Several studies (e.g., Vermunt, 2005; Guàrdia et al., 2006; Rienties et al., 2012) have been devoted to the conceptual definition of students' performance at all stage of their education, and to the identification of the factors that influence it, with the goal of promoting the development of effective strategies to increase scholastic attainments. Aim of the paper is to evaluate the presence of any effect of interpersonal relationships on performance under the assumption that students adjust their own attitudes and behavior to that of others they are connected with (i.e., social influence or contagion). This aim requires to include network's effect within a structural equation model, in which the academic performance is dealt as an endogenous latent construct considering objectives indicators (e.g., mean exam scores or final grade) as well as students' perception of learning experiences. For this reason, a simple and intuitive two steps procedure is used: first, factor scores are derived for the latent constructs, and then an individual outcome regression model with proper contiguity matrices (i.e., network autocorrelation model, Doreian et al., 1984) is estimated. The effect of students' communities on performance is then studied by means of data collected through questionnaires filled by a sample of student enrolled at an Italian University in the academic year 2008-09.
Academic performance and students' communities: including contiguity effect in Structural Equation Models
VITALE, MARIA PROSPERINA;
2013-01-01
Abstract
Several studies (e.g., Vermunt, 2005; Guàrdia et al., 2006; Rienties et al., 2012) have been devoted to the conceptual definition of students' performance at all stage of their education, and to the identification of the factors that influence it, with the goal of promoting the development of effective strategies to increase scholastic attainments. Aim of the paper is to evaluate the presence of any effect of interpersonal relationships on performance under the assumption that students adjust their own attitudes and behavior to that of others they are connected with (i.e., social influence or contagion). This aim requires to include network's effect within a structural equation model, in which the academic performance is dealt as an endogenous latent construct considering objectives indicators (e.g., mean exam scores or final grade) as well as students' perception of learning experiences. For this reason, a simple and intuitive two steps procedure is used: first, factor scores are derived for the latent constructs, and then an individual outcome regression model with proper contiguity matrices (i.e., network autocorrelation model, Doreian et al., 1984) is estimated. The effect of students' communities on performance is then studied by means of data collected through questionnaires filled by a sample of student enrolled at an Italian University in the academic year 2008-09.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
