At risk of pointing out the obvious, just about every early text of significance was oral tradition of some shape or form put in writing, with varying degree of research and reinterpretation on top -- not unlike, let's face it, oral tradition itself. Think the Bible, Herodotus, etc. all the way to the Middle Ages when sources become common enough that you're able to continually corroborate what happened through other sources. (And even those correlate significantly, because as the saying goes, history is written by victors.)
In that light, it's a perfectly valid source. Capturing the oral tradition is, of course, an important part of the value of the work if you go down that path. Putting it in context is another. It's an even more important one if anything: what is said? who is narrating? why? what might that tell you? what's omitted? why? what might that tell you? etc.
Assuming you're looking for a source that sketches out how to use such sources, and that you're open to learning by example, I found Wickham's "The Inheritance of Rome" extremely dry but thoroughly illuminating in this respect -- especially in the early chapters, where he spends enormous amounts of time discussing the merits of the sources he's referencing. Surely you'll be able to find inspiration on how to use equivalently dubious sources for your own research topic.