Response surface methodology (RSM) was applied to optimize the composition of a mixture of hydrocolloids (low-viscosity sodium alginate, high-viscosity sodium alginate, and kappa-carrageenan) used in the immobilization of beta-galactosidase for application in the hydrolysis of lactose. A five levels, a three-factor design was adopted. The activity and stability of the immobilized enzyme and the strength of the supporting matrix were optimized by using a mathematical model applied in the range of process conditions. Requirements for multifactor response surface designs were satisfied, and the correlation coefficient, R(2), was larger than 0.850, ensuring a good adjustment of the model to the experimental values. Best results were obtained for values of 1.00% low-viscosity sodium alginate, 1.40-1.60% high-viscosity sodium alginate, 0.10-0.30% kappa-carrageenan, and 10-12% enzyme.