Xu Qin

Faculty Member Xu Qin Receives NSF CAREER Award

Xu Qin, an assistant professor of research methodology at the School of Education, was awarded a 2024 National Science Foundation (NSF) Faculty Early Career Development Award. Qin is also an assistant professor of biostatistics at the School of Public Health.

The award comes with an $842,512 NSF grant for her research project, titled “CAREER: Complex Causal Moderated Mediation Analysis in Multisite Randomized Trials: Uncovering the Black Box Underlying the Impact of Educational Interventions on Math Performance.” The project will explore innovative quantitative methods in educational research that are used to evaluate the heterogeneity of mechanisms underlying the impacts of interventions in multisite randomized trials.

“By understanding how an intervention worked differently across individuals and sites, we can better understand the intervention’s efficacy and leverage the resources that work in the contexts where students learn and grow,” says Qin.

“Currently, there is a lack of statistical methods and tools for assessing how complex mediation mechanisms vary by individual and contextual characteristics in multisite trials,” says Qin. “As a quantitative methodologist, I am enthusiastic about developing methods and tools to fill the gap. These methods and tools will help researchers improve and tailor interventions to different individuals and school contexts and thus enhance educational equity.”

She hopes that methodological advances developed through this award can inform policies and programs with direct societal impacts. Through analysis of the National Learning Mindset Study and the Head Start Impact Study data sets, she aims to help practitioners make individual- and site-specific modifications of intervention designs and implementations for improving math education.

Qin intends to share and disseminate her findings from her NSF CAREER Award project by developing an R package and a user’s manual to facilitate applications of the methods. The package will allow users to specify models via path diagrams, which will ease implementations especially for those who have no experience of using R.

Qin previously received the 2022 National Academy of Education/Spencer Postdoctoral Fellowship. Her work developed methods for causal moderated mediation analysis with a focus on one single mediator. Under the support of the NSF grant, she will develop methods to further allow for multiple concurrent or sequential mediators in multisite randomized trials. As mediation mechanisms get more complex, it becomes more challenging for users to specify models. The development of the user-friendly R package with a graphical interface will greatly ease implementations and enable empirical researchers to generate a new set of thorough, precise, and valid empirical evidence regarding the heterogeneity of causal mediation mechanisms across individuals and contexts.

At Pitt Education, Qin is teaching a new course on causal moderation and mediation analysis (EDUC 3418).

“In the future, the course will cover the methods and software developed from this project,” says Qin. “The target audience is graduate students from diverse disciplines, including education, psychology, and public health. The course will equip students with powerful tools to dig more deeply into their substantive research.”