Implementation and Outcomes in Personalized Learning Schools

• In Event: Lessons Learned From the Design and Implementation of Personalized Learning in K–12 Schools

Abstract

Background and Methods

This paper examines implementation of a diverse set of personalized learning (PL) models in 40 schools that are funded by the Next Generation Learning Challenges initiative. It also presents results from an analysis of student achievement growth in 35 of those schools. We address the following research questions.

3. What features of PL did schools implement, and what practices did they adopt to promote PL?
4. What challenges and enablers did school staff encounter as they implemented PL?
5. How do students in PL schools perform on measures of academic achievement, relative to similar students in more traditional schools?

We draw on document reviews, administrator interviews, teacher surveys and logs, student surveys, and site visits to nine schools (including principal and teacher interviews and student focus groups). We examine achievement growth using data from the spring and fall administrations of the NWEA Measures of Academic Progress (MAP) in PL schools and data from a “virtual comparison group” (VCG) of students who have similar levels of initial achievement and who attend schools that are similar to the PL schools in terms of poverty and locale.

Findings

Most PL schools used learner profiles to guide instruction, provided personalized learning paths, housed course content primarily online, and used adaptive course content and assessments. Schools varied in implementation of other features, including grouping of students by learning level rather than grades, using competency-based progression, and enabling students to access instructional content outside of school hours. Student survey results largely corroborated reports from teachers and administrators; students indicated extensive exposure to instructional practices related to personalization. For instance, large majorities reported awareness of instructional goals, had opportunities for choice and for variation in pacing, and had access to technology for tracking their own progress. Common challenges reported by teachers included a lack of useful assessment data to determine competency, difficulty integrating multiple data sources to support the development of individualized student learning plans, lack of training in implementing competency-based learning, classroom management problems, and tension between the desire to have students work at an appropriate developmental level and pressures to ensure students meet grade-level proficiency on state tests.

Earlier analyses of student achievement in a sample that included some of these schools suggested that achievement growth in PL schools outpaced that of comparison schools over the two-year period examined. The achievement analysis for this paper focuses on the 35 PL schools with MAP data for the 2014-15 school year, and prior years when available.

Significance

Earlier research on PL identified implementation challenges and facilitators, and suggested that PL is associated with positive learning outcomes, at least in some settings. This paper provides new evidence from schools that vary in the maturity of PL implementation, including some that have used the models for several years. The findings will contribute to our understanding of what factors help support initial PL implantation and to sustain that implementation over time, and will also contribute to the growing body of evidence regarding learning outcomes in PL schools.

Authors

• John Pane, RAND Corporation

• Elizabeth Steiner, RAND Corporation

• Laura S. Hamilton, RAND Corporation