Abstract In two-dimensional experimental setups, tip-flow cannot be eliminated completely. In one degree-of-freedom Flow Induced Motions (FIM) of circular cylinders placed perpendicular to a uniform flow, three-dimensional effects may become significant. An ideal setup extends the cylinder to the limits of the flow-channel to minimize tip vortices, which reduce the effective length of the cylinder. Depending on how close to two-dimensional the experimental setup is, obtained results may differ. It is difficult to avoid the tip-flow in nature as well. Applications involving Vortex-Induced Vibrations (VIV) have more or less three-dimensional flow characteristics and one of the manifestations of three-dimensionality is the tip-flow. In this paper, the effects of tip-flow on VIV are investigated both experimentally and computationally. It is found that the tip-flow reduces the lift force exerted on the cylinder and narrows down the range of synchronization. Two-dimensional computational simulations become insufficient to grasp the effects of the tip-flow for a cylinder in VIV as the Reynolds number increases. Computational results for vortex-induced vibrations at these relatively high Reynolds numbers (up to 1.2 ⁎ 10 5 ) in the TrSL3 flow regime are not satisfactory when compared with experimental results. To improve the CFD predictions by introducing three-dimensional (3D) flow characteristics in a two-dimensional (2D) computational environment, a parameter called tip-flow correction factor is defined and analyzed. This parameter is introduced to compensate for any deviations from 2D flow approximation that might arise due to the 3D nature of the flow. The tip-flow correction factor is implemented as a multiplier of the force term in the vibration equation to represent the lift-force losses caused by the tip vortex. When compared to the results obtained with straightforward use of the vibration equation, it is found that the tip-flow correction factor improves the agreement between 2D computational results and experimental measurements. This method extends the validity of 2D-URANS simulations at least up to Re = 1.2 ⁎ 10 5 for which experimental results are available in this study.