Opinion Blog


Rick Hess Straight Up

Education policy maven Rick Hess of the American Enterprise Institute think tank offers straight talk on matters of policy, politics, research, and reform. Read more from this blog.

Education Opinion

Five Questions About Data Use for School Leaders

By Guest Blogger — August 24, 2018 4 min read
  • Save to favorites
  • Print

Taking over the guest blog this week is Anna Egalite, assistant professor of leadership and policy at NC State. Previously, Anna taught elementary school and did a postdoc at Harvard. She’ll be writing about education-leadership research—what we know, where we have good intuitions, and where we’re still very much in the dark.

It’s back-to-school time and education reporters are highlighting stories about how school leaders are “leaning on data” to promote student learning, making administrative decisions that are “supported by a data-driven process,” and drawing on their experience in “data-driven instruction.” This all sounds great and it’s clear that school systems—both public and private—are collecting more and better information about students, teachers, and schools. But are school leaders adequately prepared and empowered to interpret and act on these data, or is the biggest change simply a shift from cardboard to cloud-based file storage boxes?

In the spirit of self-reflection, I’ve come up with five big questions about data use for school leaders.


  1. You probably know how well the average student in your school is performing on standardized tests, but are your overall scores masking serious subgroup disparities? Thanks to federal accountability policy, school systems have traditionally broken out scores by gender, race/ethnicity, English language learner, and special education status, but if you have a sufficiently large sample size, why not get creative in how you think about categorizing students for analysis? Consider comparing the students of novice versus veteran teachers, comparing recently arrived immigrants to those who’ve been in the country for at least one year already, comparing outcomes for emotionally disturbed students to other categories of special education students, or comparing students across the different categories of English Language learners. Which specific subgroups make sense to examine is going to vary depending on your population, but that’s the point. Your student body is unique and your data analysis should reflect that.
  2. Is there a healthy amount of diversity in the types of data you’re analyzing? Everyone agrees that test-score data are easy to analyze, but they paint an incomplete picture. An effective leader is cognizant of who is chronically absent and when; who’s being disciplined in their school, how harshly, and how frequently; and which students are repeatedly overlooked for special programs, such as gifted and talented opportunities. If you’re not monitoring trends in these data, you could be selling your students short, without even realizing it. Okonofua and Eberhardt offer a simple yet powerful demonstration of educators’ potential blind spots in this regard, using experimental evidence to show that teachers interpret repeated discipline infractions as more severe when they are associated with a black student, as opposed to a white student, even though the description of the misbehavior is identical except for which student name was used (Darnell or Deshawn versus Greg or Jake).
  3. How has the demographic composition of your student body changed over time and how have you embraced the various cultures students bring to your school community? Thoughtful school leaders can make both short- and long-term plans to celebrate students’ backgrounds, such as embracing efforts to build a representative faculty to support students of color by targeting teacher hiring efforts with diversity in mind and offering targeted professional development for existing teachers that encourages an empathic mindset.
  4. How do you use data to empower parents? We already know that great principals spend a lot of time pouring over spreadsheets to gain a richer understanding of how students are progressing and to better manage teachers and resources to support learning, but have you thought about how your school system is using data to empower parents? As statewide school choice programs grow in size and scope and we try to make sense of mixed findings of their impact, parents need help making good school choices. States might lead the effort, but principals should be part of the conversations about how to refine school report cards so that families have good information at their fingertips when the time comes to choose a school, regardless of whether that’s a traditional public, charter, or private school.
  5. Who are your most impactful teachers and what are you doing to support them? Your teachers are the most valuable resource in your organization. Students assigned to high value-added teachers are more likely to attend college, less likely to have children while still a teenager themselves, and enjoy higher incomes in adulthood. It’s hard to provide workplace support and take efforts to improve their retention if you haven’t even identified the most effective teachers in your school.

Principals, if you can’t answer these questions, you’ve got a data problem. Perhaps your master’s in school administration or education leadership certification program did not offer or require rigorous statistical training, but that’s a weak excuse given the plethora of free professional development and online learning opportunities available to you today. Remember, being “data-driven” doesn’t mean you never rely on anecdote or intuition. If you’ve got a gut feeling about something that’s going on in your school, try investigating the patterns in your data with simple visualizations that summarize vast datasets into simple tables and graphs. This school year, take charge of your data.

Anna Egalite

The opinions expressed in Rick Hess Straight Up are strictly those of the author(s) and do not reflect the opinions or endorsement of Editorial Projects in Education, or any of its publications.