The aim of this paper was to review the latent structure of the Revised quality of parent-child interaction questionnaire, child version (PACHIQ-R, Lange, Evers, Jansen & Dolan, 2002) on a Croatian student sample. The study included 388 students and parallel analysis, confirmatory analysis and structural modelling procedures were implemented to determine the latent structure of the questionnaire, as an earlier administration on a varied age sample did not confirm the hypothesized theoretical model of two factors. The confirmatory analysis results do not support the two factor structure for both versions of the questionnaire, while parallel analysis indicates multiple factors. Exploratory and subsequent confirmatory analysis of the model supports the existence of one general factor with the greatest loading and three factors with a smaller number of significant loadings. The one factor model without a specific number of items differs significantly from the one factor and two factor model with all items and has better fit parameters. The reliability of the questionnaire is high; omega-h of the model on the questionnaire assessing the relationship with the mother is 0.87, with a total omega of 0.93, while the version with the father yields omega-h of the model is 0.86, with a total omega of 0.95. The contribution of this paper is evident in the finding indicating one latent dimension of the relationships toward the father and the mother, and not as hypothesized by the authors, two dimensions, accepting and resolving conflicts. In determining latent dimensions, caution is advised in the choice and interpretation of factor analysis and conducting a follow-up study with a revised number of items while using one dimension.

child-parent interaction, PACHIQ-R, factor analysis, parallel analysis