The Impact of School-Level Factors on Minority Students' Performance in AP Calculus

• Main Author: Pearson, Phillip Bruce
• Format: Others
• Published: PDXScholar 2014
• Subjects: Calculus > Study and teaching; Educational equalization; Academic achievement; Minorities > Education; African Americans > Education; Hispanic Americans > Education; Advanced placement programs (Education); Science and Mathematics Education
• Online Access: https://pdxscholar.library.pdx.edu/open_access_etds/1849; https://pdxscholar.library.pdx.edu/cgi/viewcontent.cgi?article=2849&context=open_access_etds

In recent years, Science, Technology, Engineering, and Technology (STEM) talent pool has re-emerged as a national priority. Certain racial and ethnic groups are dramatically underrepresented in STEM careers and STEM educational programs, an especially serious concern given demographic transitions underway in the United States. The College Board's Advanced Placement (AP) Calculus program provides one way in which students can gain exposure to college-level mathematics while still in high school. This study analyzed factors that contribute to the success of minority students in AP Calculus using a large, longitudinal (2007-2012), geographically distributed dataset which included important school-level variables and AP scores for 10 urban school districts. Descriptive statistics show that AP success in general and minority success in AP Calculus specifically are unevenly distributed across the dataset. A very small number of schools and school districts account for the majority of the production of passing scores on AP exams. Results from multi- variate regression and multi-level growth modeling demonstrate that school size and academic emphasis on a school level constitute important predictors of success for Black and Hispanic students in AP Calculus. The very narrow distribution of AP success across schools and school districts suggests that a specific set of school-level policies and practices are likely to be highly effective in leveraging these two predictors.