Crossley and colleagues (2015) investigated the relationship between student lan-guage on the discussion forum and course completion. This research, complementary to other analyses of MOOC forum data that looked at the presence of specific words (Wen, Yang, & Rose, 2014), used natural language processing (NLP) to examine whether the types of language in the discussion forum of an EDM MOOC is predic-tive of successful class completion.The analysis was applied to a subsample of 320 students who completed at least one graded assignment and produced at least 50 words in discussion forums. The findings indicate that the language produced by students can predict substan-tially better than chance (cross‐validated Cohen’s kappa=0.379) whether students complete the MOOC. While many students did not use the discussion forums, stu-dents who participated more frequently in the forums, who used a greater range of vocabulary in the discussion forum, and who used more concrete and sophisticated words were more likely to complete the MOOC. These results suggest that NLP can help us to better understand student retention in MOOCs