Comparing Oral and Traditional Assessments in Math Content Courses for Pre-Service Elementary Teachers

Abstract

In two sections of a number and operation course for pre-service elementary teachers, all students (n=42) took three exams, each with a traditional, written component as well as an oral component conducted the same week. Students also completed the Inventory of Learning Styles (ILS), the Abbreviated Math Anxiety Rating Scale (AMARS), and Supplemental AMARS items developed for this study. The study presented in this poster asked two questions: (1) What characteristics of a student leads her to perform better on an oral versus written mathematics assessment (relative to her peers)? (2) What opportunities for student learning can oral assessments provide? To measure relative student performance on oral versus written assessments in Question 1, we looked at the difference in a student’s z-scores on content-aligned oral assessment and written assessments questions. A linear regression analysis found that one point higher in Test Anxiety (a component of Math Anxiety) predicts a written score of 0.28 standard deviations higher than the corresponding oral score. Conversely, one point higher in Problem-Solving Anxiety (a new component of Math Anxiety suggested by a factor analysis of the AMARS and Supplemental items) predicts an oral score of 0.33 standard deviations higher than the corresponding written score. In pursuing Question 2, we conducted a detailed case study of one oral assessment that suggests that students can revise and extend their thinking in the process of taking an oral assessment and responding to instructor questions.