Many language teachers employ collaborative learning within their classrooms. However, expectations surrounding the efficacy of working collaboratively need to be empirically verified. This study employed dynamic assessment to investigate whether learners who are situated within an undergraduate Qatari EFL context learn second language grammatical structures more effectively by working either collaboratively with their peers or individually. Interventionist dynamic assessment was used to quantify the extent of the learning gains made by male Arabic undergraduate EFL learners (N = 52) three times (pretest, posttest, and delayed posttest) over a 12-week period. In between the pretest and the posttest, six form-focused treatment tasks were administered. The experimental group (n = 20) completed the tasks collaboratively with their peers; a comparison group (n = 16) completed the tasks individually; and a control group (n = 16) did not complete the tasks. The target structures were the simple past passive and the present continuous passive. A Mood's median test (Mood, 1954) found no statistically significant differences between the collaborative condition and the individual condition. Although measuring emergent abilities which are still in the process of developing provided a more complete picture of the efficacy of working collaboratively, the lack of a statistically significant difference between the performances of the experimental and comparison groups for both target structures suggests that working collaboratively is not statistically more effective in facilitating learners' linguistic development than working individually.