Face and content validity analysis of the Speech, Spatial and Qualities of Hearing Scale for Parents (SSQ-P) when used in a clinical service without interviews or week-long observation periods

Objectives: To assess the face and content validity of the Speech, Spatial and Qualities of Hearing Scale for Parents (SSQ-P) when used in a clinical setting without the recommended interviews and observation periods.

Methods: SSQ-P responses completed by 145 parents of children with bilateral cochlear implants (aged between 5 and 16 years old) were analysed. To assess face validity, the proportion of missing/ambiguous and alternative responses was recorded for each of the 23 items. Where additional written comments were included in responses, a thematic-based analysis was used to identify reasons for the missing/ambiguous or alternative responses. Content validity was assessed using item response theory (IRT), with items having information score less than 0.5 and discrimination score less than 2.0 identified as poorly performing items.

Results: All items of the SSQ-P exhibited some proportion of missing/ambiguous or alternative responses, with six items having >10% missing/ambiguous or alternative responses. IRT identified thirteen items that performed poorly in terms of information and discrimination. These included four of the six items with the most missing/ambiguous or alternative responses.

Conclusions: SSQ-P items that performed worse tended to describe scenarios that parents perceived as too specific, too vague or hazardous. Without the recommended administration via interviews following three week-long observation periods, parents found these items difficult to complete. The SSQ-P is therefore not recommended for use without the recommended administration method. However, several items performed well in terms of face and content validity, despite independent parent completion without formal observation periods. Thematic analysis suggested that minor re-wording might improve the face validity of items with high content validity but a high proportion of missing/ambiguous or alternative responses. Therefore, the results of the analyses form the basis on which a shortened version of the SSQ-P, more suitable for use in a clinical setting, could be developed in future studies.