Since past studies showing that competition has negative effects on group process were primarily conducted in traditional classrooms involving face-to-face situations, this study extends past research by investigating whether the negative effects associated with face-to-face team competition can be mitigated with the support of networking technologies where opponents' identity or proximity can be manipulated and concealed. Three fourth-grade classes were randomly assigned to three different treatment conditions to participate in the study. A synchronous competitive learning system, named "Joyce," was devised to have pairs of students compete by answering questions with another randomly assigned pair of opponents over three instructional sessions. Results from the study indicated that students in the face-to-face team competition condition rated significantly less favorably on inner-group processing than those in the decreased proximity and anonymity conditions at 0.009 level, and 0.067 level, respectively. Data analysis on students' perceptions toward classroom climate revealed that students in the anonymity condition rated significantly more favorably on classroom climate than those in the other two conditions (0.000 level). The results of this study suggest that to promote effective interactions within learning groups and to cultivate a more regulated classroom learning atmosphere, networked team competition where opponents are not identified may be the preferred instructional strategy to adopt, as compared to face-to-face team competition. The findings also suggest that designers of competitive game-based learning systems should include the feature of anonymity in their system to support student learning. The following are appended: (1) Student Perceptual Impressions of In-Group Processing; and (2) Student Perceptual Impressions of Classroom Climate. (Contains 1 figure and 1 table.)