Nickel was recovered from nickel laterite using a sulfation-roasting-leaching process and the effects of operation parameters including acid addition, roasting temperature, and roasting time on nickel extraction and iron dissolution were investigated using response surface methodology (RSM). Two second-order polynomial models of high significance were presented to show the relationship between the responses and the variables. The analysis of variance (ANOVA) showed high coefficients of determination (R²) of 0.894 and 0.980 for the two models, respectively. Optimum areas of ≥ 80% Ni extraction and ≤ 5% Fe dissolution were obtained by the overlaid contours. Verification experiments in the optimum areas were conducted and the results indicate a close agreement with the predicted values obtained from the models.