Methodological Considerations for Quantitative Content Analysis of Online Interactions entity as seen in quantity-oriented content analysis methods. He further argues that many of current analysis practices in CSCL research fail to capture knowledge building processes as a function of group cognition. If we accept that through the mediation of technological tools members in virtual com- munities can engage in shared-meaning making processes and collectively build knowledge which is greater than the sum of individual knowledge, then this theoretical view necessitates methods considering the importance of multiple units or levels of analysis. Another methodological chal- lenge in quantitative content analysis methods is related to time and sequence: how can we ana- lyze interaction in online or virtual communities unfolding over long time scales, for example, over a few months(Reiman, 2009)? Interaction in virtual communities is often unpredictable and new participation structure may emerge through Chernobilsky, & Jordan, 2008) can be used to analyze multiple processes plotted on a timeline. For the artistically inclined, Donath and Viégas (2002) have developed creative ways of using graphical representations to create user's identity, visualizing social activities and depicting expres- sive communication of online conversation. Next, advanced statistical techniques presented new op- portunities to reveal the effects of interactions on individuals and on groups. Cress (2008) presents how multilevel modeling (MLM) tests interaction- al effects of multiple within and between groups. Event-based modeling (e.g., stochastic modeling) is another statistical approach that takes into ac- count time and sequence factors (Reiman, 2009). FUTURE RESEARCH DIRECTIONS Despite the long history of content analysis as