Title: 0258 - Rasch Model for the German Child Perceptions Questionnaire (CPQ-G11-14)
Maisa Omara (Presenter)
Medical University of Vienna
Tanja Stamm, Medical University of Vienna
Maren Boecker, RWTH Aachen
Valentin Ritschl, Medical University of Vienna
Erika Mosor, Medical University of Vienna
Thomas Salzberger, University of Economics and Business
Christian Hirsch, University of Leipzig
Katrin Bekes, Medical University of Vienna
Objectives: Oral health related quality of life is essential for general health and contributes to wellbeing. The most frequently used instrument for children aged 11-14-years is the Child Perceptions Questionnaire (CPQ11-14). However, up to now the psychometric properties of this instrument have not been examined using modern statistical approaches, such as the Rasch model. Therefore the aim of the study was to further examine the German version of the CPQ11-14 by diagnosing potential measurement problems and suggesting adaptations of the scale, if needed. The specific objectives were (i) to explore and analyze individual functioning of items including analyses of differential item functioning based on personal factors as well as local dependency of the CPQ-G11-14 (ii) to determine construct validity, unidimensionality and targeting of the underlying scale and (iii) to adapt the scale of the CPQ-G11-14, if needed, based on the results of these analyses in order to establish maximum accuracy and precision.
Methods: A secondary, psychometric analyses of two population-representative datasets (total N=2658) and a randomly selected, combined smaller dataset (n=300) were carried out.
Results: Analyses showed good overall fit to the Rasch model. A number of items (ranging from 51.4% to 94.3%) had so-called disordered thresholds, meaning that the response options between the items overlapped: five initial answer options were collapsed into four, three or two, whichever pattern fitted better. Unidimensionality was achieved by splitting the CPQ-G11-14 into two subscales (socio-emotional versus oral symptoms-functioning). However, the targeting of CPQ-G11-14 is not optimal and in the future, more items that differentiate between children with little oral health problems at the “better” end of the scale should be included based on the opinion of experts.
Conclusions: Psychometric techniques as Rasch model are crucial to construct a valid and reliable instrument, particularly in an area where it is lacking such as oral health-related quality of life instruments.
The submitter must disclose the names of the organizations with which any author have a relationship, the nature of the relationship, and the clinical or research area involved. The following is submitted: NONE