Even a measurement error is at first: evidence
Yes, in a non-technical sense there is evidence of something. In the technical sense (see for example Bayesian Probability for the technical meaning used in the hard sciences) there is still no evidence that indicates the EMDrive works as advertised. If there had been such evidence then scientists across the world would flock to reproduce the results of the successful experiment just like people rushed to reproduce the results of Pons and Fleishman.
A more precise statement might be to say that even though all of the experiments reported some unexplained thrust, there is no agreement between the experiments and none of the experiments have been able to show a clear signal or lack thereof above the noise floor. Contrast this with the Pons and Fleishman experiment which did show a clear signal way above the noise floor (which ended up being non-reproducible) or the CERN experiment which showed that neutrinos traveled faster than light, again way way above their noise floor. Those CERN experimenters had the honestly and humility to say they didn't think their clear signal was real because the experiment was very complicated and there was probably something in the experimental setup they were not accounting for. The reason they said this is because if their clear signal had been real then it would have thrown a huge monkey wrench into established theoretical physics. It turns out that they eventually found their mistake, in one place in the experimental apparatus a longer cable was used instead of the short one they assumed was used. This caused an extra delay in their measurement of the light signal and thus cause the erroneous results.
If any of the EMDrive experiments provided real, scientific evidence of the EMDrive mechanism working then the scientific world would be in an uproar like they were after Pons and Fleishman. For the EMDrive experiments thus far we have the worst of both worlds. On the theoretical side, the EMDrive would upturn the world of theoretical physics much more than the faster than light neutrinos; on the experimental side, all the experimental results (except the refuted ones from China) are consistent with there being no EMDrive effect at all. This is what I meant. I didn't mean there was no evidence, I meant there was no evidence that the EMDrive actually works as advertised.