General Relativity "Tests" Not Unique
Presenting evidence that is consistent with a theory doesn't prove the theory to be true if alternative explanations exist that are equally compatible. Singling out one explanation as being correct requires a crucial test that can be reconciled with that explanation alone, thereby falsifying the others.
It is generally asserted that Einstein's Theory of General Relativity, published in 1915, has been incontrovertibly established as fact by its unique ability to account for the measured precession of Mercury's perihelion, and otherwise inexpicable predictions of gravitational redshift and the gravitational deflection of light rays. However, in an intriguing paper entitled "Some Comments on the Tests of General Relativity," posted at http://arxiv.org/abs/0806.0528, Dr. Jeremy Dunning-Davies shows that all of these effects follow from classical considerations without any recourse to General Relativity at all. An expression identical to Einstein's for the precession of Mercury's perihelion, for example, was obtained by Paul Gerber, a German schoolteacher, from classical principles that allowed for a retarded gravitational potential, in 1898.
Dr. Dunning-Davies's papers are collected at http://arxiv.org/find/all/1/au:+Dunning_Davies/0/1/0/all/0/1 and include such other titles as: "A review of redshift and its interpretation in cosmology and astrophysics"; "Reflections and thoughts on tired light"; and "A re-examination of Maxwell's electromagnetic equations."