Background/Context: Each year, large shares of students who could do well in Advanced Placement courses and exams--known as AP potential students-- do not participate, particularly students of color and low-income students. There are a number of prevailing reasons, both structural (schools do not offer the courses or teachers do no accurately identify students) as well as student-centered (lack of motivation, conflicts with other activities, or lack of self-efficacy). Purpose/Objective/Research Question/Focus of Study: This study seeks to empirically test these common reasons for foregoing AP participation with the following research questions: How are student and school characteristics related to the probabilities of students attending a high school that offers a corresponding course, enrolling in the course, and taking the exam? To what extent are explanations such as students' constraints on time, lack of motivation, or lack of self-efficacy related to the probability of AP course and exam-taking, net of student- and school-level measures? How well do AP potential estimates align with teacher recommendations into advanced coursework? We focused on differences across race and class throughout. Research Design: We define AP potential as a 60 percent probability or better of receiving at least a 3 on an AP exam in either math or English. Using a nationally representative sample of sophomores in 2002 whom we identified as having AP potential, we answered the first research question with a sequential logit. We then used post-estimation commands in Stata to examine motivation, hours working, hours in extracurricular activities, and measures of English and math self-efficacy to address the second research question. For the third research question, we modeled the probability of student misidentification--or the probability that the teacher of a student with AP potential will not identify them for honors or AP courses--using a logit. Conclusions/Recommendations: We found that large shares of students did not fulfill their AP potential, which varied by student background and subject area. We did not find support for many of the student-centered reasons for forgoing AP such as lack of motivation and constraints on time due to work or extra-curricular activities. We did find, however, that teacher identification and academic self-efficacy mattered to AP course- and exam-taking, especially for marginalized students suggesting viable policy and practice levers to improve equitable AP participation. We discuss implications for policy, practice, and research.