EMERGING DIALOGUES IN ASSESSMENT

Assessing Student Learning Outcomes in Group-Based Projects: Challenges and Best Practices

August 29, 2025

• Yi-Chin (Sarah) Wu, Ph.D., Director of Assessment, Georgia Institute of Technology, Office of Academic Effectiveness
• Wen-Chi Shie, Ph.D., Assistant Professor, National Chengchi University, Department of Public Administration

Abstract

Group-based learning in higher education significantly enhances student learning outcomes such as critical thinking, intercultural competence, and problem-solving. However, implementing group-based learning and assessment faces challenges like free-rider behaviors and varying levels of motivation in taking the course among students, which can hinder group project quality and learning development. As part of a three-year research project, a pilot study was conducted at an undergraduate course within the Department of Public Administration at a large research university. This study utilized exams, peer assessment, pre-post reflections, observations, in-depth interviews, and pre-post surveys to balance scoring fairness and promote collaborative learning to achieve higher attainment on various learning outcomes. The study results support previous research on improving learning outcomes and highlight the importance of integrating peer assessments and reflections to gauge individual contributions and learning development throughout the collaborative process. Students gradually recognized the value of collaborative projects, indicating that group-based learning assessments should consider cognitive and affective outcomes.

Impact of Group-Based Learning

As higher education shifts towards a student-centered approach, research has demonstrated that group-based learning that is created under a collaborative learning environment significantly enhances student development in various outcomes, such as critical thinking (Loes & Pascarella, 2017), intercultural knowledge and competence (Yang, 2013), problem-solving (Karantzas et al., 2013), and affective development (Loes, 2022). Group-based learning activities have a rich history applied in higher education and are rooted in theories of constructivism and critical pedagogy. Most recent studies on group-based learning have the following characteristics: (1) students work in groups of two or more; (2) a specific learning activity is assigned, such as solving a problem or creating something that students work on; and (3) there will be a common goal for the group, either created by the group or by the instructor. Therefore, not only are students expected to master the learning content, but they are also expected to produce a student artifact with their group members that would be used to assess their attainment of learning outcomes. Through interacting with other group members, students have the opportunity to hear each other's viewpoints and generate new insights during discussions.

However, the implementation of collaborative learning is not always seamless and straightforward. Once teams are formed, common “free-rider” (Maiden & Perry, 2011, p. 452) or “slackers” (Isaac, 2012, p.85) member behaviors, such as absenteeism, failure to complete assigned tasks on time, and poor-quality work, can arise at different stages of group projects and may hinder the quality of the group work (Isaac, 2012; Maiden & Perry, 2011). This article highlights some challenges for assessing student learning in a collaborative learning context and provides good practices in assessment design that can enhance student learning.   

Challenges of Group-Based Learning Assessment

Despite the prevalence of group-based learning in higher education, instructors face significant challenges and uncertainties in assessing students' learning growth (Boud & Bearman, 2024). From the students' perspective, the "free-riding" situation in group-based learning challenges the fairness of the signature assignment assessment. Every member of the group can also bring different levels of motivation and commitment to the course. Consequently, Strijbos (2016) argued that group assignment grades should not be the sole criterion for determining whether students have achieved core competencies. Moreover, due to extensive social interactions, researchers have found that affective development, which includes beliefs, values, attitudes, and emotions (Grootenboer, 2010), is closely intertwined with cognitive growth (Cabrera et al., 2002). Frequently, instructors overlook the importance of affective learning outcomes, such as “students’ ability to demonstrate a positive attitude towards actively inviting peers to share their thoughts” after experiencing a collaborative learning course design. 

Common Practices of Assessment in Group-Based Learning 

Due to various unequal efforts among group members to complete shared tasks, peer assessment is commonly used to establish personal accountability in collaborative learning contexts (Kao, 2013). Through peer assessment, instructors usually expect a more accurate understanding of student contributions. The methods of peer assessment vary depending on the classroom application, but the one usually used for moderating the free-rider is giving scores or qualitative feedback on peers' performance. Research has indicated that students evaluate their peers based on effort instead of skill (Dingel et al., 2013), which helps instructors grade students fairly. Instead of using the final signature assignment scores for every group member, some instructors prefer to weigh the peer assessment results and add them to individual students’ final scores for fairness reasons. Research also confirmed that students' scores under this approach can be seen as an objective truth only when students are completely truthful in their assessment. However, some factors could play a role that would affect the truthfulness of student scoring, such as relationships with others.

Additionally, instructors should understand that group-based assessment is based on specific tasks contributed by group members over a certain period. Individual learning development throughout the process should not be overlooked. Common tools for assessing individual student learning include reflections, systematic observations, and student interviews, which were also employed in our study.

