Factorial structure of the inCLASS: validation study in Quebec early childcare centers

As it is widely recognized children’s early childhood experiences predict their educational success, tools such as the Individualized Classroom Assessment Scoring System (inCLASS) are being used in educational settings to measure interaction quality at the child level. However, research regarding the four-domain structure of the inCLASS (including adult interactions, peer interactions, task orientation and conflict interactions) appears limited and inconsistent. This study aims to confirm the factorial structure of the inCLASS in the context of Early Childcare Centers (ECC) in Quebec (Canada). A sample consisting of 151 children (M = 61.52 months of age) attending 41 ECC in the Montreal area participated in the study. Confirmatory factor analyses were conducted to confirm whether the factorial structure of the inCLASS applied to data collected in this educational setting. Initial analyses showed an inadequate fit of the model, due to the adult communication dimension. Positive correlations were also found between peer interactions and task orientation. After allowing for the covariance of dimensions belonging to these two domains and setting the residual variance of adult communication to zero, the fit of the model improved, but was still not satisfying. These results are discussed in terms of their impact on the validity of the inCLASS scores obtained outside the United States. Findings suggest the factor structure of the tool could be improved and raise the importance of interpreting with caution the results from research conducted in Quebec using the inCLASS.