Project resources are archived on the CSforAll Teachers site at the following link: https://csforallteachers.org/resource/cooperative-learning-structures-ap-csp

The Advanced Placement® Computer Science Principles (AP CSP) course was the culmination of an eight-year NSF/College Board pilot project that exceeded all expectations in terms of enrollment in its first two official years. Broadening the accessibility of computing to include a diverse population of students has been a fundamental design goal of the AP CSP curriculum framework (CF). This project is an NSF-sponsored multi-year effort to infuse cooperative learning (CL) structures into AP CSP classrooms to improve class participation and student learning. AP CSP was designed to engage all learners.  CL structures have been demonstrated to engage diverse learners who are in the same classroom. The goal of CL is to offer an environment where every student is encouraged to participate equally toward their own academic achievement, while simultaneously working toward a positive common goal with teammates from different backgrounds and mixed abilities, thus promoting broadening participation in computer science (CS) classes.

The intellectual merit of this project is the design and assessment of the best practices of CL applied to the AP CSP CF. This project has potential broad impact by changing the way CS is viewed and taught. Through face-to-face (F2F) workshops spanning July 2015, 2016, and 2017, 143 AP CSP teachers were introduced to CL structures and guided in the integration of these structures with the AP CSP CF. The contribution of the project includes a workbook and accompanying videos of CL strategies that can be integrated in any CSP classroom, regardless of any specific CSP curriculum being used. The project's open resources will be made available on the CS10K Community of Practice for sustainability and scalability of the project to those who did not attend the workshops. Although this proposal focused on CL in the CSP classroom, the structured CL pedagogy can be used in any classroom with any discipline.

Our research efforts engaged in numerous and varied data collection activities with the participating teachers and their students. We sought to understand the effects of the professional development (PD) that focused on AP CSP pedagogical content knowledge using CL structures on teachers; specifically, how, and to what extent the teachers implemented CL in their AP CSP classrooms. We also sought to understand whether and to what extent the CL activities had an impact on student self-efficacy and AP Exam scores.  We investigated the relationships between 1) PD that infuses CL structures within the AP CSP context, 2) classroom practices with CL structures, and 3) student outcomes. Specifically, we hypothesized that the Opportunity To Learn Collaboratively (OTLC) would positively impact students' learning (i.e., AP scores) and that students with high OTLC would outperform both a national sample and a carefully matched state composite sample, particularly for females and URM students. We also hypothesized that higher OTLC classes would positively predict students' self-efficacy in computer science.  

The general findings of our project indicate that the F2F PD workshops led to gains in the understanding of and confidence in teaching with CL structures and the use of CL structures was a statistically significant and a positive predictor of student AP scores for participating classes. Student outcomes for the CL sample alone were analyzed in terms of the relationship between students' OTLC  and students' self-efficacy in computer science (CSSE). OTLC scores did not significantly predict students' post-course CSSE levels. However, the use of Pair Programming alone predicted post-course CSSE.

Students' AP scores compared favorably to national sample of  2017 + 2018 AP CSP exam-takers. AP CSP scores of students in the CL sample were higher than would be expected based on this national data. CL sample passing rates were significantly higher than expected for:

-       All students (76.8% vs. 72.3%)

-       Male students (79.9% vs. 73.7%)

-       Under-represented minority (URM) students (66.3% vs. 55.0%)

We also looked at a carefully matched state composite sample. Students in the CL sample outperformed a matched 2017+2018 state composite sample, constructed from state-level data (13 states). Passing rates were significantly higher than expected for:

• All students (76.7% vs. 70.4%,)

• Male students (79.9% vs. 72.3%)

• URM students (66.1% vs. 53.6%)

Passing rates for Female students (69.1% vs. 65.9%) and for Non-URM students (82.4% vs. 77.4%) were numerically higher but the difference was not statistically significant.

These powerful effects were realized from one week of summer F2F PD per teacher, and each teacher's daily choices of if, when, and how to incorporate CL structures into classroom activities. Teachers clearly communicated their need for additional support and resources (e.g., a learning community) throughout the academic year. The benefits of engaging in a persistent teacher learning community would have led to more effective use of CLs, prompting even stronger student outcomes.

Last Modified: 05/31/2019
Modified by: Jeffrey G Gray