May 2016 The AERA Grants Program held its annual AERA Institute on Statistical Analysis for Education Policy on May 10–13 in Washington, D.C. Twenty-nine graduate students and early career scholars participated.
This year’s institute, directed by Sarah Lubienski, Professor in Curriculum and Instruction at the University of Illinois at Urbana–Champaign, examined the “Use of Large-Scale Data to Study Mathematics Education and Outcomes.” The institutes are designed to build scholars’ research capacity to use advanced statistical techniques to examine large-scale national data sets such as those from the National Center for Education Statistics (NCES), National Science Foundation (NSF), and other federal agencies for basic, policy, and applied research.
The 2016 institute focused on three large-scale federal data sets: the Early Childhood Longitudinal Study, Kindergarten Class of 2010–11 (ECLS-K:2011), the Educational Longitudinal Study of 2002 (ELS:2002), and the High School Longitudinal Study of 2009 (HSLS:09). Each of these data sets includes items about students’ mathematics course taking, grades, test scores, and other school experiences.
Elise Christopher, a statistician and project officer at NCES, presented an overview of these large-scale data sets and discussed future NCES data collections. Hands-on training on statistical concepts and methods, led by Joseph Robinson-Cimpian (University of Illinois at Urbana–Champaign), included lessons introducing causal inference, regression discontinuity, and propensity score matching. Colleen Ganley (Florida State University) taught participants effective techniques to access data, manipulate variables, and adjust data for working with large-scale data sets. She then guided participants through propensity score matching examples designed to answer mathematics education outcome questions.
Lubienski provided a brief history of U.S. mathematics education and discussed how education policy has influenced 25 years of reform on mathematics education. The institute included a stimulating education policy discussion on “how to sensibly and sensitively report group differences,” which generated a captivating discussion on the correct use of language when referring to different student populations and groups in research.
Institute participants prepared group presentations based on their research interests. Groups explored questions examining students’ mathematics outcomes over time, such as the relationship between pre-kindergarten opportunities and math performance, math identity among high school students, math remediation in college, and the relationship between high school math courses and later STEM coursework in college. At the conclusion of the institute, participants presented their collaborative work, and shared their perspectives, questions, and concerns on applying the statistical techniques that were discussed.
“It’s a pleasure to be able to provide these important opportunities to scholars who often feel isolated in their attempts to enter the arena of large-scale analyses of education data,” said Lubienski.
AERA Executive Director Felice J. Levine, who also serves as AERA Grants Program Principal Investigator, said, “This institute reflects the purpose and aims of the long-standing AERA Grants program to provide the highest quality of training and professional development to the next generation of scholars using large-scale data sets. We are pleased to make these training opportunities available.”
Video-recorded lessons from the 2016 AERA Institute on Statistical Analysis for Education Policy will be available through the AERA Virtual Research Learning Center. The Institute was supported by the AERA Grants Program under the NSF Grant DRL-0941014.
2016 AERA Institute on Statistical Analysis Faculty