Dear AIR members, Could you point me to any research showing the distribution of the trust level of messages? In other words, not all messages exchanged in a virtual community are supportive, many will be neutral and some will be non-supportive or hostile. What does that distribution look like? What percentage of messages are non-supportive/hostile, neutral, supportive? I am developing an agent-based model of social system and need this information to govern the messages generated. I would like the messages exchanged between agents to mimic real messages distributions. Thnk you for your time and consideration -- Leo Holly Doctoral Candidate Executive Leadership Doctoral Program The George Washington University "What lies behind us and what lies before us are tiny matters compared to what lies within us." - Oliver Wendell Holmes