A way to see it is that the wedge sum of infinitely many circles is not compact while the Hawaiian earring is compact (It is bounded and it contains its limit points, so closed, and therefore you can use Heine-Borel) (you can even see it as one point compactification of infinitely many intervals.)
Also Hatcher gives a reason using the fundamental group. While in wedge sums you can use Van Kampen Theorem to obtain that the fundamental group is free. In the case of the Hawaiian earring the fundamental group is bigger as he explains.
In what matters to the topology, the wedge sum can be constructed as a quotient space while the Hawaiian earring has the subspace topology from $\mathbb{R}^2$. And they are not homeomorphic. In fact, the topology of the wedge sum of infinitely many circles is finer.
I hope this helps. If you suggest me another approach I could try to give other arguments.
See this:
page of book on google books related
And this article could very helpful:
article
EDIT:
In what matters to the definition, the wedge sum is constructed as a quotient space (you don't need the "ambient" space $\mathbb{R}^2$ and it is a topological space "by its own right") while the Hawaiian earring has the subspace topology from $\mathbb{R}^2$, it is not given a quotient topology. And they are not homeomorphic. In fact, the topology of the wedge sum of infinitely many circles is finer. I mean, the set is the same but the topologies are different.