Three graduate students with the Luddy School of Informatics, Computing, and Engineering have been busy this summer working with INsite, a partnership of about 30 school districts and the IU School of Education, on projects tracking special education progress, discrepancy in the rate of school suspensions and expulsions, and helping teachers and administrators to view changes in student data across time.

Through the Faculty Assistance in Data Science program, INsite was paired with graduate students pursuing a M.S. in Data Science through the Luddy School for summer research internships. These paid internships provide faculty with access to expertise and assistance in advanced data analytics, visualization, and development for the purpose of catalyzing their research.

“INsite has a very small staff at IU that supports our districts’ data management and visualization needs. The FADS program offered an opportunity for us to explore several areas of interest that our regular staff either does not have the time or specific data science expertise to do internally,” said Molly Stewart, Director of INsite.

Rohan Punyarthi, a master's student at Luddy, focused his project on Indicator 4A, which involves analyzing and addressing a significant discrepancy in the rate of out-of-school suspensions and expulsions totaling greater than ten days for students with disabilities. Indicator 4A is a data-driven initiative aimed at understanding and addressing disparities in disciplinary actions for students with disabilities in educational settings. The data that has been submitted for the Local Education Agency (LEA) shows that the rate at which students with disabilities are subjected to out-of-school suspensions or expulsions lasting more than ten days is significantly higher than the statewide average for three consecutive years.

“Our goal is to utilize data science methodologies and analytical techniques to identify the root causes of these discrepancies and develop targeted interventions to reduce the disparities. By doing so, we aim to create a more inclusive and equitable learning environment for students with disabilities, ensuring that they have the same opportunities to thrive and succeed as their peers,” Punyarthi explained. “In pursuit of our goal, we have created a live interactive dashboard to visualize and monitor the data in real-time. The dashboard allows stakeholders to explore the trends and patterns in out-of-school suspensions and expulsions among students with disabilities, enabling data-informed decision-making and strategic interventions at different levels of the educational system.”

“Overall, I am grateful for the opportunity to be part of this impactful project and to contribute to making a positive difference in the lives of students with disabilities. My aim is to contribute to creating a more equitable and supportive educational system for all students,” he added.

Mariah Benham currently works on the Survey and Evaluation team with Prevention Insights, part of the Department of Applied Health Sciences in the IU School of Public Health, and will complete her master’s degree with Luddy this December. Her project with INsite focused on developing a web application that allows teachers and administrators to view changes in student data across time, with visualizations she created that show student grades and associated risk scores across multiple school terms based on final numeric grades for the students at the end of each semester. In addition, there is information on how the grade, attendance, and discipline risk scores are calculated.

“I went into this project really wanting to create something that would help INsite in the long term. I knew we only had a certain number of weeks together so I wanted to focus on an area that would continue to be implemented and useful to their team after the internship period was over,” Benham said.

Haijing Tu, a master’s student at Luddy, worked with the South Bend Community School Corporation on their special education compliance data. That data consisted of analyzing over 33,000 Individualized Education Program (IEP) records. Tu worked on data cleaning, aggregation, merging, sorting, and visualization using Python pandas and Tableau, two data analysis software tools. With the overall goals of finding out the number of monitored versus unmonitored IEP goals for each student and creating interactive visualization for filtering and searching specific records based on school, teacher records, and compliance status, Tu met regularly with officials at SBCSC to report on the project’s progress and solve problems as they came across them.

“The completed project is an interactive Tableau dashboard that allows filtering by school, teacher of record, and compliance labels to see student records and the number of monitored and unmonitored goals for each student,” Tu explained. “I was really lucky to have this opportunity, because I have great interest in K-12 education.”

One of those officials Tu worked with was Joel Boehner, Assistant Director of Exceptional Learners at SBCSC.

“Haijing's work as a data science intern this summer was essential in helping us develop dashboards to better use our data to support program compliance and learning outcomes for our students with disabilities. I'm very grateful for her time and effort as she directly interacted with many of the logistical challenges that come along with doing data work in public education. I'm looking forward to implementing the tool she helped develop this school year that will help us ensure that we are appropriately monitoring student progress through their individualized education program,” Boehner said.

When INsite applied for interns through the FADS program, Stewart said the goal was to help show proof of concept in the three project areas, which could then be turned into production-ready visualizations for districts.

“The three interns that were matched with INsite proved more than capable to help us meet this goal! These three areas were all difficult if not impossible for districts to track on their own, and all of the interns made incredibly helpful progress on each of these areas in terms of getting us closer to production-ready visualizations,” Stewart added.