Advisors use placement test scores as a means of predicting students' proficiency in mathematics; however, there is a debate about how accurately these scores predict students' success. This nonexperimental quantitative study focused on one test, the Texas Success Initiative (TSI). The purpose of the study was to determine whether the test is an accurate predictor of students' success in college algebra for students in science, technology, engineering, and mathematics (STEM) majors, and whether students who took the test continued pursuing a STEM major. The theoretical framework for this study was Tinto's theory of retention. Statistical Package for the Social Sciences software was used to generate 500 random cases from 2,339 students ranging from 18 to 50 years of age who enrolled in Math 1414 during the Spring 2015 to Spring 2017 semesters at the Texas community college setting. Hierarchical multiple and logistic regression were performed to test whether the TSI scores significantly predicted students' math grade and retention. The hierarchical multiple regression revealed that the TSI score explained only 13% of the variance in math grades (R2 = .13). The logistic regression showed that the TSI score explained a variance of only 7% (Nagelkerke R2 = .07) and yielded a higher number of false positives in predicting retention in a STEM mathematics track after controlling for high school GPA, gender, ethnicity, and age. Findings revealed no significant relationship between TSI scores and students' academic success and retention. The results from this study may contribute to positive social change by providing academic advisors with additional knowledge of the best practice for placing students to achieve success in college math courses.