The objective of this study was to determine the test-retest reliability of the ‘Y-Balance Test’, a new high level test involving balancing on three directions consist of anterior, posteromedial and posterolateral directions. The ‘Y-Balance Test’ consists of a series of lower extremity reaching tasks in above mentioned three directions that challenge subjects’ postural control, strength, range of motion, and proprioceptive. A total of 13 primary schools (5 from Malay streams, 4 Chinese streams and 4 Indian streams) were randomly selected from 971 primary schools in Perak State, Malaysia. There were 351 healthy Malaysian primary school boys and girls age ranging from 7 to 12 years old being recruited using simple randomization. A pilot study was conducted on 24 subjects to find out the test-retest reliability of the ‘Y-Balance Test’ before the actual study being administrated (anterior ICC=0.96; posteromedial ICC=0.96, and posterolateral ICC=0.98). The test-retest reliability was assessed with two trials for each testing session. In this study, the test-retest Intraclass Correlation Coefficients (ICCs) were 0.98, 0.98, and 0.99 on anterior, posteromedial, and posterolateral directions respectively. The results indicated excellent test-retest reliability for all three directions. In conclusion, the ‘Y-Balance Test’ is reliable in measuring dynamic balance on Malaysian primary school children. Therefore, the ‘Y-Balance Test’ revealed excellent reliability as a dynamic balance measurement tool to assess the dynamic balance ability among primary school children. Children with high dynamic balance ability are believed could outperform in sports, games and other related physical activities.