Assessment Practices in a Pilot Study

An instructor and an assessment professional collaborated over three years to design an undergraduate required course about Research Design and Methods for sophomores within the Department of Public Administration at a large research university. The student learning outcomes were: (1) Knowledge: Students will be able to remember research methods and apply them to appropriate scenarios.  (2) Lifelong Learning: Students will seek opportunities to expand their knowledge or collaborative skills in group-based projects.  (3) Intercultural Competence: Students will be able to value their interactions with culturally different others. The following table summarizes the assessment tools used for this study.

Table 1:

The course enrolled approximately 38 students per year. At the beginning of the semester, instead of letting students choose their group members, the research team intentionally grouped them with different class levels, majors, genders, motivations, research interests, and work styles. The week after the drop-add period, students had their first group activity via an ice-breaking activity. Throughout the semester, the instructor gave each group 1-2 opportunities for weekly group discussion. Each discussion lasted approximately 30-50 minutes in this three-hour class, and students sat with their group members to discuss the assigned tasks. The discussion time allowed collaborative learning and ensured that their projects progressed during the class. 

A pre-post survey and in-class observation aimed to examine the progress and development of the individual’s lifelong skills and intercultural competence over the semester were administered to assess SLO2 and SLO3. By adopting the American Association of Colleges and Universities VALUE rubric on Teamwork, the research team developed an observation rubric and pre-post survey that covers the following aspects: (1) Individual roles in team participation, (2) Interaction between oneself and team members, (3) Contributions to fostering a constructive team atmosphere, (4) Conflict management, (5) Personal contributions outside of group meetings (Pre-post survey only) (6) Personal beliefs about cooperative learning (Pre-post survey only), (7) Self-assessment (Pre-post survey only). The instructor and teaching assistants proactively observed different groups during the designated group discussion times. As learning in a group-based context is the project design, the instructors and teaching assistants only provide general directions for each group instead of concrete answers or making decisions for the group. Additionally, the teaching assistants observed the group dynamics and scored each student's learning behavior using the pre-developed rubric. 

Peer assessment was conducted twice during the semester. The first assessment occurred in the middle of the semester, and the second was conducted at the end. These assessments allowed group members to have a chance to share their observations and feelings toward other members. The results support previous scholars’ studies that peer assessment would help students be aware of the task and give motivated students a chance to disclose any unbalanced situation. However, it is not significant to see that peer assessment would enhance motivation for low-engagement students.

Reflection assignments were required in the early weeks of the semester and near the end of the semester. The American Association of Colleges and Universities VALUE rubric on Lifetime Learning and Intercultural Competence rubrics were adopted for scoring. 

This study's preliminary results indicate that students with greater motivation to learn in the class achieve higher knowledge scores. Additionally, a statistically significant increase in lifelong learning and intercultural understanding was observed after conducting a paired t-test. As a student mentioned in her reflection, “This class requires actual participation and idea contribution to drive the tasks forward. As a result, I have learned how to collaborate and get along with my peers. My attitude has also become more proactive and positive.” While students initially found the teacher-assigned grouping worrisome, their concerns began to diminish after the middle of the semester, and students started to appreciate the value of working on a group-based assignment. In the instructor’s teaching journal, she reflected on how different assessment activities are critical to show various aspects of student learning growth, "Group 6 is probably the most concerning group at the moment, but based on [student name]'s reflection, they actually have group discussions via messaging after class, which is something we cannot observe but can be supplemented through their reflections." In addition, our study results also support the idea that peer assessment serves as a tool for high-engagement students to disclose any unfair experiences, as students were aware that group project scores would be adjusted based on observation and peer assessment.

Conclusion

Balancing scoring fairness and promoting collaborative learning to achieve higher attainment on various learning outcomes can be challenging; however, with cautious course design, this goal is achievable. We are not surprised that students still expressed concerns about peers not taking responsibility for their assigned tasks in the early stage of the semester; through the strategic use of assessment tools, students expressed a deeper appreciation for the value of group project design. Several students said they had acquired valuable skills applicable to their future careers. Our pilot study results indicate that student learning outcome assessment in a group-based learning context should not be limited to academic performance but should consider including affective or social interaction outcomes.

Recommendations

While group-based learning is a common practice, the effectiveness of student collaboration can vary significantly. It is essential to recognize that individual students' efforts may differ during the process, and the final presentation may not accurately reflect each student's learning progress. From our study results, we recommend that Assessment Professionals focus not just on cognitive growth but also consider having an affective learning outcome as a critical learning result from the course. To effectively administer multiple assessment activities, we suggest instructors review assessment results throughout the course and address any issues, such as a lack of effort in engaging in conversations, which may eventually affect students' value in collaborative learning or the quality of final artifacts.