A Harvard University professor who evaluated the CORE districts’ research on the relationship between students’ social and emotional skills and academic achievement reports encouraging initial results. And he says the work should continue, even if the districts’ controversial next step – incorporating the findings from student surveys into an index that ranks schools’ performance – has pitfalls.
The California CORE districts are doing groundbreaking work and serving as a lab for the nation, Martin West, an associate professor in Harvard’s Graduate School of Education, wrote in an article published Thursday by the Brookings Institution, where he is a senior fellow. “What is important is that we learn from what happens next. We need to let evidence speak.”
The California Office to Reform Education, or CORE, is a collaborative of nine districts, six of which received a federal waiver from the No Child Left Behind Act three years ago to design a new school accountability system. They include some of the state’s largest districts – Los Angeles, Long Beach, Fresno, Santa Ana, San Francisco and Oakland – with nearly a million students.
Part of its work involves measuring skills like perseverance, confidence and ability to collaborate – what researchers call social-emotional or non-cognitive skills – and their impact on achievement. A growing body of research, led by teams at Stanford University, has established that students with these skills perform better than students lacking them, and that these mindsets can be taught. Some of the CORE member districts, including San Francisco, Oakland and Sacramento City, initially through pilot schools like Oak Ridge Elementary, have integrated teaching social-emotional skills into their curriculums and classroom activities.
Where CORE is heading into uncharted – and some researchers say risky – territory is attempting to incorporate students’ self-evaluations of their skills into a school accountability system that ranks schools.
As West acknowledged, leading researchers Angela Duckworth, a psychology professor at the University of Pennsylvania, and David Yeager, a psychology professor at the University of Texas, who’s also affiliated with Stanford, have criticized making social-emotional factors part of schoolwide accountability systems. Doing so, they warned, will compound the significant challenges of relying on student self-assessments. Students may inflate – or be coached to inflate – their self-ratings, and they have no objective standard on which to rate themselves, West wrote.
In a New York Times article last month, Duckworth, who resigned from a Boston nonprofit that helped CORE create its social-emotional measures, said she could not support using the measures to evaluate school performance. “I do not think we should be doing this; it is a bad idea,” she said.
While agreeing the potential concerns are significant, West wrote that CORE “has approached the development of this component of its accountability framework in a cautious, thoughtful manner.”
The districts developed their surveys to measure four social-emotional skills that evidence has shown, West wrote, are valid predictors of academic success:
- Self-management – the ability to regulate your emotions and behavior in stressful situations and to set and work toward goals;
- Growth mindset – the belief that you can improve your abilities, such as in math, through effort;
- Self-efficacy – the confidence in your ability to reach a goal;
- Social awareness – the ability to understand and empathize with others, particularly those with different backgrounds and cultures.
West and his colleagues at Harvard’s Center for Education Policy Research concluded that the questionnaires that CORE developed were valid, although less statistically reliable for 3rd and 4th graders. Analyzing individuals’ responses by the middle schools they attended, West said they found strong correlations between students’ average ratings of social-emotional skills and their schools’ overall test scores in math and English language arts, along with students’ grade point averages. Self-management skills showed the highest correlation. Low social-emotional ratings corresponded to higher rates of suspensions and absenteeism in middle school.
Social-emotional factors will count only for 8 percent of CORE’s new School Quality Improvement Index, which CORE introduced this year without including the social-emotional survey ratings. That will come next fall. CORE’s hope is that schools with high ratings will share what they do well, and schools with low ratings, particularly with subgroups of struggling students, will change instructional approaches. But many researchers remain skeptical of including soft, potentially manipulable measures for school accountability.
West acknowledges the CORE results so far are “quite limited” – and not predictive of how accurate the measures will be once they are used “in a high-stakes setting or even with the very modest weight that will be attached to them. Nor can we say anything about how CORE’s focus on social-emotional learning will alter teacher practice and, ultimately, student achievement.”
Under California’s Local Control Funding Formula, districts are required to set goals and take actions for improvement in eight priority areas, including underlying conditions for learning: school climate and student engagement. And, West noted, the new federal Every Student Succeeds Act “provides both opportunity and incentive for experimentation.”
CORE is taking the opportunity and deserves watching, he wrote.