The 91涩漫-ICPSR Partnership for Expanding Education Research in STEM (PEERS) Data Hub Series in STEM Data and Data Use, features workshops that examine the structure, design, and purpose of multiple data sets that can be analyzed using quantitative, qualitative, and/or multi-method techniques. In these workshops participants can explore diverse forms of designed data, administrative data, and big data that can be used in STEM research. The publicly available datasets presented include large-scale student and school data across educational levels and contexts, such as federally supported datasets that provide a national or international scope as well as data that focus on specific states, regions, and local areas. Many of the data sets can be analyzed using custom designed analysis tools that will be presented in the webinars. Led by experts in education research, psychometrics, and the social sciences, the workshops provide an overview of the data sets and discuss potential uses to address salient topics in education and STEM research.
Upcoming Workshop(s)
Studying Math Pathways with Higher Education State Administrative Data: Opportunities and Challenges
August 13, 2026; 12:00 pm ET
Speakers:
Yuxin Lin, Maryland Higher Education Commission
Florence Xiaotao Ran, University of Delaware
State administrative data sets are a powerful tool for longitudinal analysis. This workshop will introduce graduate students to a novel analysis of Maryland Higher Education Commission (MHEC) data by Drs. Lin and Ran that looked at the impact of math course taking requirement policy reforms in community colleges on choosing a STEM major. While this study focuses on the results using Maryland community college data, the presenters will also discuss the availability and potential use of data from four-year universities within the state. As Maryland's state higher education agency, MHEC maintains statewide administrative student unit-record data that capture enrollment, course-taking, financial aid, and degree completion across the state's public and independent institutions, including all 16 community colleges, 13 public four-year institutions, and 13 independent private colleges and universities. Attendees will gain an understanding of:
- The Maryland Higher Education Dataset, requirements to access the data and available dashboards
- How Drs. Lin and Ran collaborated on the project
- How they selected variables and analysis techniques for their research and disseminated results
This workshop is designed for graduate students and early career researchers looking to improve their understanding of the research process and researchers interested in learning more about using state level administrative datasets like the MHEC for STEM analysis.
Using Large-Scale Data for STEM Education Research: Lessons from the High School Longitudinal Study of 2009 (HSLS:09)
August 18, 2026; 12:00 pm ET
Speaker:
Guan K. Saw, Claremont Graduate University
This workshop introduces graduate students and early-career researchers to using the High School Longitudinal Study of 2009 (HSLS:09) for STEM education research. Dr. Guan Saw will present his work on rural STEM opportunities and pathways as a case example, discussing the use of publicly available and restricted-use HSLS:09 data, linked with the Common Core of Data (CCD). Attendees will learn how to:
- Identify promising research questions
- Select relevant variables
- Employ appropriate analytic techniques
Attendees will leave with practical strategies for developing their own secondary-data research projects.
Using Machine Learning to Analyze SLDS Datasets
August 20, 2026; 12:00 pm ET
Instructor:
Tracy Sweet, University of Maryland, College Park
This workshop will introduce researchers to the use of machine learning techniques to address questions in STEM education research using a State Longitudinal Data System (SLDS). Dr. Tracy Sweet will discuss her recent research applying machine learning algorithms to the Maryland Longitudinal Data System (MLDS) to predict student outcomes for education research.
Machine learning is a set of algorithms used for prediction. In one study, Dr. Sweet and her colleagues used supervised machine learning algorithms to examine the impact of taking 15 credits during the first year of college on returning in the second year and attaining a college degree within 5 years. In other studies, they attempted to predict standardized test scores in Algebra and graduate school enrolment using MLDS data.
In this workshop attendees will learn about:
- Accessing and using the MLDS dataset
- Common supervised Machine Learning algorithms and evaluation metrics
- Interpreting machine learning results and understanding the potential for bias in these models
Past Workshop(s)
Unleash the Power of SLDS: Preschool through grade 20 Datasets for STEM Research
July 23, 2026; 12:00 pm ET
Speakers:
Beth Kelly, KYSTATS
Rachel Leventhal-Weiner, DataLinkCT
Tod Massa, VLDS
Moderator:
Nancy Sharkey
In recent years, many states have combined administrative datasets to create a system that can track a student's experience from early learning through postsecondary school and into the workforce. These databases are often referred to as State Longitudinal Data Systems (SLDS) or P20W (Preschool through grade 20) systems and can transform educational research. They can include data from K-12, Postsecondary, Adult Education, workforce, benefit program participation and sometimes law enforcement systems. Join experts from Virginia, Kentucky, and Connecticut as they discuss key aspects of their SLDS systems including:
- The types of data available in the system (e.g., K-12, post-secondary, program participation, etc.)
