Ali Mohammad Kyriazos Theodoros*, Galanakis Michalis, Stalikas Anastasios
The purpose of this study was to examine the construct validity of the KPSS (or KPS) in a sample of 621 Greek parents of children 7-13 years. EFA was carried out to test the KPSS factor structure. Considering the restrictions inherent to KPSS (k ≤ 3), a CFA was not possible. The KPSS construct validity was cross-validated indirectly with a compound network model of parenting satisfaction–dysfunctions with KPSS and PCQ. We expected a network with three dimensions, one corresponding to parenting satisfaction (KPSS) and two tapping parenting dysfunctions (PCQ). This hypothesis was successfully confirmed with EGA-an exploratory method of clustering dimensions within the network psychometrics framework-and then with an identical compound CFA model of parenting satisfaction–dysfunctions. The KPSS items shaped a distinct cluster (latent variable), negatively correlated with parenting dysfunctions clusters (latent variables), suggesting construct validity. Full strict measurement invariance was successfully established for the compound CFA model of parenting satisfaction-Dysfunctions across child’s gender. Internal consistency and split-half reliability were significant. Convergent and Discriminant Validity of KPSS was examined with six parenting scales, including APQ-9 Short and a newly developed questionnaire measuring Positive Psychology Parenting (Nicomachus-Positive Parenting or NPP). A consistent pattern of relationships emerged. Normative data were calculated to help parenting programs using KPSS as an outcome. Significant differences in parenting satisfaction scores were detected but with a small effect size.
