Electron–cyclotron waves are a tool commonly used in tokamaks, in particular to drive current. Their ability to drive current in a very localized manner renders them an optimal tool for MHD mode mitigation. However, such applications require high accuracy and good control of the power deposition location to efficiently target the magnetic islands. It has been indirectly observed that the suprathermal electron distribution, resulting from the wave absorption, is broader than what is expected from experimentally-constrained forward drift-kinetic modeling. The present paper explores the possibility that beam scattering through the turbulent edge of the plasma may explain this observed discrepancy. In particular, full-wave studies exhibit three beam broadening regimes, from superdiffusive to diffusive, with an intermediate regime characterized by a Lorentzian beam profile with a slightly increased full-width at half maximum with respect to the quiet plasma case. In the tokamak à configuration variable, dedicated plasma scenarios have been developed to test this hypothesis. A realistic worst-case fluctuation scenario falls into this intermediate beam broadening regime. By comparing the experimental hard x-ray emission from suprathermal electron Bremmstrahlung with the emission calculated by coupling a full-wave model to a Fokker–Planck solver, it is shown that, in the tested cases, the beam broadening is not sufficient to explain the aforementioned discrepancy between simulation and experiment and that another mechanism must play the main role in broadening the suprathermal electron distribution.