Purpose Conflicts among video head impulse testing (vHIT) and rotary chair have occurred; therefore, the purpose of this study was to determine the relationship between rotary chair and vHIT outcome parameters to understand when these two tests disagree and determine if one or both test outcomes are needed in children. Method Data from 141 child and young adult subjects (73 males, 68 females, M age = 15 years, range: 6-35) were retrospectively reviewed. Of those, 56 had a cochlear implant and 85 were normal controls. All subjects completed rotary chair and vHIT, which were categorized as (a) normal vestibular function, (b) unilateral vestibular loss, or (c) bilateral vestibular loss. vHIT tracings were analyzed to determine if gain and corrective saccade velocity, frequency, or latency were helpful parameters for determining vestibular loss. Results Of the 141 subjects, the misclassification rate was 13/141 (9%). All normal control subjects were classified as having normal rotary chair and normal vHIT. In subjects with a cochlear implant (n = 56), the misclassification rate was 13/56 (23%). There were four misclassification patterns. Using rotary chair as the gold standard, receiver operating characteristic analysis revealed optimal cut-points for vHIT gain (< 0.84), corrective saccade frequency (≥ 50%), amplitude (≥ 75°/s), and latency (≤ 320 ms). Using these vHIT cut-points improved the agreement between rotary chair and vHIT, resulting in an overall misclassification rate of 10/141 (7%) and 9/56 (16%) in subjects with a cochlear implant. Conclusions When testing children, caloric testing is often not an option due to tolerability or time. However, discordant results occur between rotary chair and vHIT. These data suggest vHIT is a sufficient first-tier assessment. If abnormal, rotary chair is not necessary. If normal, rotary chair can be helpful for uncovering other indicators of vestibular loss. When interpreting vHIT, including gain and all corrective saccade outcomes may improve sensitivity.