Simulated pseudo-second-order kinetic adsorption data were analyzed by different methods of least-squares regression. The methods used were non-linear regression and four linearized forms of the pseudo-second-order equation. The simulated data were compromised with three different homoskedastic and heteroskedastic error distributions. In the presence of all types of error distributions, non-linear regression was the most robust method and provided the most accurate and efficient estimates of the kinetic parameters. (C) 2010 Elsevier B.V. All rights reserved.