The purpose of the study was to investigate which form of straight leg raise exercise might be most effective for strengthening vastus medialis oblique: performing the exercise with neutral, internal or external rotation of the hip. A single group, cross-sectional, repeated measures study design was selected. Since the relationship between isometric muscle tension development and surface EMG amplitude has been shown to be both near linear (Basmajian and De Luca, 1985) and highly repeatable (Weir et al, 1992), bipolar electromyography was used for this investigation. Electrical activity of the vastus medialis oblique muscle was monitored while subjects were lying in a supine position, performing straight leg raise exercises with three different angles of hip rotation (neutral, internal and external rotation), with and without ankle weights. Data from 30 male healthy subjects were analysed by repeated measures with three independent variables (leg dominance, hip rotation angle and ankle weight) and three interactions (hip rotation angle and ankle weight, leg dominance and hip angle, leg dominance and ankle weight). Leg dominance had no effect on the EMG amplitude signal. However, there was a significant effect of hip rotation angle and the ankle weight in the straight leg raise (p < 0.05). Significant differences were found only between the external rotation and the neutral hip position or the internal rotation (p < 0.05). All the interaction effects were not significant (P > 0.05). Performing a straight leg raise exercise with external hip rotation was the most effective of the three techniques as a specific strengthening exercise for vastus medialis oblique. This finding may be particularly relevant for therapists or trainers who are interested in a strengthening method for people with weak or under-conditioned vastus medialis oblique muscle, such as those with patello-femoral pain syndrome.