PERC 2025 Abstract Detail Page
Previous Page | New Search | Browse All
| Abstract Title: | Mining and Measuring 19 Misconceptions: Which Are Improved By Instruction? |
|---|---|
| Abstract Type: | Contributed Poster Presentation |
| Abstract: | We have discovered ~19 coherent, student misconceptions and confusions using multi-dimensional IRT plus machine learning. We find some not previously identified in the PER literature, e.g. "impetus force on circular path" and "air exerts significant drag and downward force". We can find the amount of each misconception held by an individual or class, allowing the use of pre- and post-testing to find the diminishment of each misconception by instruction vs. pretest score (1-30) for 17k student results on the FCI. We find 11 Standard Misconceptions that are strongly held by average and below students and are well remediated for well above average students only. We also find 8 Naive Misconceptions that are strongly held by below average students and are remediated for all students and 3 "confusions". This work provides formative assessment as well as standards for developing pedagogies to address specific misconceptions. |
| Session Time: | Poster Session A |
| Poster Number: | A-158 |
Author/Organizer Information | |
| Primary Contact: |
David E. Pritchard MIT Cambridge, MA 02139-4307 Phone: 617669 2347 |
| Co-Author(s) and Co-Presenter(s) |
Martin Segado, MIT Aaron Adair, MIT John Stewart, UWV |
