(Finite dimensional) Linear transforms have matrix representations. These are maps $T:V \to W$ such that $$ T(cv_1) = cT(v_1) $$ and $$ T(v_1+v_2) =T(v_1) + T(v_2) $$ for a constant $c$ and any $v_1, v_2 \in V$.
Since linear maps are special functions from one vector space to another, their compositions are associative just like compositions of any functions.