Disentangling Support for Violent and Non-violent Radicalization among Adolescents: A Latent Profile Analysis.

Although support for violent and non-violent radicalization can co-occur, only a few adolescents who support non-violent radicalization also support or engage in violent acts. Yet, little is known about what factors are associated with adolescents' paths towards or away from violent and/or non-violent radicalization. Within a socio-ecological and positive youth development framework, this study investigates profiles of support for violent and non-violent radicalization among adolescents attending high schools in Quebec (Canada) and whether such profiles are differently associated with experiences of social adversity, school-, family- and peer-related factors and psychological distress. Adolescents (N = 1911; Mage = 15.7; SDage = 0.98; 48.7% girls) completed an online survey during school hours. A Latent Profile Analysis on scores of support for violent and non-violent radicalization was conducted. A multinomial logistic regression was used to explore the associations between profiles and variables of interest. We identified six profiles of adolescents. The heterogeneity of profiles suggested multiple and complex combinations of support for violent and non-violent radicalization as well as their co-existence in some but not all profiles. Adolescents who reported less discrimination, more positive school experiences and more family support were less likely to belong to profiles that supported violence. Primary prevention efforts in the field of support for violent radicalization must adopt a socio-ecological and social justice approach and consider the diversity of adolescents' profiles, attitudes and experiences.

Cultural differences in the measurement of self-curiosity within Mexico: a person-centered and variable-centered study.

Background: This study assessed the role of culture in the measurement of the Self-Curiosity Attitude-Interest scale (SCAI-M), a measure of attitude and interest in increasing one's knowledge of self, adopting both a person-centered and variable-centered approach. Methods: The study was conducted on a Mexican sample composed of 484 adult participants who completed both the SCAI-M and a series of instruments that measure cultural dimensions through Qualtrix. Data were collected between November 9, to December 18, 2020, and respondents were contacted using advertisements on social media platforms (Facebook and WhatsApp). Analyses included multigroup confirmatory factor analysis and latent profile analysis. Results: A latent profile analysis allowed for the generation of three groups featuring distinct cultural orientations that were similar to previously found cultural profiles (Consensus-oriented Egalitarians, Flexible Individualists, and Rules-based Competitors). Multigroup Confirmative Factor Analysis showed partial metric and scalar invariance for the SCAI-M between groups; moreover, we found proofs of convergent validity with other cultural dimensions besides the ones linked with the Hofstede model. Our results indicate that cultural profiles and cultural variables are associated with both the level and meaning of self-curiosity among Mexican citizens. Conclusions: Finally, the discussion includes considerations on self-curiosity divergence among minority cultures and relevant clinical applications; a field for which we propose future research.

Latent Congruence Model to Investigate Similarity and Accuracy in Family Members' Perception: The Challenge of Cross-National and Cross-Informant Measurement (Non)Invariance.

Several methods are available to answer questions regarding similarity and accuracy, each of which has specific properties and limitations. This study focuses on the Latent Congruence Model (LCM; Cheung, 2009), because of its capacity to deal with cross-informant measurement invariance issues. Until now, no cross-national applications of LCM are present in the literature, perhaps because of the difficulty to deal with both cross-national and cross-informant measurement issues implied by those models. This study presents a step-by-step procedure to apply LCM to dyadic cross-national research designs controlling for both cross-national and cross-informant measurement invariance. An illustrative example on parent-child support exchanges in Italy and Germany is provided. Findings help to show the different possible scenarios of partial invariance, and a discussion related to how to deal with those scenarios is provided. Future perspectives in the study of parent-child similarity and accuracy in cross-national research will be discussed.