Innovative Schools

Ongoing challenges caused by the COVID-19 pandemic have widened learning gaps. Recent 2022 NAEP results indicated that two decades of growth for American fourth-grade students in reading and math were wiped away by just two years of pandemic-disrupted learning. And while elementary and middle school students have made up significant ground to address learning loss, they are still not fully caught up. According to an Overdeck Family Foundation-supported analysis of standardized test score data led by Stanford and Harvard researchers, students have made up about a third of what they lost in math and a quarter of what they lost in reading. Meanwhile, the gap in student proficiency between high- and low-income communities continues to widen.

Pre-pandemic, a classroom likely contained students who performed at seven different grade levels; post-pandemic, the range was predicted to span up to nine grade levels. In a June 2020 survey conducted by YouthTruth, teachers in schools with at least 75 percent of students who qualify for free and reduced-price lunch were most likely to say their students work on different content at different paces and need additional support.

Unlike the traditional classroom model, which is often a teacher lecturing at the front of a classroom, a student-centered learning model shifts the focus onto students and how to best meet their needs through coaching, small group facilitation, and the use of educational technology. Studies show students attending student-centered learning environments outperform their peers, especially Black, Latino, economically disadvantaged, and English learner students.

Learning is most effective when content is meaningful and relevant, when adults build caring relationships with their students, and when students are in classrooms that foster a sense of belonging and a growth mindset (Duke et al., 2017; Dweck, 2006; The Science of Learning; Quay, 2017). Overdeck Family Foundation-supported studies by researchers in the Mindset Scholars Network showed that ninth grade students with stronger reports of growth mindset earned higher core course GPAs than students reporting a more fixed mindset. A second study found that math teacher messaging that supported growth mindset, belonging, purpose, and affirmation explained roughly 29 percent of the variance in student test score growth, exceeding the predictive power of established classroom practice inventories.

Research consistently shows the potential of high-dosage tutoring at improving student outcomes, but now we know that virtual and hybrid tutoring models can provide cost-effective alternatives to traditional tutoring, while continuing to show strong results for student impact. Preliminary data from an evaluation of Saga Education’s blended learning model indicate that participation improves student math test scores by the equivalent of an extra one to two years of additional math learning for the typical American high school student, in a single year. And findings of OnYourMark Education’s high-impact virtual literacy tutoring program revealed a positive impact on students’ early literacy skills, including effects on kindergarteners’ knowledge of letter sounds and first graders’ decoding skills.

Educators overwhelmingly see technology as a key resource to solve the above problems and meet students’ expanding needs. On average, schools used 2,591 edtech tools in SY 2022-23, but little evidence is available to help them decide which products are best for their students. Moreover, teachers are stressed, overwhelmed by technology, and reluctant to invest their limited time in a potentially unproven tool or approach. Only 15 percent of K-12 education companies said their customers had asked for results of a correlational study and, according to CEMD, only 36 percent of districts analyzed by CEMD selected exclusively high-quality math curricula for elementary school. Many other districts select curricula that are not fully aligned with quality criteria or are state-specific and of indeterminate quality, which increases the risk that ineffective edtech solutions are implemented, continuing to increase learning loss versus reverse it.