This assignment is by Brian Macdonald and is included on the AI Teaching Examples on the Poorvu Center for Teaching and Learning website.
This assignment involves evaluating ChatGPT’s explanations of Poisson Regression and its suitability compared to linear regression. Students will identify inaccuracies in ChatGPT’s responses and provide corrections, enhancing their understanding of statistical modeling and critical evaluation skills. You can easily customize this activity by asking AI to respond to a concept related to your discipline.
Key Features of This Assignment
- Identifying Inaccuracies
- Students pinpoint inaccuracies in ChatGPT’s explanations, focusing on the assumptions and applications of linear and Poisson regression models, which improves their analytical skills and understanding of statistical concepts.
- Clarifying Statistical Assumptions
- Students will correct misunderstandings related to the distribution assumptions of linear regression and Poisson regression, such as the normality of response variables and the equality of mean and variance in Poisson regression.
- Application of Quasi-Poisson Regression
- Students will distinguish between Poisson and Quasi-Poisson regression models, explaining when each model is appropriate based on the relationship between the mean and variance of the response variable, thereby enhancing their application skills in statistical modeling.
Find the full version of this resource on the Poorvu Center for Teaching and Learning Box site.