This article presents an experimental study aiming to provide reliable experimental data for the prediction of the shear strength of reinforced concrete beams without stirrups. The shear strength results obtained from the proposed model as well as those from the design codes (ACI 318, Eurocode 2 and BS 8110) are compared with the database containing 700 beams made of both high strength concrete and normal strength concrete. Furthermore, the experimental results were used to assess the contribution of each shear mechanism obtained from Chen’s, Cavagnis’s and Fernández Ruiz’s models. The contribution of these various shear-transfer actions is also quantified experimentally, using a digital image correlation. As a result, the measured contributions of dowel action, aggregate interlock and compression zone to the total shear resistance were estimated as 45% to 50%, 20% to 35% and 17% to 31%, respectively, for high strength concrete beams. On the contrary, the average test-to-predicted contribution of the shear-transfer action ratio determined by the Chen formula is 1.15 for short beams, whereas the Fernández Ruiz and the Cavagnis models yielded average ratios of 1.04 and 1.52, respectively, for slender beams. The proposed formulae give a rational prediction for either both short and slender beams, and yield accurate and consistent results compared to the other models used in this study, with a lowest average value of the test-to-predicted at 1.08 and that of the coefficient of variation at 23.06%, particularly for short beams. However, ACI 318 is the only code that does not take into account the size effect, leading to a severe underestimation of the shear strength for short beams.