- Where to find documentation on using the system and variable names
- Requirements for researchers to access the system and privacy protections
- Dashboards and publicly available access tools
This workshop is ideal for graduate students, early career and senior researchers looking for a powerful source of longitudinal and outcome data on STEM education issues.
Learning Opportunities Early and Often: Longitudinal Findings from the Study of Early Child Care and Youth Development
July 16, 2026; 12:00 pm ET
Instructors:
Andres Bustamante, University of California, Irvine
Eric Dearing, Boston College
Deborah Vandell, University of California, Irvine
Read instructors' bios.
Longitudinal datasets like the Study of Early Child Care and Youth Development (SECCYD) are uniquely powerful tools for educational and child development researchers who study STEM education and learning. With 26 years of data, the SECCYD follows a large cohort of children from birth, providing rich assessments of cognitive, social, emotional, and relational development, as well as the quality of the environments in which children grow—including their homes, early education settings, schools, afterschool programs, and communities.
Drawing on their multiyear collaboration, Andres Bustamante, Eric Dearing, and Deborah Vandell will demonstrate how the SECCYD can be leveraged to uncover new insights into the impact of developmental and educational opportunities across childhood and adolescence. They will discuss their work, the evolution of their collaboration, and practical guidance for accessing and using the SECCYD.
Attendees will gain an understanding of how the researchers:
- Identified and accessed the SECCYD data rather than other longitudinal datasets
- Selected variables and analytic approaches to address their research questions in STEM education and learning
- Sustained and expanded their research collaboration over time
This workshop is designed for graduate students and early career researchers seeking to strengthen their understanding of the research process, as well as scholars interested in using SECCYD data to support STEM education research.
Unlocking the STEM Education Data Vault
June 11, 2026; 12:00 pm - 1:15 pm ET
Instructors:
Sandra Tang, ICPSR at the University of Michigan
Julia Kaufman, RAND Corporation
Elizabeth Steiner, RAND Corporation
Sean Simone, Council for Opportunity in Education
Learn about the breadth and depth of STEM education related datasets available for research from experts. In this webinar, attendees will:
- Gain an overview of the variety of STEM related datasets available through ICPSR, RAND Corporation and Federal agencies
- Learn the requirements to access and use the data in their own research
- Receive an introduction to NCES’ DATALAB and POWERSTAT tools for conducting analysis
This workshop is ideal for graduate students, early career and senior researchers that are looking for large scale datasets to answer STEM related research questions.
Introduction to the Stanford Education Data Archive (SEDA)
March 9, 2021; 12:30pm -2:30pm ET
Instructors:
sean f. reardon, Stanford University
Andrew D. Ho, Harvard University
Benjamin R. Shear, University of Colorado, Boulder
Erin M. Fahle, St. John’s University
This workshop introduces attendees to the Stanford Education Data Archive (SEDA), a database of nationally-comparable academic achievement data made publicly available by The Educational Opportunity Project (EOP) at Stanford University. SEDA is based on roughly 430 million standardized test scores from 3rd-8th grade students in U.S. public schools, districts, counties, and states from the 2008-09 through 2017-18 school years. These data enable nationwide research to better understand student achievement and educational opportunity. In this workshop, the EOP team will provide a description of SEDA’s contents, showcase examples of correlational and causal research that can be conducted with SEDA, and highlight relevant validation studies that support the use of SEDA in different types of research. Attendees can explore the SEDA data before and during the training using our interactive data explorer () or by downloading the data files () and using statistical software (e.g., R or Stata). For more information on SEDA and research using SEDA please visit the Research section of the EOP website ().
Prior to the workshop please complete the online Data Use Agreement: .
The PEERS Data Hub is joint effort of 91涩漫 and the Inter-University Consortium for Political and Social Research (ICPSR) that is NSF funded (Award ECR 1937612). It provides collaborative space for STEM education research communities to build and advance knowledge by sharing innovative ideas, methods, and tools. Further information about the PEERS Data Hub is available .