A PRIISM Seminar by Chapman University's Chelsea Parlett-Pelleriti
Join PRIISM and Dr. Chelsea Parlett-Pelleriti to learn about Bayesian Item Response Theory, illustrated with applications to metamemory data.
Abstract
Item Response Theory (IRT) is commonly applied to validate surveys/measures and do computer adaptive testing. However IRT model structure which combines item level and subject level parameters can be applied to a broader range of models and analyses. For example, Beta Regression, its extension of Zero-One-Inflated Beta Regression, and Ordered Beta Regression which provide modeling frameworks for continuous bounded data. These models will be presented in a Bayesian Framework, and modifications to these models to include IRT parameterization will be suggested. Applications to metamemory data will be shown, and the basic ideas behind using cumulative logit, inflation, and logistic models with IRT parameterization will be discussed.
Bio
Dr. Chelsea Parlett-Pelleriti is a faculty member in the Fowler School of Engineering at Chapman University. Her core courses include topics in Statistics, Data Science, and Machine Learning. She received her Ph.D. in Computational and Data Science, with an emphasis on applying Machine Learning and Bayesian Statistics to behavioral data (such as Psychology), which paired her computational and technical skills with her background in Psychology. As an instructor, Chelsea values using new, engaging technology and other pedagogical techniques (such as flipped classes, educational Shiny apps, and guided simulation activities) in order to help students develop a sense of passion, rigor and engagement with their courses. Chelsea is also passionate about effective science communication, both in the classroom, as well as on social media, because effective science communication encourages diversity and welcomes people into the fields of Statistics and Data Science.
