This study investigates the comparative effect of teaching collocations through practicing them in literary and non-literary contents. The participants were composed of 30 EFL students selected from a cohort of 52 intermediate students in an English language institute in Sanandaj City, Iran. They were selected based on their scores on the Preliminary English Test (PET) and a collocation test, developed and piloted in advance. The selected participants were randomly divided into two experimental groups. Collocations were taught through literary content in one group; the others were instructed via non-literary content. At the end of the period of treatment with both groups, a collocation test was administered to both groups as a posttest. The analysis of collected data, using One-way ANCOVA and Descriptive Statistics, reveals that teaching the new collocations through literary contents proved significantly more effective than teaching them through non-literary contents.