This paper describes a sequence of lessons based on a modified version of the ESS Colored Solutions problem posed in an urban multicultural middle school setting in which students often worked in collaborative groups. As students encountered observable phenomena, they were encouraged to record techniques and observations, search for patterns in their data, and eventually develop explanations. The researchers examined discourse and interactions in working pairs, groups of four, and individual logbook entries to see how various tasks, settings, and interactions influenced engagement for each of the members of the group studied. Each member of the student working groups was observed to be fully engaged in the activities of developing and recording techniques and observations and finding patterns. The activity of developing explanations, however, revealed differential engagement based upon whether or not each student had focused on properties of the solutions as the salient feature of the system being studied. The researchers hypothesize that this differential may have substantial consequences when students encounter formal instruction in related scientific concepts. Those more engaged in generating explanations appear more ready to appropriate and use concepts as intellectual tools to explain observed phenomena. (Author)