The alkali digestion pretreatment method in the United States Environmental Protection Agency (USEPA) Method 3060A could underestimate the content of Cr(VI) in Cr-contaminated soils, especially for soils mixed with chromite ore processing residue (COPR), which leads to a misjudgment of the Cr(VI) level in soils after remediation, causing secondary pollution to the environment. In this study, a new pretreatment method to analyze Cr(VI) concentration in contaminated soils was established. The impacts of soil quality, particle size, alkali digestion time and the rounds of alkali digestion on Cr(VI) detection in contaminated soils was explored and the alkali digestion method was optimized. Compared with USEPA Method 3060A, the alkaline digestion time was prolonged to 6 h and multiple alkali digestion was employed until the amount of Cr(VI) in the last extraction was less than 10% of the total amount of Cr(VI). Because Cr(VI) in COPR is usually embedded in the mineral phase structure, the hydration products were dissolved and Cr(VI) was released gradually during the alkaline digestion process. The amount of Cr(VI) detected showed high correlation coefficients with the percentage of F1 (mild acid-soluble fraction), F2 (reducible fraction) and F4 (residual fraction). The Cr(VI) contents detected by the new alkaline digestion method and USEPA Method 3060A showed significant differences for soil samples mixed with COPR due to their high percentage of residual fraction. This new pretreatment method could quantify more than 90% of Cr(VI) in Cr-contaminated soils, especially those mixed with COPR, which proved to be a promising method for Cr(VI) analysis in soils, before and after remediation.