[Purpose] The purpose of this study was to examine the test-retest reliability of the modified Thomas test using lumbo-pelvic stabilization. [Subjects] Thirteen subjects (male=10, female=3) with hip flexor tightness voluntarily participated in the study. [Methods] The participants underwent the modified Thomas test under three conditions: 1) the general modified Thomas test (GM), 2) active lumbo-pelvic stabilization (ALS), and 3) passive lumbo-pelvic stabilization (PLS). Intra-class correlation coefficients (ICC) were used to determine the test-retest reliability of the knee joint angle measurement under three conditions. The standard error of measurement (SEM) and minimal detectable difference (95% confidence interval) (MDD⁹⁵) were calculated for each measurement to assess absolute consistency. [Results] The ALS (ICC = 0.99) and PLS (ICC = 0.98) methods for the modified Thomas test were more reliable than GM method (ICC = 0.97). The MDD⁹⁵ score for the ALS method, 2.35 degrees, indicated that a real difference existed between two testing sessions compared with the scores for the PLS (3.70 degrees) and GM methods (4.17 degrees) [Conclusion] Lumbo-pelvic stabilization is one of the considerations for precise measurement and may help to minimize measurement error when evaluating hip flexor tightness using the modified Thomas test.