Grade expectations: the motivational consequences of performance feedback on a summative assessment

Abstract

The current study investigated the impact of performance feedback on female students’ motivation following a summative classroom assessment in STEM. Seventh through ninth grade students (N = 161) were randomly assigned by their math or science class to one of the following conditions: grades, written comments, grades and comments, and no feedback. Results revealed complex interactions between feedback stage (anticipation vs. receipt) and form. Students in the grades-and-comments condition experienced the most optimal effects (i.e., increased mastery goal orientation and intrinsic motivation) despite a decline in self-efficacy. Students in the comments condition experienced a similar increase in intrinsic motivation. Finally, students in the no feedback condition reported a heightened performance-approach goal orientation when they received assessments with no feedback. Theoretical, methodological, and practical implications are discussed.

Keywords:

• Assessment
• goal orientations
• intrinsic motivation
• performance feedback
• self-efficacy

Notes

1 Although task-focused comments can vary, an example of a more specific task-focused comment delivered in a middle school math course would be, “The way you calculated the volume of the cube is perfect! When you calculate the volume of cylinder, it is a tricky operation to remember to square and not double the radius as you have done…”; Koenka & Anderman, 2019, p. 18.

2 15 classes took part in the study (n = 187). However, a fidelity error occurred that led to data from two classes (n = 26) being excluded from the analysis. As a result, the remaining sample size and demographic information reported in this section reflects the 13 classes (5 science and 8 math) that were retained for analysis.

3 The remaining 15% of students did not report their race or ethnicity.

4 In addition to reporting their age, year in school, and race/ethnicity, students were also asked to report their average grade in (a) their current math or science course and (b) the relevant subject from the previous year .This achievement information was not used in analyses because many students reported not knowing their current grade and simply guessing their average grade from the previous academic year.

5 The possible levels for this grading system are R, 1, 2, 3, or 4. Similar to a conventional letter grade, the levels ranging from 1 through 4 could be accompanied by a “+” or “–”. This type of quantitative feedback was provided to students in the grades and grades plus comments conditions because it was the quantitative feedback students at the school regularly received.

6 This data collection effort also measured students’ perceptions of the classroom goal structure, perceptions of assessment fairness, and reports of time spent studying. However, information on these measures are not reported given that they are beyond the scope of the article. Results involving these variables are available upon request.

7 Because PALS subscales were administered, this means that performance-approach goal orientations were measured as a desire to demonstrate one’s competence. This diverges from some other measures in the literature, which conceptualize and measure performance-approach goal orientations as a desire to outperform others (e.g., Elliot & Murayama, 2008).

8 This effect should be interpreted with caution given that its p-value of .05 was equal to, rather than less than, the alpha cutoff of .05.

9 It is important to note that the change in performance-approach goal orientation from anticipation to receipt only reached marginal significance for students in the comments condition (p = .05). As a result, this finding should be interpreted with caution.