| In this dissertation, I present a mixed-methods case study that focuses on gaps in average academic achievement among students of different socio-economic statuses (Reardon, 2011; Sirin, 2005). I examine the impact on student achievement of one administrator's decision to provide all faculty members with a list that identified students from low-income families in their classes and caseloads (the intervention).;Specifically, I used administrative student data in both a one-year (n=735) and two-year (n=1487) window surrounding the intervention to conduct a differences-in-differences analysis to obtain an unbiased estimate of the causal impact of the intervention on student achievement. I also administered a descriptive survey (58.2% response rate) to the faculty (n=201), in which faculty members' self-reported changes in practice and their reactions to knowing the identity of their students from low-income families.;In my analysis, I found that the intervention led the faculty to help these students gain access to additional academic resources and caused an improvement in achievement of the students from low-income families. This positive impact held true across several outcome measures including grades and state tests, and across two comparison groups. For instance, on average, English grades improved by 0.28 of a letter grade (p=.057) for students from low-income families. For a prototypical student, this increase represents moving from an English grade-point average of 2.15 to 2.43, a 13% increase. The intervention also reduced the school's gaps in student achievement, by socio-economic status. The causal impact of sharing the list of students from low-income families was no greater, on average, for White and Asian students than for Black and Hispanic students.;The majority (65.81%) of the faculty who responded to the descriptive survey said that knowing about these students' income levels had either some impact or a strong impact on how they connected these students to additional resources. Of the faculty who responded to the survey, 75.2% said knowing the names of their students from low-income families had no or minimal impact on their expectations of these students, and 57.27% reported the intervention had no or minimal impact on their perceptions of students from low-income families. |
