This project sought to improve not only students' ability to communicate science effectively, but to help further deepen understanding of physical concepts with minimal instructor load. By adapting Calibrated Peer Review and implementing it so that students wrote papers based on group activities completed during class, the project hoped to facilitate these goals. What was found was that not only did student writing noticeably improve in their attention to detail and their ability to explain the mathematics involved in solving the physics problem, but also that students in general were able to identify this change themselves. It was also discovered that explicit instructions outlining what should be included when writing about science, provided in the form of the guiding questions, were largely responsible for improvement over the previous year's students who had received very little instruction on what to write.
There is no overwhelming data to support that deeper understanding of the physics involved in writing the papers definitely occurred. However, while data showed no improvement in students' ability to convey an understanding of the physics of the problem, or to explain what was learned and which insights were gained, the lack of improvement does not necessarily mean that deeper understanding of the physics did not occur. Because each activity was conceptually more difficult than the previous, and the mathematics involved more complex, the constancy of the quantitative data acquired may attest to the fact that the project actually did facilitate a better understanding of the physics involved in each assignment. Furthermore, three of the seventeen students openly identified that a deeper understanding of physical concepts had occurred over the course of the project.
Implementation of this project required intense effort on the part of project collaborators. Development of the guiding questions, rubric, writing prompts, data collection and analysis, and the structuring of the stages required a dedicated amount of time on the part of the course instructor. To achieve the goal of minimal instructor load for potential adopters of this writing unit, several considerations and revisions should be made. Data show that the guiding questions themselves had a positive impact on student writing, and the rubric and peer evaluation process were effective in facilitating certain change in student writing, attesting to the fact that they are both useful in the state that they are in. Question 8 of the guiding questions asks students to explain their mathematical manipulations. While the other criteria which showed improvement were more vaguely derived from the rubric, this question corresponds directly with a rubric criteria. Students were not addressing this question well in the beginning of the process, but by the final paper, this guiding question was addressed in a much more complete way, suggesting that the rubric and the peer review process played an important role in bettering students' ability to write.
Time could be taken to adapt these tools for specific course goals, but an interested instructor would not have to devote considerable time to such changes. Further work is needed to improve the online submission tools used in stages 2 and 4 to automate the feedback process. Ideally, students should be able to receive instant feedback on how well they scored instructor examples as well as instant feedback from their peers. The original CPR handled this extremely effectively; adopting the official CPR on line structure may be the answer to addressing these concerns. Criticisms from two students identified a lack of instructor feedback as being a hindrance to their learning, so it may be impossible to maximize student learning while completely removing instructor feedback of papers.