A meta-analysis on the impact of grades and comments on academic motivation and achievement: a case for written feedback

Abstract

This research synthesis examined the impact of grades, comments, and no performance feedback on academic motivation and achievement in elementary and secondary school. Four meta-analyses were conducted, with two each exploring the impact of (a) grades versus no performance feedback and (b) grades versus comments on academic motivation and achievement, respectively. Overall results indicated that grades positively influenced achievement but negatively influenced motivation compared to no feedback. However, compared to those who received comments, students receiving grades had poorer achievement and less optimal motivation. Moderator analyses generally suggested that overall effects varied as a function of the type of motivation (i.e. the specific construct, internal vs. external motivation), context (e.g. academic subject; comment type), student characteristics (e.g. achievement level), and methodology (i.e. grade anticipation versus receipt), though it was not possible to test these moderators in all analyses. Theoretical and methodological contributions and implications for education practice are discussed.

Keywords:

• Grades
• written feedback
• assessment
• motivation
• achievement

Acknowledgements

An earlier version of this paper was presented at the American Educational Research Association (AERA) Annual Meeting.

Disclosure statement

Alison C. Koenka is now at Virginia Commonwealth University, and Lisa Linnenbrink-Garcia is now at Michigan State University.

Notes

1 Eccles and Wigfield (2002) group motivation constructs into four, rather than three categories. However, the fourth category of constructs – those that integrate motivation and cognition – is beyond the scope of this synthesis.

2 We intended to explore how these factors moderated effects in all meta-analyses (grades versus no feedback and grades versus comments). However, there was insufficient variation to do so in the syntheses comparing grades and comments. As a result, we focus our discussion of hypotheses related to these variables on the syntheses comparing grades and no feedback.

3 Effect sizes for motivation constructs that were hypothesized to be influenced by performance feedback in opposing directions (i.e. internal and external motivation processes) were both included when computing the overall effect size. As a result, we do not substantively interpret this overall effect size.

4 All of the aforementioned motivation construct effect sizes reached statistical significance. However, it is important to note that the magnitude of some effect sizes were quite small according to Hattie’s (2009) standards. The same is true of the specific motivation construct moderator analyses for the Comments vs. Grades meta-analysis reported on p. 18. Please see Supplemental File C for more information.

5 Effect sizes for motivation constructs that were hypothesized to be influenced by performance feedback in opposing directions (i.e. internal and external motivation processes) were both included when computing the overall effect size. As a result, we do not substantively interpret this overall effect size.

6 As we noted earlier, previous theory and research suggest that the motivational consequences of grades for students who are in secondary school and/or lower-achieving may depend on the broader evaluative context such as whether grades are normative or criterion-referenced (e.g. Ames, 1992). This may help to explain the existing pattern, but it is beyond the capabilities of the current synthesis to speak to this possibility. Educators should thus consider this caveat along with the final recommendation.

Additional information

Funding

This research was supported by a grant from the W.T. Grant Foundation [#181179] to Harris Cooper, Principal Investigator.