Low-stakes testing, or quizzing, is a formative assessment tool often used to structure course work. After students complete a quiz, instructors commonly encourage them to use those quizzes again to retest themselves near exam time (i.e., delayed re-quizzing). In this study, we examine student use of online, ungraded practice quizzes that are reopened near exam time after a first graded attempt 1-3 weeks prior. We find that, when controlling for preparation (performance in a previous science, technology, engineering, and mathematics [STEM] course and incoming biology knowledge), re-quizzing predicts better performance on two cumulative exams in introductory biology: a course posttest and final exam. Additionally, we describe a preliminary finding that, for the final exam, but not the posttest, re-quizzing benefits students with lower performance in a previous STEM course more than their higher-performing peers. But unfortunately, these struggling students are also less likely to participate in re-quizzing. Together, these data suggest that a common practice, reopening quizzes for practice near exam time, can effectively benefit student performance. This study adds to a growing body of literature that suggests quizzing can be used as both an assessment tool and a learning tool by showing that the "testing effect" extends to delayed re-quizzing within the classroom.