IF you believe 'closure' of addition and scalar multiplication is false, yes, you could look for values of x and y which make the statement untrue. That is, look for a "counter-example".
However, if it is really a subspace, closure of addition and scalar multiplication must be true for all x and no matter examples you tried, that would never prove it true for all.
Instead use algebraic theorems to show the set is closed under addition and scalar multiplication. For this paraticular example, two "generic" members of the set would be of the form (2x+ 3y, x, 0, -x+ 2y) and (2x'+ 3y', x', 0, -x'+ 2y'). The sum of those is ((2x+ 3y)+(2x'+ 3y'), x+ x', 0+ 0, (-x+ 2y)+ (-x'+ 2y'))= (2(x+x')+ 3(y+ y'), x+x', 0, -(x+ x')+ 2(y+ y')). Do you see that this is in the same subset? It must be of the form (2x"+ 3y", x", 0, -x"+ 2y"). What are x" and y" in this case?
To see that it is closed under scalar multiplication, take a general member of the set, (2x+ 3y, x, 0, -x+ 2y) and multiply by the scalar "r":
(r(2x+ 3y), rx, r0, r(-x+ 2y))= (2(rx)+ 3(ry), rx, 0, -(rx)+ 2(ry)).
Again, do you see that is of the form (2x"+ 3y", x", 0, -x"+ 2y")? What are x" and y" here?
To find a basis, note that any member of this set is of the form (2x+ 3y, x, 0, -x+ 2y)= (2x, x, 0, -x)+ (3y, 0, 0, 2y)= x(2, 1, 0, -1)+ y(3, 0, 0, 2).