Recreational Off-Highway Vehicles (ROVs), since their introduction onto the market in the late-1990s, have been related to over 300 fatalities with the majority occurring in vehicle rollover. In recent years several organizations made attempts to improve ROV safety. This paper is intended to evaluate ejection mitigation measures considered by the ROV manufacturers. Evaluated countermeasures include two types of occupant restraints (three and four point) and two structural barriers (torso bar, door with net). The Rollover protection structure (ROPS) provided by the manufacturer was attached to a Dynamic Rollover Test System (DRoTS), and a full factorial series of roll/drop/catch tests was performed. The ROV buck was equipped with two Hybrid III dummies, a 5th percentile female and a 95th percentile male. Additionally, occupant and vehicle kinematics were recorded using optoelectronic stereophotogrammetric camera system. Two distinctive rollover conditions (200deg/s and 360deg/s at 3.5-5g vertical drop acceleration) used to simulate vehicle rollover were extracted from the field testing presented in the literature. The results show that for both rollover conditions the unbelted occupants experienced the largest excursion from the vehicle which resulted in occupant ejection from the ROV. In unbelted tests Anthropomorphic test devices (ATDs) came into contact with ROPS or structural barrier (door) and in three out of eight unbelted tests sustained high neck loads (Nij = 1.09-1.35). Seatbelt use eliminated complete occupant ejection, but in several cases partial ejection was observed. Occupant size, occupant restraints, and structural barriers used in respective tests influenced the kinematic response of ATDs during rollover.