Validation of a French-Language Version of the Spatial Hearing Questionnaire, Cluster Analysis and Comparison with the Speech, Spatial, and Qualities of Hearing Scale

Objectives: To validate a French-language version of the spatial hearing questionnaire (SHQ), including investigating its internal structure using cluster analysis and exploring its construct validity on a large population of hearing-impaired (HI) and normal-hearing (NH) subjects, and to compare the SHQ with the speech, spatial, and qualities of hearing scale (SSQ) in the same population. Design: The SHQ was translated in accordance with the principles of the Universalist Model of cross-cultural adaptation of patient-reported outcome instruments. The SSQ and SHQ were then presented in a counterbalanced order, in a self-report mode, in a population of 230 HI subjects (mean age = 54 years and pure-tone audiometry [PTA] on the better ear = 28 dB HL) and 100 NH subjects (mean age = 21 years). The SHQ feasibility, readability, and psychometric properties were systematically investigated using reliability indices, cluster, and factor analyses and multiregression analyses. SHQ characteristics were compared both to different literature data obtained with different language versions and to the SSQ scores obtained in the same population. Results: Internal validity was high and very good reproducibility of scores and intersubject variability were obtained across the 24 items between the English and French SHQ for NH subjects. Factor and cluster analyses concurred in identifying five correlated factors, corresponding to several SHQ subscales: (1) speech in noise (corresponding to SHQ subscales 7 and 8), (2) localization of voice sounds from behind, (3) speech in quiet (corresponding to SHQ subscale 1), (4) localization of everyday sounds, and (5) localization of voices and music (corresponding to parts of the SHQ localization subscale). Correlations between SSQ subscales and SHQ factors identified the greatest correlations between SHQ factors 2, 4, and 5 and SSQ spatial subscales, whereas SHQ factor 1 had the greatest correlation with SSQ_speech. SHQ and SSQ scores were similar, whether in NH subjects (8.5 versus 8.4) or in HI subjects (6.6 for both), sharing more than 80% of variance. The SHQ localization subscale gave similar scores as the SSQ spatial subscale, sharing more than 75% of variance. Construct validity identified better ear PTA and PTA asymmetry as the two main predictors of SHQ scores, to a degree similar to that seen for the SSQ. The SHQ was shorter, easier to read and less sensitive to the number of years of formal education than the SSQ, but this came at a cost of ecological validity, which was rated higher for the SSQ than for the SHQ. Conclusions: A comparison of factor analysis outcomes among the English, Dutch, and French versions of the SHQ confirmed good conceptual equivalence across languages and robustness of the SHQ for use in international settings. In addition, SHQ and SSQ scores showed remarkable similarities, suggesting the possibility of extrapolating the results from one questionnaire to the other. Although the SHQ was originally designed in a population of cochlear implant patients, the present results show that its usefulness could easily be extended to noncochlearimplanted, HI subjects.