Previous work quantified a 59% reduction in injury risk for children in belt positioning booster seats (BPB) compared to those restrained in seat belts using a sample of crashes of all directions of impact. Experimental sled tests have highlighted the potential for extreme occupant excursion out of the BPB in side impact crash conditions. Using data from a large child specific crash surveillance system, the present study built upon these previous studies and quantified the relative effectiveness of BPB as compared to seat belts in reducing the risk of injury among 4-8 year olds in side impact crashes. Children in BPB were at a 58% reduction in risk of injury than those in seat belts in side impact crashes. This result varied by booster seat type: those in high back BPB were at a 70% reduction in injury risk while those in backless BPB did not experience a statistically significant reduction in injury risk compared to those in seat belts. This differential performance of the two types of BPB provides direction for future research into the design and performance of these restraints.