This paper presents a systemic intervention approach as a way to overcome the methodological challenges involved in research on Computer-Supported Collaborative Learning (CSCL). These challenges include how to choose between qualitative and quantitative analysis, and between predictive and descriptive studies. In addition, there is a lack of agreement in the research community regarding the appropriate methods and measures to analyze several dimensions of the learning process, including the cognitive, motivational, social and technological dimensions. The final challenge is a need for further research on how participants should engage in defining and re-defining learning purposes, interests and outcomes. The up-front focus of systemic intervention is on the process of making boundary judgments during inquiry. Therefore, it facilitates questioning about whose views and what issues should be considered pertinent in an analysis. Definitions of improvement, the purposes of inquiry, the context of application of methods, and participants’ roles are all important in systemic intervention. The focus is on both the issues ‘out there’, and the identities and roles of the people (including the researcher) who get to frame those issues. In addition, the systemic intervention approach advocates methodological pluralism: mixing methods from different traditions (quantitative and qualitative; predictive and descriptive) to address the purposes of multiple stakeholders. In line with this focus on methodological pluralism, a design for CSCL research will be presented that includes the use of several methods drawn from other methodologies (Critical Systems Heuristics, Interactive Planning, Social Network Analysis, Formative and Summative Tests and Content Analysis). This methodological design will be used in future research to analyze and reflect on both a Colombian CSCL case study and the roles and identities of the people who participate in it.