Teachers' Sense of Efficacy Scale (TSES), since its development, has gained much popularity because of its adequate psychometric properties. Yet, scholarly information on the soundness of this instrument in the African context has not been documented. Besides, the previous validation studies on the TSES have yielded mixed results with different factor structures, calling for further studies to test its utility and use across different cultures. The purpose of this study was to revalidate the TSES in Ghana using item response theory (IRT) complemented by factor analyses. Through a validation study approach, 693 basic school teachers were purposively sampled to participate in the research by responding to the Teacher Efficacy Scale. The between-item and within-item multidimensional IRT analyses combined with factor analysis were performed to understand how the items functioned with intentions to propose a well-fit instrument within the Ghanaian context. The outcome of this research revealed that the original TSES with a three-factor structure, 24 items measured on a 9-point scale did not provide an optimal measure of teachers' self-efficacy within the Ghanaian context. Rather, a three-structure scale with 18 items and a 6-point scale offered a maximal measure of teachers' self-efficacy. The practical implications of the findings are discussed.