In this paper correction factors are derived for a cylindrical NE2571 ionization chamber, for absorbed dose determinations in medium-energy x-rays. These new correction factors are proposed as weighted mean values of the factors derived from two methods. The first method is based on a comparative study of dosimetry of medium-energy x-rays with the cylindrical ionization chamber and a water calorimeter. The calorimetric results show that when the ionization chamber, calibrated free in air, is used to measure the absorbed dose in a water phantom, solely applying mass energy absorption coefficient ratios for the conversion, the ratio of the absorbed dose calorimetry to ionometry is 1.007 +/- 0.015 at 250 kV (HVL: 2.5 mm Cu), 1.026 +/- 0.018 at 150 kV (HVL: 0.82 mm Cu) and 1.040 +/- 0.020 at 100 kV (HVL: 4.54 mm Al). The second method is based on an experimental investigation of the NE2571 ionization chamber in connection with Monte Carlo calculations. Using this method three components in the overall correction factor are investigated: the effect of the displaced volume, the combined angular-energy dependence of the chamber response within the water phantom and the stem correction. The values of the correction factors derived from this investigation were within the uncertainties, in agreement with the values from the first method and varied from 1.015 +/- 0.010 at 2.50 kV (HVL: 2.5 mm Cu) to 1.019 +/- 0.010 at 100 kV (HVL: 4.54 mm Al) with a maximum of 1.025 +/- 0.010 at 150 kV (HVL: 0.82 mm Cu). The correction factors, obtained as a weighted mean of the two methods, are systematically lower than the values proposed in the IAEA TRS-227 report.