Near-surface turbulent kinetic energy dissipation rates are altered by the presence of sea ice in the marginal ice zone, with significant implications for exchanges at the air-ice-ocean interface. Observations spanning a range of conditions are used to parameterize dissipation rates in marginal ice zones with relatively thin, newly formed ice, and two regimes are identified. In both regimes, the turbulent dissipation rates are matched to the turbulent input rate, which is formulated as the surface stress acting on roughness elements moving at an effective transfer velocity. In marginal ice zones with waves, the short waves are the roughness elements, and the phase speed of these waves is the effective transfer velocity. The wave amplitudes are attenuated by the ice, and thus, the size of the roughness elements is reduced; this is parameterized as a reduction in the effective transfer velocity. When waves are sufficiently small, the ice floes are the roughness elements, and the relative velocity between the sea ice and the ocean is the effective transfer velocity. A scaling is introduced to determine the appropriate transfer velocity in a marginal ice zone based on wave height, ice thickness and concentration, and ice-ocean shear. The results suggest that turbulence underneath new sea ice is mostly related to the wind forcing and that marginal ice zones generally have less turbulence than the open ocean under similar wind forcing. Plain Language Summary The rate of mixing near the ocean's surface determines the rate of exchange of material, heat, and energy between the air and the ocean. Determining the rate of mixing under all conditions is necessary to be able to make global estimates of exchange, such as how much carbon dioxide is going into the ocean. In the open ocean, this rate can be predicted using measurements of the wind. In oceans where sea ice is present, the rate is lower. The current methods for predicting the rate of surface mixing in sea ice are too simple and do not well describe the range of conditions that occur. We propose that this rate can be predicted in thin, new sea ice using measurements of waves, wind, and the ice. When ocean waves are present, the rate of mixing is predicted to increase with increasing wave height and decrease with increasing ice. When there are no waves, we can predict the approximate rate of mixing due to wind stress using the speed of the ice relative to the ocean. Under all conditions, the rate of mixing in sea ice-covered oceans is expected to be lower than in open ocean with the same wind.