students marks using SQLite which is a non-relational database.

The information I have is
Course Name & Code (for example 'International Business', 'BUS800')
Course Section (for example '001')
Course Term (for example 'Fall'
Year Taught ('2007')

Student name & number ('John Doe', '000 000 000')

Assignments (there maybe 4 assignments and 2 exams)

Student Grade (grade for each assignment)

So I was thinking this might be the optimal way to make the database so I can compare the different classes or just print out marks from one course.

Table course (c_RowID, course_name, course_code) // 'International Business' 'US800'
Table section (s_RowID, section_number) // '001'
Table terms (t_RowID, term_name) // 'fall'
Table year (y_RowID, year_taught) // '2008'

TABLE course_relations (cr_rowID, c_rowID, s_rowID, t_rowID, y_rowID) //where cr_rowID is course_relations rowID so I can then compare this to Students, and Assignment Marks

Table student(student_ID, first_name, last_name)
Table assignments(a_rowID, assign_name)

TABLE assign_course_relations (acr_rowID, cr_rowID, a_rowID) //this would then control which assignment is for which course

Table student_assign_relations (student_ID, acr_rowID, Grade)

---------------
Is this the correct way of going about it ?

Any advice is appreciated TIA

here is an exercise for you:

try to see how many of those tables make sense if you do not use an "id" but rather use some other identifying unique key for each table

Sorry I have no clue what that means, I'm not really a database person, I can just fudge my way through it on the best of days. (come downtown to Ryerson and show me what you mean )

to answer your original question ("Is this the correct way of going about it ?") you are very very close

about coming down to ryerson, that's a wonderful idea, contact me via my site and maybe we can set something up