Patterns of Control-Value Beliefs as Predictors of Emotions: A Latent Class Analysis

• In Event: New Empirical Insights on What Energizes Learners: A Session on Emotions and Engagement

Abstract

Background. According to Pekrun’s control-value theory of achievement emotions (Pekrun, 2006), control and value judgments are key in predicting the nature and intensity of various discrete emotions learners experience in achievement situations. Indeed, there is cumulative correlative evidence that control and value beliefs are linked with students’ achievement emotions. Specifically, control beliefs are positively related with positive emotions and negatively related with negative emotions, whereas value relates positively to both positive and negative emotions except boredom (for a summary of supporting evidence, see Pekrun & Perry, 2014). Less research attention has been paid to another key assumption of the theory: Different discrete emotions are thought to be characterized by specific combinations of these beliefs. For example, enjoyment should occur when someone holds high control beliefs coupled with high value, whereas anxiety should be characterized by low control beliefs coupled with high value. The present research explores through latent class analyses whether and how frequently such different patterns of control and value beliefs really exist among students, and if so, how those relate to achievement emotions.

Methods. We used data from N = 1,404 5th grade students (49.8% female) from all three German secondary school tracks who self-reported their control and value beliefs towards mathematics (6 items for control, e.g. “I can solve mathematics problems easily”; 5 items for achievement value, e.g. “It’s very important for me to do well in mathematics”). Students’ achievement emotions were assessed using the AEQ-Mathematics (Pekrun et al., 2011). Herein, we report results for enjoyment, anxiety, anger, and boredom. We used Mplus Version 8 (Muthén & Muthén, 1998-2017) to explore the number of latent classes underlying the participants’ responses to the control and value items, and to investigate how those patterns were related to the achievement emotions.

Results and Discussion. Results indicated that a 5-class solution showed the best fit to the data (entropy=.831). Three of the groups were characterized by more or less balanced levels of both control and value beliefs (low/low, n=228; medium/medium, n=505; high/high, n=290); and two groups emerged with opposing levels of control and value beliefs (high control/low value, n=206; low control /high value, n=175). In line with predictions of control-value theory, the high/high group reported significantly more enjoyment than all other groups, and the low control/high value group reported significantly more anxiety than all other groups. For anger and boredom, results were less clear-cut, and group differences were less pronounced. Both the low control/high value and the low control/low value groups reported relatively high levels of anger, thus feeling out of control seems to be a key driver of achievement-related anger. For boredom, in line with expectations, low value seems to be the driver: Both the low control/low value and the high control/low value groups reported high boredom. However, surprisingly, the low control/high value group also reported relatively high levels of boredom. Overall, our analyses provide unique insight to the existence and frequency of belief patterns and their links with emotions as proposed control-value theory.

Authors

• Anne C. Frenzel, University of Munich (LMU)

• Reinhard Pekrun, University of Essex

• Julia Maria Gertrud Hoelzl, University of Munich