OBJECTIVE:: Current practice during pediatric colonoscopy is to obtain random colonic biopsies, even from normal appearing tissue. The majority of literature published on colonic biopsy practice analyzes adults; however, limited data examines the pediatric population. The lack of standardization regarding tissue sampling during pediatric colonoscopy reinforces the necessity to study this question objectively. The aim of this study was to assess the value of obtaining mucosal biopsies from grossly normal colonic tissue. METHODS:: A retrospective study was performed to examine the utility of obtaining colonic biopsies from normal appearing tissue. Subjects included patients who underwent colonoscopy over a two-year period. Descriptive analyses and logistic regression models were used to determine endoscopic and histologic agreement and to characterize predictors of agreement. RESULTS:: The predictive value of agreement between normal appearing colonoscopies (n = 237) and histopathology was 81%. Excluding patients with inflammatory bowel disease (IBD) increased the predictive value to 85%. Two and a half percent of grossly normal colonoscopies had histological findings consistent with newly-diagnosed IBD. Predictors of normal histology included abdominal pain (odds ratio [OR] 3.68, 95% CI: 1.84–7.37) while a known diagnosis of IBD was predictive of abnormal histology (OR 0.22, 95% CI: 0.08–0.67). Amongst the full sample, elevated inflammatory markers (OR 2.02, 95% CI: 1.06–3.86) or fatigue (OR 2.60, 95% CI: 1.13–5.94) were predictive of abnormal histology. CONCLUSION:: Results suggest that colonic biopsies may not need to be routinely obtained from all pediatric colonoscopies, particularly in those patients with complaints of abdominal pain. Biopsies should continue to be obtained from patients with a known diagnosis of IBD, elevated inflammatory markers, or fatigue. Further studies are needed to standardize protocols for biopsy practice in pediatric colonoscopy.