The purpose of the study is to evaluate the extent to which a retrospectively conducted translatability assessment (TA) could identify the items previously singled out during the validation study as having poor content validity or poor measurement performance. This study was performed with the intent of supporting evidence of the usefulness of TA early in the PRO development process. The Weight module of the Youth Quality-of-Life Instrument (YQOL-W) was used for this appraisal of translatability.
Two linguists, blinded to the design and results of content validation and psychometric analyses, conducted a TA on the 32-item version of the YQOL-W taken into cross-sectional validation for item reduction. TA results were categorized into (1) issues relating to target culture (cross-cultural issues) and (2) issues relating to the structure of the original questionnaire (structural issues). Items for possible revision or deletion were identified. We compared the results of the TA with the content validity and psychometric results and decisions to eliminate items after cross-sectional validation.
Content validation identified seven of the 32 items to be dropped, and psychometric analyses including the 25 remaining items highlighted an additional four to be eliminated, yielding a final instrument with 21 items. Out of these 11 dropped items, TA had identified nine as problematic (82 %) and the developer was advised to drop five of them (45.4 %). In addition, TA results highlighted the need to change the original formulation of eight items for semantic reasons and identified two instances where alternative wording should be used for translation purposes without any change to the original formulation.
Our study showed that translatability assessment confirmed problematic issues in items previously identified as having poor content validity or poor measurement performance. In general, a translatability assessment offers the possibility for the identification of alternative formulations for translation purposes, modifications of original formulations to optimize subsequent translations efforts, and the early detection and discussion of irrelevant or inappropriate items.
Authored by Katrin Conway, Catherine Acquadro & Donald L. Patrick