Abstract
The study aims to validate the shortened version of the Person's Relating to Others Questionnaire (PROQ3), a measure of negative and maladaptive relating to others, for data collection via the Internet across 4 national samples. The psychometric properties of the Internet-administered (IN) format of the PROQ3 in 4,802 participants (169 British; 360 Irish; 1m110 Dutch; and 3,163 Greek) were compared with that of the standard-written (SW) version in 1918 participants (338 British, 403 Irish, 204 Dutch, and 973 Greek), in respect of its measurement and structural equivalence. Internal consistency, as estimated by alpha coefficient and item-to-scale homogeneity, were consistently acceptable across nationality and modality. There was agreement in regard to the interscale correlations across nationality and modality. Lower mean scores for the British sample in the SW format, and lower mean scores for the Irish sample in the IN format were found. The structural equivalence across modality and nationality was also supported: A consistent 8-factor underlying structure, as supported by a multiple group factor analysis, and an octagonal higher order, as supported by a 3-way multidimensional scaling procedure, were found. It was concluded that the PROQ3 can be administered via the Internet with maintained psychometric properties for clinicians to screen people with interpersonal relating deficiencies and for researchers to collect data.
