The objective of this paper is to investigate the implications of the load separation criterion for evaluating ductile fracture mechanics parameters. This criterion allows the load to be represented as the multiplication of two separate functions; a material deformation function and a crack geometry function. Load separation implies a method for J-integral evaluating using only a single load-displacement record. The original method for evaluating J, proposed by Begley and Landes, used the energy rate interpretation of Rice which requires several load-displacement records for identical specimens with varying crack lengths. A method based on load separation introduced a new parameter η, ηel and ηpl which greatly simplified J calculation. This parameter which can be a function of geometrical factors is generally evaluated experimentally using the energy rate interpretation of J. In this paper the load separation criterion is used to imply a simple method for evaluating η experimentally. Using blunt notched specimen load versus load point displacement results from the literature, four different configurations with a wide range of stationary crack lengths are evaluated. Also included are several different materials varying from low work hardening to high work hardening. The data include thin and thick sections so that both plane stress and plane strain conditions are evaluated. A new method for η-estimation derived from the implication of the load separation is proposed. This method avoids most of the errors that accumulate in the classical methods of estimation. Both the separation method and the energy rate method are evaluated by comparing the techniques and the results. The results show some new trends in ηpl results for the different configurations evaluated in this paper.