The study explored the effect of cooperative learning on Grade 12 learners' performance in projectile motions. A quasi-experimental research design with non-equivalent control group was used. Two schools were purposively selected from Maleboho Central circuit in South Africa based on their performance in Physical Sciences Grade 12 results of 2011. The sample consisted of 49 learners from two schools. School A was used as the Experimental Group (EG) and was taught using cooperative learning technique while school B was the Control Group (CG) taught using traditional teaching methods. Pre- and post-tests were used to collect data. Data were analysed using descriptive statistics: arithmetic means and standard deviations; and inferential statistics: independent student t-test, Analysis of Covariance (ANCOVA) and Cohen's d. Results show that EG outperformed the CG suggesting that cooperative learning technique enhanced learners' performance more than the traditional talk-and-chalk teaching approach.