A bstract We analyzed the evolution data of the Daya Bay reactor neutrino experiment in terms of short-baseline active-sterile neutrino oscillations taking into account the theoretical uncertainties of the reactor antineutrino fluxes. We found that oscillations are disfavored at 2.6 σ with respect to a suppression of the 235 U reactor antineutrino flux and at 2.5 σ with respect to variations of the 235 U and 239 Pu fluxes. On the other hand, the analysis of the rates of the short-baseline reactor neutrino experiments favor active-sterile neutrino oscillations and disfavor the suppression of the 235 U flux at 3.1 σ and variations of the 235 U and 239 Pu fluxes at 2.8 σ . We also found that both the Daya Bay evolution data and the global rate data are well-fitted with composite hypotheses including variations of the 235 U or 239 Pu fluxes in addition to active-sterile neutrino oscillations. A combined analysis of the Daya Bay evolution data and the global rate data shows a slight preference for oscillations with respect to variations of the 235 U and 239 Pu fluxes. However, the best fits of the combined data are given by the composite models, with a preference for the model with an enhancement of the 239 Pu flux and relatively large oscillations.