<p>Peer led team learning (PLTL) has served as a successful pedagogical innovation for decades. The more recent implementation of cyber peer lead team learning (cPLTL) has shown promise in student outcomes within the online space. This study utilized the combination of social constructivism, reflective problem-solving processes, observational learning and cognitive development frameworks to underpin the interactions between the student groups. The California Critical Thinking Disposition was adapted into a qualitative analysis rubric aimed at investigating the critical thinking development of cPLTL students. The scales were graded according to how students demonstrated their problem-solving skills as they related to the levels of explanation, interpretation, and analysis. Nested within these categories, student engagement was scored on a three-unit scale according to basic, intermediate, and advanced levels of observed problem-solving skills. The recordings of two student cPLTL groups were used in this study of approximately 7 students per group, meeting once a week for five weeks. The interpretation category showed the greatest number of codable instances among the students as well as growth areas for their problem-solving skills. Analysis was the most consistent area that was maintained throughout the recorded weeks. Explanation, the lowest critical thinking level, decreased over the weeks. These differences observed from lower ordered into higher ordered thinking provides a brief snapshot suggesting student participation in cPLTL provides growth in student problem-solving skills within the social learning environment. The influence of peer leaders emerged as an essential element in facilitating cognitive development among students, supporting Bandura’s concepts of observational learning and modeling. The implications of this work provide PLTL and cPLTL practitioners with a theoretically grounded method to qualitatively evaluate the development of problem-solving skills in collaborative online environments. However, contextual differences relating to the student online environment, issues with consistent student attendance, and the small sample size limit the generalizability of the results.</p>