The criterion asks for a range of structures used accurately, and it is read as “write longer sentences” more often than any other instruction in the exam. The result is predictable: writers stack subordinate clauses onto sentences that were working, and the errors they introduce cost more than the range they gained. A response of accurate simple and compound sentences scores better than one of ambitious complex sentences with agreement errors in them.
Each shape does something the others cannot. A simple sentence lands a point — it is the right structure for a thesis, and for the sentence you most want the reader to remember. A compound sentence balances two things of equal weight. A complex sentence subordinates: it says explicitly that one part matters more than the other, which is why it is the natural shape for a concession or a cause.
The practical consequence is that variety follows from function. If your paragraph has a claim, a reason, a concession and a consequence, the sentences will vary on their own, because those four jobs do not fit the same shape. Variety pursued directly — long, short, long, short — produces sentences whose form has nothing to do with their content, and readers feel that even when they cannot name it.
One clause, no subordination. The right shape for a thesis and for the line you want remembered.
Two independent clauses joined by “and”, “but”, “so”. Equal weight on both sides.
One clause subordinated to another. It says which half matters more, which is why concessions take this shape.
An accurate simple sentence beats an ambitious complex one with an agreement error in it.
