Discrete Emotions in Summative Feedback: Differentiated Effects of Anger, Enjoyment, and Boredom When Assigning Grades

• In Event: Channels of Feedback: Examining Sources, Types, and Contingencies

Abstract

Objectives and Theoretical Framework
Grading can be considered a form of summative feedback (Dixson & Worrell, 2016). Since grading takes up a considerable portion (approximately 13%; OECD, 2014) of teachers' professional lives and is perceived as an emotional task (Steinberg, 2008), it is important to shed light on the processes involved in grading. Previous research showed that externally induced emotions influence grading in emotion congruent ways. Negative emotions lead to lower and positive emotions lead to higher grades for the same essays (Brackett, Floman, Ashton-James, Cherkasskiy, & Salovey, 2013). This picture misses if naturally occurring cues can elicit teacher emotions, which in turn influence grades. Therefore, we sought to show that a material-inherent cue (handwriting) can naturally elicit discrete emotions (anger, enjoyment, and boredom), which influence grades differently.
Method and Data Sources
Our experiment involved 74 pre-service teachers (62 female; Mage = 25.31(SD= 4.30)) from two universities in southern Germany grading two different essays of similar quality written in different handwriting (good vs. bad) one after the other. After reading an essay, they directly assigned a grade and self-reported their emotions when grading the respective essay. Participants experienced both levels of each variable (e.g., essay A in good handwriting was combined with essay B in bad handwriting). The discrete emotions of anger (e.g., "I was angry while grading"), enjoyment (e.g., "I had fun grading the text"), and boredom (e.g., "I was so bored during grading, that time seemed to stand still") were measured on a 4-point Likert scale (1 = strongly disagree, 4 = strongly agree) with three items each. Grades were assigned according to the German grading system on a scale from 1 (very good) to 6 (insufficient) and were then reverse-coded. All variables were transformed to represent the participants' emotions and the assigned grade in the good versus the bad handwriting condition – irrespective of the order and the essay in which the good and bad handwriting were presented in the study. Two-condition within-participant mediation analyses (Montoya & Hayes, 2017) were used to estimate the unstandardized path coefficients, representing the mean difference within each participant by comparing the good and the bad handwriting condition.
Results
Figures 1 and 2 revealed that the unstandardized effect of handwriting on grades was 0.32 (CI [0.03, 0.61]), which equals one-third of a letter grade. The relationship was mediated via anger (0.44, CI [0.21, 0.67]) and enjoyment (0.26, CI [0.06, 0.43]) in emotion congruent ways, but not via boredom (0.02, CI [- 0.04, 0.10]). That is, good handwriting predicted a lower level of anger and a higher level of enjoyment, whereas it did not predict boredom. A higher level of anger and enjoyment, in turn, predicted lower and higher grades, respectively, whereas boredom did not predict grades.
Significance
Results imply that it is important to investigate grading as an emotional task to prevent negative consequences for teachers' well-being on the one hand and students' future careers through biased grades on the other.

Authors

• Carolin Schwab, University of Munich
  Presenting Author

• Anne C. Frenzel, University of Munich
  Non-Presenting Author