We analyze the impact of turbulent metal diffusion on observables related to chemical evolution for simulated dwarf galaxies (\mstar $\sim$ 7 $\times$ 10$^{5}$ - 2 $\times$ 10$^{8}$ \msun) from the Feedback in Realistic Environments (FIRE) project and the associated Latte project. In particular, we study the effects of diffusion on the metallicity distribution functions and $\alpha$-element abundance patterns for both isolated dwarf galaxies and satellites of a Milky Way mass galaxy. Our implementation of enhanced metal mixing is based on a turbulent model of diffusion driven by Kelvin-Helmholtz instabilities. Compared to simulations without diffusion, we observe a narrowing of both the metallicity distribution function and abundance ratio distributions, which results from individual particles being driven toward the average metallicity. This effect results in better agreement with observations of Local Group dwarf galaxies. The inclusion of metal diffusion more accurately reproduces the chemical evolution of low-mass galaxies, further justifying the use of simulations as analogs in detailed studies of galaxy formation and evolution.