Current second language (L2) instruction research has encouraged the use of collaborative output tasks in L2 classrooms. This study examined and compared the effectiveness of two types of output tasks (reconstruction cloze tasks and reconstruction editing tasks) for learning English phrasal verbs. Of interest was whether doing the tasks collaboratively led to greater gains of knowledge of the target verbs than doing the tasks individually, and also whether the type of task made a difference. The study was conducted in two intact low-intermediate adult English-as-a-second-language (ESL) classrooms. The effectiveness of the tasks was determined by how successfully learners completed the tasks and also by means of a vocabulary knowledge test administered before and after the treatment. The results showed that completing the tasks collaboratively (in pairs) led to a greater accuracy of task completion than completing them individually. However, collaborative tasks did not lead to significantly greater gains of vocabulary knowledge than individual tasks. The results, however, showed an effect of task type, with the editing tasks being more effective than the cloze tasks in promoting negotiation and learning. The findings contribute to the research that has examined the effectiveness of pedagogical tasks in L2 classrooms. (Contains 5 figures and 6 tables.)