I suppose the real question here is, "What exactly makes a quality reference, and when should I use one?"
In general, if you're quoting actual facts (e.g. ahadith) those don't need an actual reference beyond the citation of the hadith itself, even if you read about these hadiths from a third-party. However, facts alone rarely if ever speak for themselves; they need to be understood, interpreted and applied to actually answer the question, otherwise they're just items. If you personally lack the ability and/or expertise to do so yourself, but base your answer on the interpretations and opinions of someone who has said ability/expertise, then it is incumbent on you to reference that work.
However, the quality of any given reference still requires a critical analysis of that reference: Why should any future readers believe the opinion of the guy you quoted is any more useful than your own? Or any random person on the Internet's, for that matter?
A common fallacy (especially on this site, I find) is the Appeal to Authority, which presumes that just because an argument is made by an expert (or a perceived expert), it's automatically and unassailably right. And any answer that is fundamentally fallacious is pretty much by definition a bad answer.
In other words, just quoting a reference may make your answer appear more valuable, but it's really no better than if you just voiced your own personal opinion. Regardless of source, such a reference is still essentially an item, and again if your answer relies on this item being valuable but you personally lack the ability and/or expertise to critically analyse and present it as such, it is again incumbent on you to refer to an authority that has that ability/expertise.
It's turtles all the way down, really, until you reach a point where you, personally, can answer according to your own experience and authority. Blindly citing and summarizing references is no better than copy-paste; if you don't understand what you've posted well enough to defend it if/when called on it, you probably shouldn't be posting at all.
In short, one cannot give a blanket "All references of this type are good/bad," whether it's analysing the primary sources directly, or analysing other people's analysis of the primary sources, or analysing other people's analysis of other people analysing the primary sources; the quality of any given reference needs to be understood individually to determine its value, and the onus of this is on the person who posts it. If you don't know if the reference you're citing is "a true reference" then you're better off not using it as any sort of basis for your answer until you've (a) studied it enough to actually recognize its value or (b) found a better reference.