Although students' motivational frameworks (entity vs. incremental) have been linked to academic achievement, little is known about how early this link emerges and how motivational frameworks develop in the first place. In a year-long study (student N = 424, Teacher N = 58), we found that, as early as 1st and 2nd grade, children who endorsed an incremental framework performed better on a nationally normed standardized math test than children who held an entity framework (i.e., believe ability is stable, prefer easy tasks). Furthermore, teachers' self-reported instructional practices (mastery- vs. performance-oriented) played an important role in the development of students' motivational frameworks. The more a teacher reported emphasizing that children demonstrate competence in the classroom (i.e., performance-oriented instructional practices), the more students endorsed an entity framework at the end of the school year, even after controlling for students' beginning-of-year frameworks. These findings have significant implications for theory as well as practice, as they show that even in the early elementary grades, teacher-reported instructional practices are linked to the development of students' motivational frameworks, which in turn, are linked to students' mathematics achievement.