Can new methods solve old problems in the assessment of well-being? Item response theory, itembanking, and computer adaptive test exemplified by the KIDSCREEN study

Most of today's multi-item instruments for the assessment of well-being and health-related quality of life are rather long - whereas shorter instruments often lack measurement precision. Furthermore, the measurement results of the different tools are hardly comparable. These shortcomings hinder the practical application of well-being assessment for clinical and scientific purposes. Exemplified by the international KIDSCREEN project, it is shown how modern methods of item response theory (IRT), item banking, and computer adaptive test ( CAT) could help solve these problems. The 6 items of the IRT scaled psychological well-being dimension of the KIDSCREEN-52 Quality of Life instrument and 20 additional items from 5 other quality of life instruments could be calibrated on a common metric. The developed item bank increased measurement precision from alpha = .81 to alpha = .92 over a wide range of measurements and enables individual diagnosis and the comparison of measurements results across instruments. The item bank can be used for individual diagnosis as part of a CAT.