Shortened hearing aid performance inventory for the elderly (SHAPIE): A statistical approach

The shortened hearing aid performance inventory for the elderly (SHAPIE) is an adaption of the Hearing Aid Performance Inventory (HAPI) (Walden, Demorest and Hepler, 1984, J. Speech. Hear. Res., 27, 49-56). The SHAPIE was derived from the HAPI in three stages. First, nine items that were clearly inapplicable to many elderly people were deleted. Second, published data were used to delete a further 15 items on the grounds of low item-total correlation, low inter-subject standard deviation, and low factor loadings. Third, 18 of the remaining 40 items were modified and the 40-item questionnaire was administered to 107 hearing impaired subjects. Fourth, a further 15 items were deleted using similar statistics derived from these Australian hearing impaired subjects. The number of 'not applicable' responses was also considered. The resulting 25-item questionnaire, recommended for general use, had test-retest reliability equivalent to that of the 40-item questionnaire, and a Cronbach's alpha of 0.91. In addition to overall score, it provides subscale scores for listening in noise, listening with reduced cues (e.g. no visual cues) and a miscellaneous subscale incorporating listening in quiet, listening up close, and listening to non-speech sounds.