We study the question of the gauge dependence of the quantum gravity contribution to the running gauge coupling constant for electromagnetism. The calculations are performed using dimensional regularization in a manifestly gauge invariant and gauge condition independent formulation of the effective action. It is shown that there is no quantum gravity contribution to the running charge, and hence there is no alteration to asymptotic freedom at high energies as predicted by 11.10.Gh, 11.10.Hi Until fairly recently the belief that quantum gravity lay outside the realm of any conceivable experimental test was pervasive in the physics community. Important work by Donoghue [1] helped to change this pessimistic viewpoint. Donoghue proposed that the effective field theory methodology be applied to quantum gravity, with Einstein's theory viewed as an effective low energy approximation to some as yet unknown more complete theory. Since Donoghue's pioneering work, there has been considerable interest in this viewpoint. A beautiful review of the subject has been given by Burgess [2].A notable calculation was performed recently by Robinson and Wilczek [3] for Einstein gravity coupled to a gauge theory. It was claimed, as a consequence of quantum gravity corrections, that all gauge theories become asymptotically free at high energies. This includes the Einstein-Maxwell theory, and occurs below the Planck scale at which perturbative quantum gravity calculations become suspect. If the gravitational scale is sufficiently low, as predicted by many higher dimensional theories, then it is conceivable that the predictions of [3] on the running gauge coupling constant might be experimentally testable [4]. A recent analysis of the calculation [5] has cast doubt on the results of [3] by claiming that the quantum gravity correction to the running gauge coupling constant is gauge dependent, and that there is really no such effect. The calculations of [5] do demonstrate that when computed using traditional background-field methods the effective action does depend on the choice of gauge condition. Choosing a particular gauge condition, and demonstrating independence of parameters that enter this condition is not sufficient to demonstrate gauge condition independence. As we will discuss, with the choice of gauge conditions made in [3, 5] a gauge condition independent result for the effective action necessitates additional terms that are not present if the standard background-field method is employed. Because of the interest in this problem, its potential as an experimental test of quantum gravity, and the controversy surround- * URL: http://www.staff.ncl.ac.uk/d.j.toms ; Electronic address: d.j.toms@newcastle.ac.uk ing the details, we will present a different analysis to that of [3,5] that removes all question of any possible gauge dependence.There are two basic problems that need to be dealt with in any calculation in a gauge theory. The first is to ensure that the calculations are done in a way that is gauge invariant. The second is ...