This study examined, through a designed simulation of a performance assessment of a typical teaching act, the possibility of measuring elementary preservice teachers' mathematics content knowledge without relying solely on the candidates' performance on a standardized mathematics test. Data collection involved a series of disclosed mathematics problems designed by the National Council of Mathematics, which were used as prototypes for the National Assessment of Educational Progress (NAEP) tests. The mathematics problems were prepared for the fourth grade NAEP assessments. In this study, 24 fifth graders were required to solve six two-step problems, show their work, and explain how they solved them. The same problems were given to 69 preservice elementary teachers to examine their mathematics knowledge. They had previously taken the standardized test of mathematics skills. The preservice teachers were then given the students' responses to the problems and directed to analyze them and feedback to each fifth grader. The feedback was categorized into four types (general, specific, student performs, or teacher performs). Results indicated that the correlation between the standardized test (as the test of preservice teachers' knowledge of mathematics) and the quality of the feedback variable (the performance assessment) showed a positive but low relationship. (SM)