Files
Abstract
This study examined whether children's social standing, as defined by peer reputation (i.e., popularity), affective-regard among peers (i.e., social acceptance and social rejection), and network position (i.e., degree, betweenness, and closeness centrality), predicted their accuracy in detecting affiliation patterns within the classroom-based social network. It also examined demographic factors including gender and grade. Fourth- and fifth-grade participants (n = 387) were recruited from 21 classrooms. The current study uses a hybrid SCM/bbnam approach (E. Lee et al., 2022) to calculate children's network accuracy based on the collective perception of the affiliative network gathered from peer-report. Results suggest considerable variability in children's accuracy in detecting classroom-based, peer-reported affiliation networks. Correlations between predictor variables revealed significant associations between network accuracy, gender, grade, betweenness centrality, and closeness centrality. When analyzed with mixed effects regression models, gender, popularity, and closeness centrality were unique predictors of overall accuracy within the multivariate context. Analysis of error rates revealed individuals closely positioned to all others within the network (i.e., closeness centrality) demonstrate better accuracy by making fewer errors in reporting non-existent relationships (i.e., fewer false negatives). In addition to closeness centrality, popular children and girls consistently demonstrate a more accurate perception than their counterparts. Current findings suggest predictors of network accuracy might vary depending on the network size and warrant examination of how structure might moderate the relationship between social standing and accuracy.