This dissertation examines the extent to which a U.S.-developed theory of teacher knowledge, Mathematical Knowledge for Teaching (MKT), is applicable in a Ghanaian context. To address the problem of Ghanaian students' poor student achievement in mathematics, caused in part by the quality of teaching, this dissertation is grounded in the premise that teacher knowledge influences teaching quality and consequently impacts students' learning. Progress made on MKT in the U.S. to help advance the quality of teachers' training motivates this study to examine whether the theory and measurement of MKT can support similar improvement of teaching quality in Ghana. This study focuses on the question of the transferability of the MKT concept and its measures.

Specifically, this study is guided by the question: "To what extent can empirically derived U.S.-developed measures of MKT be used to study MKT held by a sample of primary teachers in Ghana." In particular, do the U.S.-developed Ghanaian-adapted MKT measures validly measure MKT in Ghana? I address this with the two sub-questions: a. What is the relationship between teachers' MKT scores and their reasoning about their responses to the adapted MKT measures? b. What is the relationship between teachers' MKT scores and the mathematical quality of their instruction?

This study first adapts U.S-developed measures of MKT to make them usable in Ghana, without altering their substantive content. The adapted measures were then administered to 60 conveniently sampled practicing teachers. Three fourth-grade teachers among them were selected for in depth analysis to examine the validity of their MKT scores in two ways. First, they were interviewed to determine the consistency of their reasoning with their mathematical knowledge as assessed; and second, two consecutive mathematics lessons are analyzed for the mathematical quality of the instruction. Findings from the three teachers were extended to three sixth-grade teachers to determine the consistency of these findings.

Results from the data analysis show that although the MKT construct is valid in principle in the Ghanaian context, there is strong evidence to suggest that the instruments that assess MKT and the mathematical quality of instruction need further adaptation to suit the Ghanaian context.