Eating frequency has been implicated in the risk of colorectal cancer (CRC) in several epidemiological studies with contradictory and inconclusive findings. We performed a meta-analysis to evaluate their relationship. The pooled relative risk (RR) with 95% confidence interval (CI) was calculated to estimate the effects. A total of 15 eligible studies with 141,431 subjects and 11,248 cases were retrieved after a comprehensive search of the PubMed, Cochrane Library, and Web of Science databases up to October 2013. The overall meta-analysis revealed no strong significant association between eating frequency and risk of CRC in different eating occasion categories (1 meal/day): RR = 1.01, 95% CI 0.94-1.09, P = 0.709; 3 vs. <3 daily meals: RR = 1.17, 95% CI 0.93-1.46; 4 vs. <3 daily meals: RR = 1.13, 95% CI 0.92-1.38; ≥ 5 vs. <3 daily meals: RR = 0.95, 95% CI 0.61-1.47; 4 vs. ≤ 3 daily meals: RR = 1.18, 95% CI 0.92-1.51; and 1-2 vs. 3 or 4 daily meals: RR = 0.82, 95% CI 0.63-1.06). However, modest evidence of an increased risk of CRC in case-control studies (RR = 1.30; 95% CI, 1.11-1.52) and ≥ 5 vs. ≤ 3 meals group (RR = 1.30; 95% CI, 1.11-1.52) was observed. Our meta-analysis results do not support the hypothesis that eating frequency strongly reduced or increased the risk of CRC. Clinical randomized trials are required to evaluate this relationship further.