Pilot Study

The objectives of the pilot test were to analyse the content of the DISABKIDS draft questionnaires from the respondent’s perspective, to collect data for first psychometric analyses, and to simulate the field test. The main psychometric aim of the pilot test analyses was to reduce the number of items and select the best performing items according to predefined criteria. Two methods were chosen to check the appropriateness of the DISABKIDS pilot questionnaires: a quantitative psychometric analysis and a qualitative cognitive debriefing. The questionnaire was administered in each of the project countries during an interview with children drawn from the different conditions. The children completed the questionnaire and were then interviewed using a standard procedure to ascertain the acceptance of the questions, the relevance of the questions to them, their understanding of the questions, and if the wording of the questions could be improved / made more acceptable to children.

Data files were provided by the Hamburg study centre and data entry was done in each country. After data entry, the different data sets were merged and checked in terms of data plausibility. First of all, an analysis of the pilot study was carried out at the international level using classical multi-scaling as well as modern psychometric methods. The criteria applied for the item selection were chosen from a psychometric and clinical perspective. In detail, the results of frequency analyses, cognitive debriefing, exploratory as well as confirmatory factor analysis, item correlations, and Rasch analysis were used to decide upon the inclusion or omission of items. Overall, the item selection process resulted in a 56-item-version of the chronic generic module. The field test version of the chronic generic HRQoL module consists of six domains (independence, emotion, social inclusion, social exclusion, limitation and treatment) and showed a good psychometric performance with reliability coefficients ranging from α= .71 to α= .90.

With regard to the development of the disease-specific instruments a working group was formed including members of the German, Dutch, and UK team. It was considered important to make use of both a clinical and a statistical perspective during item selection. Data from various sources were first utilised in order to select variables that performed poorly, such as the feasibility and difficulty of the item, its care disease-specific character and it's association with health status variables. The condition specific questionnaires consist of 14-19 items divided over 2-3 domains.

 

 

Visit our partners website: