Studies investigating the association between cytochrome glutathione S-transferase P1 (GSTP1) Ala114Val polymorphism and colorectal cancer (CRC) risk report conflicting results. The aim of this study was to quantitatively summarize the evidence for such a relationship. Two investigators independently searched the Medline, Embase, China National Knowledge Infrastructure, and Chinese Biomedicine databases. Summary odds ratios (ORs) and 95 % confidence intervals (95 % CIs) for GSTP1 polymorphism and CRC were calculated in a fixed effects model (the Mantel-Haenszel method) and a random effects model (the DerSimonian and Laird method) when appropriate. The pooled ORs were performed for co-dominant model (ValVal vs. AlaAla, AlaVal vs. AlaAla), dominant model (ValVal + AlaVal vs. AlaAla), and recessive model (ValVal vs. AlaVal + AlaAla). This meta-analysis included seven case-control studies, which included 3,173 CRC cases and 3,323 controls. Overall, the variant genotypes (ValVal and AlaVal) of the Ala114Val were not associated with CRC risk when compared with the wild-type AlaAla homozygote. Similarly, no associations were found in the dominant and recessive models. When stratifying for ethnicity, Hardy-Weinberg equilibrium in controls, study sample size, and source of controls, a significantly increased risk was observed among Asians (AlaVal vs. AlaAla, OR=1.67, 95 % CI=1.08-2.59; dominant model, OR=1.74, 95 % CI=1.14-2.67). No heterogeneity or publication bias was found in the present study. This meta-analysis suggests that the GSTP1 Ala114Val polymorphism may not be associated with CRC risk, while the observed increase in risk of CRC may be due to small-study bias.