A new method was developed for determination of damaged starch in wheat flour. The method was based on the idea that the difference between the soluble matter contents of the extracts prepared from the same flour with and without alpha-amylase is mainly due to the action of alpha-amylase on damaged granules. The extraction stage of the method is similar to the AACC method (76-30A). However, after this stage a refractometer was used to measure Brix values of solutions instead of determination of reducing sugars. The sample size and incubation period of the AACC method were optimized for the proposed method using response surface methodology. The damaged starch levels of twenty-five commercial samples were determined by the proposed method as well as the two AACC Methods (No: 76-30A and 76-31). The results of the proposed method are compared with the results of the AACC Methods by using regression analysis. The correlation coefficients between the proposed method and both of the AACC Methods were 0.99. The average absolute differences between the proposed method and the AACC methods were 3.9 and 5.0%, respectively. The proposed method can be completed in 30 min with low number of experimental steps. The results showed that the starch damage values measured by the new method could be viewed with as much confidence as the traditional AACC Method 76-30A. (C) 2003 Elsevier Ltd. All rights reserved.