RESULTS OF THE WORLD'S FAIR HEARING TESTS 541 



such distributions for the 35,589 test scores that were used in calcu- 

 lating the mean hearing loss values of Table 2. It is arranged to show, 

 for each of the five tones, the cumulative distributions separately for 

 men and women in the five age ranges. It gives the percentage of 

 tests failing to show a correct response at the test step indicated, or 

 the percentage of individuals having a greater hearing loss than that 

 corresponding to the indicated step. For example, the table shows 

 that only 0.7 per cent of the men in the 20-29 age group have hearing 

 losses greater than 32 db for a 1760-cycle tone, while' 11 per cent of 

 those in the 50-59 group have this much loss. 



Zero hearing loss falls between steps 7 and 8 for the first three tones, 

 and between steps 6 and 7 for the last two tones. The last step corre- 

 sponds to very good hearing, and individuals able to hear this step 

 have hearing acuities at least 10 db better than average. Some 10 

 or 15 per cent of the youngest age group, but only 1 or 2 per cent of 

 the oldest group, were able to hear the tones on the last step. For the 

 tone of lowest frequency, there were seven young persons for every 

 older person who could hear the last step, but at 7040 cycles, there 

 were 18 young persons for every such older person. 



The tabular data for the age groups 20-29 and 50-59 are shown 

 graphically in Fig. 4 for four of the tones, beginning at 880 cycles.'* 

 The curves are cumulative distributions and the ordinate gives the per- 

 centage of individuals having hearing losses greater than the value 

 indicated by the abscissa.^ At 880 cycles hearing losses in excess of a 

 given amount tend to be more prevalent among women than among 

 men. At 1760 cycles the distribution curves for men and women are 

 much the same. At the two higher frequencies, the prevalence of 

 deafness in excess of a given amount is greater among men than 

 among women. 



A hearing loss of 25 db at frequencies up to 1760 cycles begins to 

 be a handicap. The individual will usually be aware of such an im- 

 pairment, and will experience difficulty in understanding speech under 

 conditions of public address, such as in the church or theater or around 

 the conference or dinner table. The distribution curves show that 

 only about 1.5 per cent of the young people taking the test, or three 

 out of 200, have a hearing loss of 25 db or more for tones of these 

 frequencies. In the oldest age range almost ten times as many, or 

 every seventh person, shows this much impairment. 



^ The distributions for 880 cycles may be used for 440 cycles as well, it being as- 

 sumed that the small difference shown in Table 4 is due to practice. 



^ The ordinates are shown on an arithmetic probability scale, which has the prop- 

 erty that a normal distribution plots as a straight line whose slope is proportional to 

 the standard deviation of the distribution. It is convenient because it shows the 

 small values more accurately and because on this scale the standard errors of the 

 ordinates are approximately equal in all parts of the range. 



