Education Policy Seminar Series - Spring 2020

For the current Education Policy Seminar Series, please visit our website. 

Spring 2020

Eric TaylorUsing Survey Scores to Estimate Social-emotional Growth: Implications for Psychology, Education, and Policy​

Jim Soland, Assistant Professor, University of Virginia
Monday January 27, 2020, 12:00-1:30 PM  
Holloway Hall (Rm 116), Bavaro Hall

Bio: Jim Soland is an Assistant Professor of Quantitative Methodology at the University of Virginia and an Associated Research Fellow at NWEA, an assessment nonprofit.  His research is situated at the intersection of measurement and policy.  Particular areas of emphasis include measuring social emotional learning, quantifying and correcting for test disengagement, and growth modeling more generally.  His work has been featured by the Collaborative for Academic, Social, and Emotional Learning (CASEL) and the Brookings Institute.  Prior to joining UVA and NWEA, Jim completed a doctorate in Educational Psychology at Stanford University with a concentration in measurement and policy.  Jim has also served as a classroom teacher, a policy analyst at the RAND Corporation, and Senior Fiscal Analyst at the Legislative Analyst’s Office (LAO), a nonpartisan organization that provides policy analysis to support the California Legislature.

Abstract: A huge portion of what we know about how humans develop, learn, behave, and interact is based on survey data.  In education, the vast majority of our knowledge on students’ social-emotional development is based on survey scores because related mindsets and competencies are not observed and difficult to quantify. For example, how children develop skills to regulate their behaviors and how adolescents develop confidence in their academic abilities have both been studied primarily using surveys.  Evidence suggests healthy development in such social-emotional competencies is integral to long-term educational attainment outcomes like finishing high school and attending college, as well as later life outcomes like earnings and happiness in adulthood (Dweck, Walton, & Cohen, 2011).  Yet, while we know that measurement bias common to surveys can affect scores at a given timepoint (especially self-report bias), little is known about how these biases affect growth estimates.  This series of studies will examine how much measurement bias affects our understanding of students’ social-emotional growth, including in a quasi-experimental context.  Implications for education, policy, economics, and psychology research will be discussed.

Three peopleRelationship Between Income Inequality and Achievement Since 1990​

Tom Kane, Professor, Harvard
Thursday February 20, 2020, 12:00-1:30 PM 
Holloway Hall (Rm 116), Bavaro Hall
Cosponsored with the Department of Economics

Bio: Tom Kane is an economist and Walter H. Gale Professor of Education at Harvard. He directed the Measures of Effective Teaching project for the Bill & Melinda Gates Foundation-- the largest study of classroom practice ever undertaken.  He has studied the design of school accountability systems, charter schools, teacher effectiveness, financial aid for college, race-conscious college admissions and the earnings impacts of community colleges.  Along with colleagues from Harvard and Dartmouth, Kane will be working with a set of districts from rural New York and Ohio on strategies for lowering student absenteeism and increasing college enrollment.  From 1995 to 1996, Kane served with President Clinton's Council of Economic Advisers.  Kane has also been a faculty member at Harvard’s Kennedy School and at UCLA and has held fellowships at the Brookings Institution and the Hoover Institution.

Abstract: Trends in income-based achievement gaps are central to current concerns over intergenerational mobility.   Yet there is no consistent measure of academic achievement linked to student-level family income over time and no national data collection effort to measure it directly.  Researchers have had to find other ways to infer changes in income-based achievement gaps over time.   For example, Reardon (2011) combined data from a series of longitudinal studies—using different assessments and different measures of student-reported family income—to conclude that there’s been a 40 percent widening in the gap in achievement for those at the top and bottom of the income distribution for 4th and 8th grade students since the late 1980’s.

We take a different approach, combining the NAEP data (with its consistent scale over time) with Census measures of family income at the block group and census tract level.    We use school-level aggregates—on the mean as well as the variance in income and achievement—to reconstruct the student-level relationship between income and achievement.   We do not find evidence of a widening of gaps in achievement in 4th or 8th grade, math or reading since 1990.   On the contrary, we infer that there has been a substantial narrowing of income-based achievement gaps in 4th grade math and English and stable gaps in 8th grade—as well as a sizeable increase in mean achievement at all income levels in math.