In this paper, we present new partnering strategies to pair students for formative assessment in a programming course. These strategies, collaborative and cooperative learning, are two forms of active learning. Both forms have been widely tested in previous classroom experiments and show evidence of effectiveness. This study focused the application on how formative assessment (i.e. a weekly quiz) is administered, rather than on learning activities in general. A collaborative quiz is one based on the idea of paired programming, a technique that has become well-received in computer science. This type of quiz requires the pair of students to program together, using one computer to turn in one final solution. The second method, a cooperative quiz, allows students to collaborate as much or as little as they prefer. For a cooperative quiz, students must each use their own computer to write and submit their own solution, but any collaboration with their partner is allowed. To account for student preferences for active or independent learning, these strategies were also guided by how students are paired. To do this, the Felder-Silverman learning style inventory was used to categorize students along the model’s active-reflective dimension as reflective, active, or “in-between” (neutral) learners. Students were partnered throughout the semester with a variety of partners with attention to mixing and matching their preferred learning styles to determine any effects of partnering.
This study took place across three sections of a sophomore-level programming class in a mid-size, public university in the western United States. The sample included approximately 100 primarily second-year mechanical engineering undergraduate students. The course is a required programming course using MATLAB. A weekly quiz was administered across each of the sections differently. This was done to test the relationship between strategy and assessment performance. Therefore, each week there was a section of students taking a collaborative, cooperative, and independent quiz. All sections were given the opportunity to take the same total number of each quiz type and all sections were taught by the same instructor. A standardized rubric was used to score student performance and compare across sections, across quiz styles, and across learning styles.
The research question guiding this paper is: how does the quiz style and/or partner learning style preference affect student performance on formative assessment? Both quantitative and qualitative methods were used to address this question. Statistical analysis was used to determine the average differences in quiz performance based on quiz style, mixed or matched learning style partnering, and the combination of the above. Several quiz sessions were videotaped, and this data was used to discern the interpersonal dynamics of different partnering conditions. This paper will focus on those results. The results from the statistical analysis demonstrates that working with a partner improved student performance over individual assessment. Video observations confirmed the statistical results, and provide more detail regarding the interpersonal dynamics of each type of student pairing. Additionally, we will present student narrative feedback.
Are you a researcher? Would you like to cite this paper?
Visit the ASEE document repository at
for more tools and easy citations.