F. Veiga1, V. Robu2, J. Appleton3, I. Festas4, D. Galvão1

1Universidade de Lisboa (PORTUGAL)
2University of Bucarest (ROMANIA)
3University of Georgia (UNITED STATES)
4Universidade de Coimbra (PORTUGAL)
Conceptual Framework:
The value and current relevance of the construct students’ engagement in school (SES) have been highlighted in literature, despite of the lack of empirical studies and validated multidimensional instruments.

The purpose of this study is to study how the relationship between SES and the student’s concept of self (self-concept), varies throughout the adolescence school years.

The sample consisted of 685 students from different regions of the country, of both sexes, divided by grade (6th, 7th, 9th and 10th).
Data were collected in classroom context through a survey that included items from “Piers-Harris Children’s Self-Concept Scale” (PHCSCS) and the questionnaire “Student’s Engagement in School - A Four Dimensional Scale (SES-4DS)” which includes the cognitive, affective, behavioral and agentic dimensions (Veiga, 2013), and shows high psychometric qualities.

Results from variance analysis of engagement (anova two-way 2x2), according to school year (6th and 7th versus 9th and 10th grades) and self-concept (low and high), allowed to find a main effect of the school year on the cognitive dimension of SES and total scale (p <0.001); the effect of self-concept (ac) manifested itself in all dimensions of SES, with a high level of significance (p <0.001); the significant effects of the interaction of the variables year and self-concept emerged in cognitive and agenctic dimensions, as well as in the total scale, and were due to a greater differentiation in the 6th and 7th grades, comparing with the 9th and 10th grades, and also a greater decrease, over the years, of such dimensions in the higher self-concept group than in in the lower self-concept group. The study of this same variable in the modality of anova two-way 2x3 (low, medium and high self-concept) confirmed the main effects but not the variables interaction.

Considering the lack of studies on these concepts, results are framed within the context of social-cognitive perspective of adolescence development, emphasizing the importance of the activation of variables such as self-concept.