Presentation Title

Assessing Reliability and Validity of Peer and Self-Graded Assessments

Faculty Mentor(s)

Dr. Ruth Reynolds

Project Type

EdD Colloquium - ONU

Scholarship Domain(s)

Scholarship of Discovery

Presentation Type

Presentation

Abstract

The purpose of this study was to determine the usefulness of peer and self-grades as proxies for the teacher’s grades by evaluating the reliability and validity of evidence-based grading practices in a high school setting. The results of this study were intended to reveal grading strategies, if any, that were useful for teachers in order to bridge the gap between theoretical models of learning and explicit pedagogical techniques. This quantitative correlational study involved one participating teacher from a public suburban high school who administered a written assessment to 77 student participants in an Advanced Placement science course. The students self-graded and peer-graded the assessments using a publicly-released grading rubric from College Board. The researcher employed descriptive and inferential statistics to probe the relationships between peer-grades, self-grades, and teacher grades. The teacher’s grades were highly, positively, and statistically significantly correlated with averaged peer-grades (r(75) = 0.75, p < .001) and self-grades (r(74 )= .66, p < .001). Peer-graders demonstrated good inter-rater reliability (Cronbach’s α = 0.85). Peer-grading accuracy did not statistically significantly improve as students continued to peer-grade. No statistically significant difference in self-grading accuracy was detected when students peer-graded before self-grading. Peer-grades were more acceptable measures of student performance compared with self-grades and can be used as substitutes for teachers’ grades in the limited methodological and demographical context of this study. A larger and more diverse sample of student and teacher participants is recommended for future studies.

Cohort XX

Permission type

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

This document is currently not available here.

Share

COinS
 
Apr 17th, 11:40 AM

Assessing Reliability and Validity of Peer and Self-Graded Assessments

Wisner Auditorium

The purpose of this study was to determine the usefulness of peer and self-grades as proxies for the teacher’s grades by evaluating the reliability and validity of evidence-based grading practices in a high school setting. The results of this study were intended to reveal grading strategies, if any, that were useful for teachers in order to bridge the gap between theoretical models of learning and explicit pedagogical techniques. This quantitative correlational study involved one participating teacher from a public suburban high school who administered a written assessment to 77 student participants in an Advanced Placement science course. The students self-graded and peer-graded the assessments using a publicly-released grading rubric from College Board. The researcher employed descriptive and inferential statistics to probe the relationships between peer-grades, self-grades, and teacher grades. The teacher’s grades were highly, positively, and statistically significantly correlated with averaged peer-grades (r(75) = 0.75, p < .001) and self-grades (r(74 )= .66, p < .001). Peer-graders demonstrated good inter-rater reliability (Cronbach’s α = 0.85). Peer-grading accuracy did not statistically significantly improve as students continued to peer-grade. No statistically significant difference in self-grading accuracy was detected when students peer-graded before self-grading. Peer-grades were more acceptable measures of student performance compared with self-grades and can be used as substitutes for teachers’ grades in the limited methodological and demographical context of this study. A larger and more diverse sample of student and teacher participants is recommended for future studies.

Cohort XX