The 100-level Introduction to Meteorology course is a requirement for Natural Sciences and secondary science education majors. It also attracts a significant proportion of non-majors. This seems to be the perfect opportunity to teach undergraduates basic data gathering and analysis skills. The course's final project requires the students to analyze the accuracy of Punxsutawney Phil's Groundhog Day predictions. We start out with a discussion of the history of Groundhog Day and Punxsutawney Phil, then review previous attempts to assess Punxsutawney Phil's accuracy. We discuss how different authors have approached their analysis, and potential shortcomings or advantages of their methods. After the discussion, the students design their own analysis. They have to first decide what type of data to gather, and from where. This means they needed to determine the geographic and temporal extent of their data and which of the publically available datasets to use. They then have to decide on parameters for Punxsutawney Phil's "success" – how do you measure spring? How do you measure winter? What would constitute an accurate prediction? The students conducted an analysis of the data, and presented it as a poster during the final week of lab. These posters were evaluated both by the instructor and their peers, using a rubric.f