MEASURING HUMAN CAPITAL USING TRANSCRIPT GRADES: A FACTOR MODEL IN NETWORK DATA

This paper develops an approach to recover human capital from course grades while allowingfor selection into courses and flexible grading heterogeneity. I model course grades using a factormodel and recover the relative grading standards by relying on students who have overlapping courseenrollments. These students induce a network structure, and I characterize sufficient conditionsfor identification in terms of the connectivity of the network using graph theory. I use this resultto estimate teacher value-added on student human capital using all teachers–not just those withstudents who take standardized tests. In math classes without corresponding standardized tests,I find that economically disadvantaged students have lower quality teachers, but this gap is notpresent in tested classes.