How schools are getting smarter about analytics.

GUEST COLUMN | by Aaron Feuer

The field of analytics, though well established in the corporate world, is rather new to schools. Districts don’t typically employ data scientists, and most schools track only attendance and test scores. But this is changing – driven partly by teacher and parent requests, schools are getting smarter about analytics. No one would try to argue that attendance records and test scores provide a complete picture of the quality of the learning environment. But what metrics should schools track? In our work with K-12 school districts, we’ve developed a list of approximately 50 softer metrics that impact learning, including:

–   Student Perceptions of School Safety

–   Parent Involvement

–   Perceived Quality of School Leadership

–   Student Grit

–   Perceived Quality of Teacher-Student Relationships

So what should school administrators do to get started with analytics? First, of course, you need to start collecting data.

The value of Teacher-Student Relationships was demonstrated in a study conducted last year by Dr. Hunter Gehlbach, Associate Professor at the Harvard University Graduate School of Education and now Director of Research at Panorama. By providing feedback that showed teachers and students a few of the things that they had in common, Dr. Gehlbach and his colleagues saw relationships and academic performance improve in the classroom.

For the study, Dr. Gehlbach and his research team surveyed students and teachers about their interests and values. Based on prior research that highlights how learning about similarities can strengthen relationships between two parties, they hoped that providing feedback about shared commonalities might help teachers and students better connect with each other. Sample questions asked of students and teachers included what were the most important qualities in a friend, which sporting event they would most like to attend, and when students learn the most.

A week after this initial “get-to-know-you” survey, students and teachers were randomized into different experimental conditions. The treatment group received feedback on what they had in common with the other party, and the control group did not. Five weeks after that, both students and teachers completed a more comprehensive survey to measure the effect of receiving this similarity feedback. The research team also gathered classroom grades at the end of the first quarter.

The study found that five weeks after the intervention:

• Students and teachers who learned what they had in common with the other party perceived themselves as being more similar.
• When teachers learned that they shared commonalities with their students, they rated their relationships as more positive. (By contrast, the intervention did not significantly affect students’ perceptions of their relationship with their teachers).
• When teachers received feedback about being similar to their students, the students earned higher grades (though the effect was far more profound in Black and Latino students – something Dr. Gehlbach’s team will explore further in future research).

These early results suggest that similarities between students and teachers may be a promising lever to improve teacher-student relationships, and demonstrate that small interventions can have outsized impact on the learning environment.

So what should school administrators do to get started with analytics? First, of course, you need to start collecting data. Surveys of students, teachers, administrators and parents are a great way to do this. There are free, open-source survey tools that can help schools measure student perceptions; that may be a great place to start.

Once you have some data collected, here are three actions to take:

1. Draw relevant comparisons – Look for ways to draw comparisons between key stakeholders within your school– for example, look for commonalities or differences between students by demographics or grade level, or compare schools in aggregate to others within the district. Are some groups being better served by the school than others? Which populations seem most important to focus on?
2. Look beyond one source of data – Survey results can highlight the unique perspectives of students. But you may consider combining that information with third-party/demographic data from external sources as well. For instance, does data on income or ethnicity predict which parents feel as though the school provides the best fit for their children? Perhaps after-school or extracurricular participation is associated with students’ performance in school.
3. Measure changes over time – Collecting feedback through surveys should be an ongoing activity. Make a plan to follow up with your results in order to measure how response data changes over time. How are student perceptions changing over the course of a school year? How does feedback from the same grade level change as students advanced to higher-grade levels? Have the programs you implemented last year been well received by students?

The use of analytics within schools is only going to become more prevalent. Districts that understand this, and take action to harness and leverage all of that data, will be better positioned to improve the learning environment and prepare today’s students to be tomorrow’s leaders.

—

Aaron Feuer is founder and CEO of Panorama Education, which helps 6500+ K-12 schools improve through data analysis and feedback survey.