Let's try to get at the implicit assumptions (well, corresponding issues that we should get on the table) undergirding the question:
1. Are species homogeneous enough to be evaluated in aggregate? If so, when evaluating whether a species is 'good' or 'bad', which characteristics shared among individuals will be pertinent?
2. Good/Bad toward what end? Which characteristics or capacities for action do we care about and why?
3. Are goodness and badness to be found in overt, observable action? We can hardly get into an animal's mind and infer (with reliability) the pertinent 'motives', desires, influences, etc. (yes, language, observed expressions, intuition, etc. are most imperfect, but the provide information worlds better than our meagerly educated guesses about how animals experience).
And then, as we'd do whenever exploring ethics:
1. What is the ontological standing of "the good" or "the bad"? In what 'stuff' of reality are ethics to be found?
EGs:
What weight should be given to consequences of acts, intentions behind acts, generalized maxims that sometimes guide acts, personal characteristics etc.?
When looking for the good and the bad, should we turn attention to feelings of pleasure and pain, expressions of freedom, development of rich potentialities in individuals (self-actualization), 'fairness', arbitrary preferences etc. ?
What is the scope of ethics? Do we allow universal maxims, more particular ethical frameworks developed in particular contexts, of particular communities and situations, overriding quantitative logic to assess quantities of good and bad (yes, utilitarianism), or is it just arbitrary whims?
Anyway....yeah. big can 'o' worms...and I lack the patience for it right nao.
