I'm currently completing an assignment for my linear algebra class, and I feel that I don't completely understand the dimension of an orthonormal basis. For reference, this is the question I am dealing with:

Let W = {(2a+4c, 2b-c+d, 2b-2c+2d, 2a+5c-d, a+b+c+d) ∈ $\mathbb R^5$ | a, b, c, d ∈ $\mathbb R$}

a) Find an ordered basis for W. What is the dimension of W? What is the dimension of W⊥?

b) Find an orthonormal basis for W.

I already proved that W is a subspace of $\mathbb R^5$. Furthermore, I was able to obtain an ordered basis for W of

B = {(2,0,0,2,1),(0,2,2,0,1),(4,-1,-2,5,1)}

If this is a correct basis, then obviously dim(W) = 3. Now, this is where my mistunderstanding lies.

Using the Gram-Schmidt Process to find an orthogonal basis (and then normalizing this result to obtain an orthonormal basis) will give you the same number of vectors in the orthogonal basis as the original basis. This is according to my book. So, does this mean that dimW⊥ = dim(W)? I read somewhere else online, however, that dimW + dimW⊥ = dim(V), where W is a subspace of V. So, is dim(W⊥) = 3? Or is it 5-3 = 2? (As W is a subspace of $\mathbb R^5$). I am clearly misunderstanding a concept here. Any help would be greatly appreciated!


You are simply confusing the orthonormal basis for W which can be found by GS process starting by the basis you already have found (are you aware about that) with the subspace orthogonal to $W$ and indicated with $W\perp$ which has dimension equal to $n-dim(W)$.

  • $\begingroup$ I see. So the dimension of W⊥ is not the same as the number of vectors in the orthogonal basis? I thought that the two were essentially the same thing. Am I wrong to say that the orthogonal basis for W found by G-S Process spans the subspace orthogonal to W? $\endgroup$ – Melissa Mar 21 '18 at 1:13
  • $\begingroup$ @MahaEssa Yes, you are wrong to say that, as gimusi has already pointed out. The basis produced by the Gram-Schmidt process has the same span as its inputs. $\endgroup$ – amd Mar 21 '18 at 1:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.