Uncertainty can be viewed as a mechanism by which researchers can examine speaker commitment in both bona-fide and deceptive communication. We apply the examination of varying levels of uncertainty in text to the online grooming domain. Within online grooming literature, researchers use data sets consisting of real victims, vigilantes, and law enforcement officers. However, psychology researchers question whether internet stings involving law enforcement officers (LEO) and vigilantes can be used to approximate the grooming process of children. In this paper, we examine differences between the three groups within the stages of the grooming process. We approach the problem through the lens of uncertainty language. We find the use of uncertain language differs the most between victims, vigilantes, and LEO in the meeting, sexual, and risk assessment stages.