Abstract The additive model of shear strength of reinforced concrete (RC) members, i.e. shear strength (V) is equal to the sum of the contribution from concrete (Vc) and that from transverse reinforcement (Vs), has been widely accepted in the literature and engineering practice. Shear span-to-depth ratio (a/d) is known to be a significant factor affecting V of RC members. However, very few quantitative studies on the influence of a/d on Vc and Vs have been reported in the literature. Another issue is related to the controversial relationship between the shear force at first diagonal cracking (Vcr) and Vc, for which different guidelines are given in ACI, AASHTO LRFD and CSA codes. Through direct measurement of Vc and Vs from 11 RC beam tests, this work provides experimental evidence for these issues. The experimental results show that Vc can be very different from Vcr. At small shear span-to-depth ratio (a/d), Vc is much larger than Vcr, while it is the opposite for beams with a large a/d value. Not all stirrups crossing the critical shear crack yield at ultimate shear strength, and Vc as well as Vs are not constant under increasing member deformation. Although design codes give a conservative prediction of V, they predict an un-conservative value of Vc at large a/d.