Purpose. The main purpose of this study is to verify the accuracy of body composition measurements by the BIA method, using TANITA 418 MA unit. Basic procedures. For purposes of our study we decided to use the type-A uncertainty evaluation method (statistical processing of measured data). To formulate diagnostic error for the BIA method we used typical (standard) error of measurement from differences between pairs of repeated measurements - which is in principle a mean error of the test. The research was carried out on a group of 10 students whose field of study is physical education and sport. These students were chosen from the total number of 74 students in such a way that they represent as wide range of fat fraction as possible. Representation of their fat fraction oscillated between 9.0 to 31.2% of body fat. Input value of body fat, which was the basic criterion for selection of students to the measurement etalon, was determined by the DEXA method. Main findings. We described the uncertainty (type A) in selected file of etalons by an average value and by a standard deviation in percentage of fat: 0.13 ± 0.05. A diagnostic error of the method expressed by means of typical (standard) error of measurement was 0.38% of body fat. Conclusion. Considering the detected values of uncertainty (type A) and of typical error of measurement, we can conclude that measurement of body composition by the BIA method using TANITA 418 MA unit is sufficiently accurate for requirements of practical use. The necessary condition for the above mentioned measurement accuracy is a strict observance of standard measurement conditions.