The Influence of Autonomy Support on the Trajectory of Boredom Over a Semester

• In Event: Exploring the Antecedents of Students’ Academic Emotions

Abstract

Boredom is a commonly experienced and debilitating emotion (e.g., Goetz et al., 2006). Most studies examining boredom and its precursors have concentrated on measuring boredom in a single event. Although Ahmed et al.’s (2013) finding of an increase in trait academic boredom among Grade 7 students over a school year advanced our understanding, little is known about the trajectory of state boredom among university students, and the extent to which it relates to antecedents. In Pekrun’s (2006) control-value framework, a lack of support for students’ autonomy is expected to influence cognitive appraisals, which in turn influence the experience of boredom. Autonomy support is an important element of the college learning environment and appears to reduce the experience of negative emotions, including boredom (e.g., Kaplan & Assor, 2012). The purposes of this study were to examine the pattern of change in class-related boredom over an academic semester and to evaluate the influence of perceived autonomy support on this pattern of change.
161 Canadian university students completed five online questionnaires in exchange for research credits. At Time 1 students’ perceived autonomy support from their teachers was measured as a trait dimension. Times 2 to 5 assessed state perceived autonomy support and class-related boredom. Surveys were administered approximately three weeks apart. Perceived autonomy support was reverse-coded to match the direction of boredom measures. To examine changes in boredom over a semester, latent growth curve modelling (LGCM) was used (MPlus 7.0: Muthén, & Muthén, 2012). To examine the impact of perceived autonomy support, a conditional LGCM with time-invariant (i.e., general perception of autonomy support) and time-varying predictors (i.e., triweekly rating of perceived autonomy support during a course) was tested.
Class-related boredom showed a linear growth with good model fit, Chi2(5) = 11.772, p = .038, Chi2/df = 2.354 CFI = .967, RMSEA = .097. The mean intercept (b0 = 27.95, p < .001) and slope (b1 = .71, p = .032) were significant, indicating that the initial level of class-related boredom was significant and the rate of change followed a linear increase trajectory. The initial level of class-related boredom did not influence the linear increase. While there were significant individual differences in the initial levels of class-related boredom, there was no statistical difference in the change in levels of boredom. The conditional model showed an excellent model fit, Chi2(19) = 23.343, p = .223, Chi2/df = 1.229, CFI = .980, RMSEA = .043. The prediction (b1 = .40, p = .01) from general perception on the initial level of class-related boredom was significant, but the time-invariant predictor did not significantly influence the rate of change of class-related boredom. This means that the more students perceived being supported as an autonomous learner at a given time, the lower the level of class-related boredom. Our findings highlight the importance of understanding the antecedents of boredom because supporting autonomy may mitigate otherwise increasing levels of boredom. Reeve (1998) has shown that autonomy-support can be taught to teachers making this an area for future research on boredom intervention.

Authors

• Virginia Man Chung Tze, University of Alberta

• Lia Marie Daniels, University of Alberta

• Rob Klassen, University of York