Why are we 'weighting' An assessment of a self-weighting approach to measuring oral health-related quality of life

Abstract

Objective: To determine whether or not self-weighting at an item level contributes to the performance of an oral health-related quality-of-life measure. Design: Data were collected in two national surveys conducted a month apart, one using the 'weighted' measure and the other an 'unweighted' version of the UK oral health-related quality-of-life measure. In addition, sociodemographic and self-reported oral health status were recorded. Results: The UK oral health-related quality-of-life measure discriminated between groups based on age group (<65, 65 and older) and social class (higher and lower) irrespective of the version of the questionnaire used. Both versions also showed significant associations with self-reported oral health: denture status (P < 0.01) and number of teeth possessed (P < 0.01). In addition, both versions demonstrated predictive ability in identifying those in prosthetic need (<20 teeth and without recourse to a denture, P < 0.01). Conclusion: Weighting the UK oral health-related quality-of-life instrument does not improve the psychometric properties of the instrument and thus raises questions about the value of self-weighting at an item level. © Blackwell Munksgaard, 2004.