Background : The purpose of this study was to determine whether cleansing the perineum and urethral meatus and using midstream urine affect the rate of bacterial contamination of urine specimens, and to determine the optimum urine collection method. We studied 41 asymptomatic healthy nursing school students. Women who were menstruating were not excluded from this study. Method : The first and midstream urine samples were collected during consecutive urinationsby each woman. The first sample was not a clean-catch specimen, and the second one was a clean-catch specimen. Both specimens were studied by urinalysis and bacterial culture with standard methods. Results : 41 women met the study criteria and 39 successfully completed the study. None of the urine cultures were positive. 68.3% of the non clean-catch first urine cultures, 53.7% of the non clean-catch midstream cultures, 33.3% of the first clean-catch urine culteres and 30.8% of the midstream clean-catch urine were found to be contaminated. There was a significant difference in the bacterial contamination rates between the first and midstream urine, and the clean-catch and non clean-catch urine(p=0.035, p =0.001 respectively). On urinalysis, 7.3% of the non clean-catch first urine, 7.3% of the non clean-catch midstream urine, 2.6% of the clean-catch first urine and 2.6% of clean-catch midstream urine were found to be above grade 2. Conclusions : According to our results, the bacterial contamination rate was the lowest in midstream and clean catch urine specimens. Threrfore it is recommended that the midstream clean-catch technique is the standard practice for collecting urine specimens for bacterial culture in women.