The Role Of Within-Class Consensus On Classroom Climate Constructs In The Context Of Multi-Level (Structural Equation) Modelling
27th January 2020 : 12:45 - 14:00
Research Group: Quantitative Methods Hub
Speaker: Lisa Bardach, University of York
Location: Department of Education, Seminar Room B
Convener: Lars-Erik Malmberg
Classroom characteristics, such as motivational climate, are by their very nature classroom level constructs, as students provide ratings on features of the classroom that are common to all students (‘climate constructs’). Hence, statistical methods that allow for properly disentangling effects occurring on the individual and the classroom level –i.e., multilevel (structural equation) models– have become a widely used approach.
To date, numerous multilevel studies have linked class mean levels of student ratings on climate constructs to a range of educational outcomes, such as motivation and achievement. Compared to the immense body of research focusing on the mean levels of student ratings in the context of multilevel modelling, only very little attention has been paid to a further feature inherent in student ratings of classroom climate constructs: the degree to which students within a given class agree in their perceptions of the climate (within-class consensus). From a methodological perspective, the aggregation of individual student ratings to evaluate the climate of a class requires establishing sufficient consensus among the students within this class. In addition to methodological considerations concerning within-class consensus, researchers have started treating within-class consensus as a construct with a substantive meaning, and thus, to complement their analyses of mean levels of classroom climate constructs with analyses of consensus.
In this talk Lisa Bardach introduces her research program on within-class consensus and give an overview of four empirical studies. In her research, she assesses within-class consensus within the influential framework of goal structures. Goal structures constitute the motivational climate in a class that is manifested in teachers’ instructional strategies, e.g., concerning the design of task, autonomy-supportive behaviour and their evaluation practices. Relying on multilevel modelling and multilevel structural equation modelling, she shows that within-class consensus on several dimensions of goal structures predicts secondary school students’ achievement and motivation in the sense of achievement goals (study 1), and socio-emotional variables, such as a positive error climate and students’ engagement in cooperative learning strategies (study 2) –above and beyond any direct effects of the mean levels of goal structures. In study 3 she revisits the relation between consensus on goal structures and secondary school students’ achievement to test their temporal ordering. Results from cross-lagged multilevel structural equation models reveal that for two of the three investigated dimensions of goal structures, achievement precedes consensus and not vice versa. Including additional teacher data in multi-level structural equation models, study 4 demonstrates that consensus on goal structures in secondary school classes is associated with specific teacher behaviour, such as differential teacher treatment of higher vs. lower achieving students, and that higher levels of negative emotional teacher factors (teachers’ emotional exhaustion, teaching-related anxiety) decrease consensus. Finally, she discusses promising future directions for research on within-class consensus building on and expanding the presented work.