What does the design of effective CSCL require and how do we study its effects?

Abstract

looks at patterns or clusters of results, syndromes (my word) using "Small Space Analyses" (guttman, 1969, psychometric) argues against group performance as a measure but rather "individual cognitive residues" learnign in all but the most restrictive situations requires mindful engagement under learner control computers start a 'chain reaction' that leads to a whole host of changes (package) collaborative learning requires'well orchestrated interdependence'...not just cooperative, makes some of the collaboration isn't always good arguments. good citation for collaborative premise (my word))