Swift solution: Taylor Swift sparks innovation for education researcher
LAWRENCE — Taylor Swift speaks to human experiences through emotion, interpretation and expression. Researchers investigate those experiences using analysis, measurement and science. Jake Thompson connected these approaches to fill a blank space in education research.
Thompson, assistant director of psychometrics at Accessible Teaching, Learning, and Assessment Systems (ATLAS), has a strong appreciation for music. He played the violin as a kid and said he gravitates to music because there is a song for every feeling. Growing up with parents who listened to Swift in the country music stage of her career, Thompson’s earliest memory of the now mega pop star is being an anti-fan.
“I will not like the things that my parents like. ... I will not like Taylor Swift,” Thompson said while reflecting on his rebellious teen era.
With a newfound grace, Thompson said everything changed for him in college when Swift released “Red.” Two years later, Swift left a permanent mark on Thompson with the release of “1989.” After 10 years of new and rereleased albums, Thompson is a self-proclaimed “Swiftie” who appreciates Swift’s whole vibe. While becoming a loyal fan, Thompson also became a trusted expert in his field and found a way to combine his love for Swift with his professional expertise.
For three summers now, Thompson worked to build measr (pronounced “measure”), a tool that simplifies the process of analyzing and evaluating diagnostic classification models (DCMs) making them accessible to a wider range of researchers and practitioners. DCMs are statistical models used in student assessment to measure knowledge, skills and understanding based on a predefined set of skills or attributes.
Large-scale assessment systems such as the Dynamic Learning Maps (DLM) alternate assessments use DCMs to offer shorter and more reliable assessments for students with the most significant cognitive disabilities.
Thompson implements DCMs to analyze assessment data. These models provide educators with a clearer picture of a student's strengths and weaknesses to support their instructional decisions. However, using DCMs in applied settings often presents challenges.
"Before measr, there were few user-friendly software options," Thompson said, “and the options that were available required extensive coding knowledge.”
Creating a new tool from scratch comes with learning curves. Thompson knew he could be fearless in a low-stakes, trial-and-error environment and needed a place to workshop ideas before applying them to measr. This is where “taylor” took center stage.
In his free time, Thompson created the taylor package, which provides access to a dataset of Swift’s lyrics, audio characteristics and album cover themes. This package picked up interest among the data science community last year when hundreds of practitioners used taylor for what they call a #TidyTuesday spotlight.
With taylor as a training ground for his programming skills, Thompson could create a powerful and user-friendly tool to support researchers across skill levels.
“Researchers can leverage DCMs with measr to help uncover more precise and actionable data for educators,” Thompson said, “ultimately leading to improved learning outcomes for students.”
In finding an impossible harmony between music and data science, Thompson created an innovative tool to reduce barriers for education researchers. From the album that sparked his fandom, “Don’t you dream impossible things?”
About ATLAS
ATLAS, a center within the Achievement & Assessment Institute at the University of Kansas, promotes learning and improved outcomes by creating accessible and academically rigorous technology-based learning and assessment systems. Visit the ATLAS website to learn more.