As people solve problems together that they cannot do alone, automatic analysis of conversations can inform and enhance their design to aid learning and teaching. Such analyses must traverse the obstacle course of voice transcription, complex categorization, and statistical analysis. Automated transcription feeds automatic categorization via computational linguistics to create a database (Big Data). Automated statistical analysis integrates statistical discourse analysis (SDA) and artificial intelligence. SDA models (a) pivotal actions that radically change subsequent processes and (b) explanatory variables at multiple levels (sequences of turns/messages, time periods, individuals, groups, organizations, etc.) on multiple target actions. The artificial intelligence expert system translates my theory into a statistical model, tests it on the data, interprets the results, (if needed, rewrites itself to execute revised analyses), and prints a table of results. I showcase automated SDA on 321,867 words in 1,330 messages by 17 student-teachers in 13 weekly online discussions of lesson designs.