Several intraocular lens (IOL) power calculation formulas (either theoretical or empirical) are used to determine the emmetropic IOL power) The Sanders-Retzlaff-Kraff (SRK) linear regression formula is among the most widely recognized empirical ones. In the present study intraocular lens power calculation aiming at emmetropia was performed, using SRK formula, in 145 cataractous eyes undergoing lens implantation. The final refraction was evaluated at 8 to 12 months after surgery. The purpose of this study was the identification and quantitative evaluation of the factors which influence significantly the accuracy of SRK in the intraocular lens power calculation. The following factors were studied: (1) the error in preoperative biometry with regard to the difference between post and preoperative axial length measurements, (2) the position of the implantation of the intraocular lens (anterior versus posterior chamber), (3) the intraocular lens style, (4) the intraocular lens power level, (5) the preoperative corneal astigmatism, (6) the surgically induced corneal astigmatism, and (7) the postoperative astigmatism. Multiple regression and stepwise regression analysis showed a strong correlation (R2 = 0.65; p < 0.001) between postoperative refractive error (Rf) and error in preoperative biometry (delta AL), surgically induced corneal astigmatism (SIA) and postoperative astigmatism (Ap) only. This correlation is expressed by the following equation: Rf = 0.07 -2.55 delta AL -0.42 SIA + 0.34 Ap. This equation indicates the quantitative effect of each factor on the accuracy of the SRK formula, by defining the pattern of the fluctuations of the amount or state (myopic or hyperopic) of refractive error induced by changes of variables delta AL, SIA and Ap.