No just expressing a preference.
The question really should be: When moral claims are made, i.e., racism is wrong, murder is wrong, what do these claims express?
It's fully possible that while moral claims express factual propositions, Ivaj is an incompetent English speaker and uses moral language to express his own preferences. The non-cognitivist has to give an account of moral language in general, not just of what
they are saying when they speak about morality. So the question really is: What are competent English speakers saying when they make moral judgments?
And the answer to that question is clearly not an expression of preference, since it's possible to make moral judgments without expressing a preference. E.g., "There is a right thing to do, but I don't know what it is", "that's the wrong thing to do, but I honestly don't care", etc. That these sorts of statements are coherent, and not conceptually inconsistent, proves that moral judgments are not mere expressions of preference.