Collaborative learning, particularly in the context of team-based, project-based learning, is common in undergraduate engineering education and is associated with deeper learning and enhanced student motivation and retention. However, grouping students in teams for project-based learning sometimes has negative outcomes, which can include lowered participation by sub-groups of students, unwillingness or inability of students to critique peers' ideas, and decreased tendency of students to provide externalized rationales in support of their perspectives. Online chat was investigated as a medium for improving group conversations because it allows multiple students to contribute simultaneously, because it allows students to avoid some of the social pressures associated with meeting face-to-face (F2F), and because its text-based nature leaves a record of the conversation. Student design teams of 4 or 5 students were assigned to F2F (n = 73 students on 17 teams) or Google Drawing tool with synchronous chat (n = 158 students on 37 teams) contexts for a single design negotiation conversation at the beginning of a design-build-test cycle in a first year "Introduction to Engineering" course. Participants also completed a post-conversation survey and end-of-semester team evaluation. Transcripts of the design conversations and student reflections were analyzed to compare student participation and characteristics of the design conversation in F2F versus online environments. The analyses showed that team member contribution levels were more even in the synchronous chat tool compared to the F2F conversations. Student perspectives on the post-conversation survey supported this quantitative result, with some F2F students reporting lowered ability to contribute due to academic intimidation and to unwillingness to interrupt team members. The same pattern was observed with female students. Many women contributed to F2F team conversations at particularly high or particularly low levels relative to their teammates; women online participated at more average levels. Non-native speakers on teams meeting F2F contributed at significantly lower levels than their native-English-speaking peers, but non-native speakers on teams meeting via synchronous chat contributed at average levels on their teams. The conversations were similar in terms of what the students discussed and how they did so rhetorically, with a few exceptions: Online groups spent less time on course logistics and more time discussing the to-be-built object compared to their peers meeting F2F. Students meeting in synchronous chat were significantly more likely to express disagreement and more likely to provide explicit rationales in support of their positions. This study suggests that instructors should consider moving some group conversations of first-year engineering project-based learning teams to synchronous chat contexts, to better balance participation and to increase characteristics of collaboration that lead to learning (disagreement and explicit rationalizations). Further research should investigate whether the discourse patterns initiated in the synchronous chat meetings persist when teams move to F2F contexts, such as the laboratory. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]