This paper explores the impact of teachers on student reading comprehension. Participants of the study were 550 students in 35 first grade classes and their respective 35 teachers. The study aims at determining 1) whether there is significant between class residual variance in reading comprehension at the end of first grade, 2) whether there is evidence of differential effectiveness and 3) whether dimensions of instructional quality assessed by the CLASS instrument are related to reading comprehension. After controlling for initial code skills and initial language comprehension skills, multilevel models reveal a 10% residual between-class variance in reading comprehension at the end of school year. This result is congruent with the presence of teacher effects. The estimates do not show differential effectiveness. Instructional quality as assessed by the CLASS instrument reveal that emotional support is negatively correlated with reading comprehension. Results are discussed and limits of the study are pointed out.