Background: Colorectal cancer (CRC) is the third common cancer and the fourth leading cause of cancer death worldwide. A single nucleotide polymorphism (SNP), rs961253 located in 20p12, was firstly described to be associated with the increased risk of CRC in a genome-wide association study; however, more recent replication studies yielded controversial results. Methodology/Principal Findings: A hospital-based case-control study in a Chinese population was firstly performed, and then a meta-analysis combining the current and previously published studies were conducted to explore the real effect of rs961253 in CRC susceptibility. In the Chinese population including 641 cases and 1037 controls, per-A-allele conferred an OR of 1.60 (95% CI = 1.26-2.02) under additive model. In the meta-analysis including 29859 cases and 29696 controls, per-A-allele have an OR of 1.13 (95% CI = 1.09-1.18) under a random-effects model due to heterogeneity (P = 0.019). Nevertheless, the heterogeneity can be totally explained by ethnicity, with the tau(2)reduced to 0 after including ethnicity in meta-regression model. In stratified analysis by ethnicity, per-A-allele had ORs of 1.34 (95% CI = 1.20-1.50) and 1.11 (95% CI = 1.08-1.14) for Asian and European, respectively, without heterogeneity. Modest influence of each study was observed on overall estimate in sensitive analysis, and evident tendency to significant association was seen in cumulative analysis over time, together indicating the robust stability of the current results. Conclusions/Significance: The results from our study and the meta-analysis provided firm evidence that rs961253 significantly contributed to CRC risk in both Asian and European population.