Equal-loudness level contours are assumed to reflect the sensitivity across frequency of the auditory system. However, studies on equal-loudness level contours show considerable differences between laboratories even when the same measurement method is employed. This study investigates how experimentally obtained equal-loudness level contours using the method of constant stimuli are affected by the individual listener, the measurement room and the exact choice of test-tone levels. Concerning the individual listener, measurements revealed that equal-loudness level contours can be lower in the initial than in subsequent measurements. The measurement room does not significantly influence the results and the distribution of test-tone level presentations only slightly affects equal-loudness level contours. In addition, the size of the range of levels employed also has no influence, provided that the range includes the point of subjective equality (PSE). However, the absolute position of the test-tone level range strongly affects the resulting shape of equal-loudness level contours. The experiments provide strong evidence that the differences between recent studies are caused by the different test-tone level ranges employed. Furthermore, it is suggested that discrepancies between recent equal-loudness level contours and the contours specified in ISO 226 are also due to the level range of the test tones employed.
