The instructional and assessment environments that students experience can have an enormous impact on their mathematical success, their understandings of what mathematics is, and their views of themselves as learners and doers of mathematics. While there has been ample research conducted about how teachers use assessment results to inform their instructional practices, the few studies specifically addressing the relationship between the instructional approaches and the types of assessments that teachers use have yielded inconsistent results. The aim of this dissertation was to explore this relationship and how it may be impacted by the racial composition of a class. Systemic racism and white supremacy have an enormous, yet often invisible, impact on all aspects of life in the U.S., which certainly extends to the classroom. Consequently, this dissertation used the Critical Race Theory (CRT) and QuantCrit frameworks to examine whether and how classes with different racial compositions are exposed to differing instructional and testing environments in U.S. mathematics classes. CRT provided this work with a solid theoretical justification for focusing on race and the QuantCrit extension of CRT guided my methodological decisions. This work used nationally-representative data from the 2012 administration of the National Survey of Science and Mathematics Education (NSSME) to explore (1) what, and how strong, the relationships are among teachers' educational backgrounds, areas of instructional emphasis, level of reform-based instructional practices, and testing practices, (2) whether there are distinct groups (profiles) of classes with various combinations of learning objectives and instructional practices and how those groups are distributed across teacher-, class-, and school- level characteristics, (3) whether classes with different racial compositions are given different types of tests, and whether that relationship is maintained when class profile is accounted for, and (4) whether the relationship between instructional practices and test use varies based on class racial composition. The first research question was addressed using structural equation modeling, the second question was addressed using latent profile analysis, and the remaining questions were answered using multinomial regression. The results indicated that reform-based learning objectives, in which teachers focus on developing students' use of mathematical practices rather than on rote learning, were strong predictors of reform-based instructional practices, while traditional learning objectives were not. Teachers who valued mathematical practices and the application of knowledge were more likely to select test formats that allowed students to demonstrate their abilities in these areas, while those who valued computation were more likely to use test questions with limited response options. The latent profile analyses indicated that traditional learning objectives and reform-based learning objectives and instructional practices all tended to move together, such that a class with a high score in one of these constructs was likely to have high scores in the other two. The grade-level of the class and the number of years of experience the teacher had were the strongest predictors of class profile membership. Controlling for class profile membership, however, did not mitigate the strong relationships observed between class racial composition and test use. Classes with large proportions of Black students were most likely to be tested frequently regardless of the class's instructional environment. Finally, classes that received more reform-based instruction were more likely to be tested frequently, specifically with tests that were predominantly short-answer/multiple-choice but did contain some constructed-response questions. Furthermore, the more that teachers with high percentages of Black students in their class used reform-based instructional practices, the more frequently they tested their students. I interpret these results as indicating that teachers with a more student-centered approach to instruction likely want formative information about what their students know and can do, and they are using frequent tests and quizzes to obtain that information. While a student-centered approach to instruction would ideally include the use of curriculum-embedded formative assessments rather than formal examinations, teachers may be constrained by school- or district-wide policies that mandate the frequent updating of online gradebooks, or they may simply not have adequate time or training to implement alternative forms of assessment. The prevalence of exams that are mostly short answer/multiple choice, in particular, may be the result of teachers using commercial testing products that were purchased by their school district, or teachers may be choosing to develop their own exams and opting for item formats that are quick and easy to grade. More research is needed to determine the extent to which classroom testing practices are influenced by the decisions of individual teachers as opposed to by larger policies. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]