Given that the fourth and last step in the factorial invariance test is strict invariance, which involves adding another constraint to those used to test strong invariance, and bearing in mind that the previous model with fewer constraints failed to remain invariant, it can be concluded that the model also fails to maintain strict invariance in any of the subsamples analyzed.
In sum, the data extracted in relation to invariance enable us to state that the four-dimensional factorial configuration with liberation of covariances between items, and therefore the factorial structure of the CIAF questionnaire, remains stable regardless of age, sex and physical activity.
This lack of invariance indicates that people distinguish between the four types of sociocultural influence, although the importance attached to each varies during different stages of adolescence.
55 Table 5 Goodness-of-fit indexes of the models of invariance Invariance [chi df p RMSEA NNFI CFI AIC square] Sex Configural 667.
Our first test involved determining the existence of configural invariance, which was established with the indicators shown in Table 2 (Model CCi).
Table 2 shows the results of metric invariance testing (Model [C.
This validation required confirming the factorial invariance hypothesis for the model across both groups.
Despite the differences in adjustment between the Brazilian and French models (Models A1 and B1), the findings related to invariance suggest that the meaningfulness model is partially similar for the two groups.
The most powerful and versatile method for testing such cross-cultural invariance is the multigroup, confirmatory factor analysis (CFA) approach (Steenkamp & Baumgartner, 1998).
Configural invariance addresses the question of whether the same pattern of factor loadings exists across different groups or countries (i.
Prior to conducting the measurement invariance testing, we assessed two potential biases in our data.
Prior to beginning tests of measurement invariance, it was necessary to determine if a unidimensional or multidimensional factor structure best fit the data.