Eric.ed.gov – Measuring Student Progress and Teachers’ Assessment of Student Knowledge in a Competency-Based Education System. REL 2017-238

Competency-based education has received growing attention in recent years as a way to address preK-12 learning goals. In competency-based education, students are promoted to the next course of study or grade level in each subject area after demonstrating mastery of identified learning targets aligned to standards. Westminster Public Schools in Colorado began the transition to a competency-based education system in 2009. In the Westminster Public Schools system, courses of study are organized according to performance level rather than according to traditional grade levels. Westminster Public Schools, a member of the Regional Educational Laboratory Central College and Career Readiness Research Alliance, asked for assistance in examining how long it takes students in the district to progress through their performance levels, especially students who are below their traditional grade level. Westminster Public Schools also asked for assistance in examining how well teachers’ ratings of student competency (learning target scores) align to external assessments of student academic achievement. Educators may use the approach described in this report to assess the degree to which teachers’ judgments of student competency relate to student academic achievement measures in their own school or school district. Using data from the Westminster Public Schools learning management system, this study examined how long elementary and middle school students took to complete math and literacy performance levels 3-8 during the 2013/14 school year. To examine the relationship between students’ learning target scores and Colorado’s standardized achievement test scores, a student’s learning target scores within a performance level were combined to create an overall performance-level competency score for each student. The performance-level competency scores represent an aggregate measure of student competency within their given performance levels based on teachers’ judgments. These performance-level competency scores were then used to predict students’ scores and proficiency levels on the Transitional Colorado Assessment Program. A majority of students completed their courses of study in math and literacy in approximately one academic year. Although a majority of students who were in a math or literacy performance level below their traditional grade level also completed their course of study in one academic year, a larger percentage of them (43-47 percent) completed their level in three or fewer quarters compared with students in a performance level at their traditional grade level (17-22 percent). These results suggest that competency-based education in Westminster Public Schools provides students who are behind academically an opportunity to complete performance levels in less time than in a traditional education system. Students’ performance-level competency scores had statistically significant and positive relationships with Transitional Colorado Assessment Program scores, but the relationships were weak. The performance-level competency scores accounted for only a small proportion (3-4 percent) of the variance in students’ scores on the state achievement test. Math performance-level competency scores accurately predicted math proficiency levels on the state achievement test for 40 percent of students, and literacy performance-level competency scores accurately predicted reading proficiency levels on the state achievement test for 59 percent of students. The performance-level competency scores of students who were in a performance level below their traditional grade level were more likely to predict that their state achievement test proficiency level would be higher than it actually was. In contrast, for students above grade level, performance-level competency scores were more likely to predict that their state achievement test proficiency level would be lower than their actual level. The relatively weak relationships between performance-level competency and state achievement test scores suggest that teachers’ judgments of student competency under competency-based education in Westminster Public Schools are not good predictors of academic performance, as measured by the Transitional Colorado Assessment Program. The following are appended: (1) Westminster Public Schools learning target competency scale; (2) Data and methodology; and (3) Additional study findings.