Full Idea
It is plausible to think that an explanation, when successful, captures or represents (by argument, or a why? question) an underlying real-world relation of dependence which obtains among the phenomena cited.
Gist of Idea
A good explanation captures the real-world dependence among the phenomena
Source
Kathrin Koslicki (Varieties of Ontological Dependence [2012], 7.6)
Book Reference
'Metaphysical Grounding', ed/tr. Correia,F/Schnieder,B [CUP 2012], p.213
A Reaction
She cites causal dependence as an example. I'm incline to think that 'grounding' is a better word for the target of good explanations than is 'dependence' (which can, surely, be mutual, where ground has the directionality needed for explanation).