Studies in research on Computer Supported Collaborative Learning (CSCL) usually gain their scientific findings from pre‐/post‐tests, video or logfile analyses. Although eye movements have proved to be a valuable source of information for the study of cognitive processes, they are hardly regarded in the field of CSCL. A crucial reason for this is the lack of suitable observational schemes. To bridge this gap, we propose a categorial coding scheme for global text processing in CSCL on the base of established well‐defined eye movement measures. The empirical examination showed its high inter‐rater reliability (κM = 0.91). Implications for CSCL are discussed.