ABSTRACT Traditional approaches to the strengthening of dimensional wood roof joist systems are often intrusive and require a substantial amount of additional labour to remove ceilings or roof sections. In an effort to design a less intrusive system for reinforcing over-stressed dimensional wood joists, a roof joist strengthening system was designed consisting of a tension steel plate installed along the bottom of the existing joists with bolted side plate connections. An experimental program was conducted at the University of Manitoba to assess the viability of the reinforcement design. A total of 15 joists were tested to failure under four-point bending conditions, consisting of 5 unreinforced control joists, 5 joists with steel reinforcement, and another 5 reinforced joists with an artificial defect introduced at mid-span along the tension side. The purpose of introducing an artificial defect at mid-span was to simulate the failure modes observed at R.W. Bobby Bend School in Stonewall, Manitoba, where certain roof joists contained knots, splits or cracks along the bottom of the joist within the mid-span region. Service loads based on the applied snow and dead loads applied to these joists were calculated using the National Building Code of Canada [1] and were approximated in the experiment as 8.5 kN. The results indicated that the steel reinforcement produced a 46.8% increase in average load-carrying capacity compared to control joists, and a 33.1% increase in capacity for reinforced joists with artificial defects. The average deflection at service loads was 19.5% lesser than the deflection of the control joists, and the apparent stiffness was determined to increase by 50.0%. The results from this research support the conclusion that the steel reinforcement system for dimensional wood joists is a viable alternative to traditional systems that effectively increases the load-carrying capacity, stiffness, and ductility of the structure.