We work with a lot of teacher teams and leadership teams. They often feel unsure about having deep conversations about data and evidence. “Data” refers to raw facts, figures, or information collected from observations, measurements, surveys, tests, or other sources. In an education setting, data can include test scores, attendance records, survey responses, or classroom-observation notes. “Evidence” refers to data that have been analyzed, interpreted, or used to support a claim, conclusion, or decision. Evidence goes beyond the raw numbers to show meaningful patterns, trends, or findings that can inform action.

And that is where Michael’s 3-year-old granddaughter comes into the equation. OK, almost 3 years old.

“It’s mahogany, Papi.”

Looking at crayon colors with Maisie, he asked her to name the color of a particular crayon.

There were many colors and several in the brown tones, so given that she had so many shades of brown, he was a bit taken aback that she named the crayon so specifically. Mike rolled the crayon in his hand to find the name. Maisie was correct. It was mahogany.

“How do you know this, Maisie?”

This time she looked at him with an expression that said, “How can you not know this, Papi?”

She answered, “It has more red in it, Papi. I like to color hair with it.”

Later on, Peter heard Maisie say the color at breakfast and was completely blown away that such a young child knew the name “mahogany.” Besides being a charming story, what does this have to do with the adults in school who will one day teach Maisie and be open to her teaching them a thing or two?

Three-Legged Stool
In our work with school leadership teams, we explain that shared understanding, joint work, and evidence of impact are what we call the three-legged stool of collective leader efficacy. During the evidence of impact “leg,” we often share data expert and researcher Victoria Bernhardt’s four types of data: Demographic, Perceptions, Student Learning, and School Processes.

We poll our participants, asking such questionsas:

A middle school reports an increase in chronic absenteeism. Surveys show that students feel disconnected from school. Which type of data does this scenario best align with?

• A) Demographic
• B) Perception
• C) Student Learning
• D) School Processes

The group quickly discovers that the answers vary widely. Our workshop participants banter (sometimes emotionally) about their selections. Some will share emphatically that their answer is the correct one. We let the conversation continue and draw them back to some of our opening slides that contain quotes from the book Both/And Thinking by Wendy K. Smith and Marianne W. Lewis. Immediately, participants acknowledge that there may be one best answer depending on the purpose or reason for gathering the data, or it may be powerful to understand how two or more measures might work together for better understanding. We then lead the group to the power of intersection of these four types of data.

Bernhardt states, “It is not until we intersect all four circles, at the school level, and over time that we are able to answer questions that will predict if the actions, processes, and programs that we are establishing will meet the needs of all students.” With this intersection, we can answer the foremost question:

• Based on whom we have as students, how they prefer to learn, and what programs they are in, are all students learning at the same rate?

Looking Deeper
Maisie didn’t just name the crayon color “mahogany,” she understood it. She saw something within the color that many adults might overlook. She recognized that mahogany had more red in it and she could articulate why that detail mattered to her: “I like to color hair with it.” For her, the crayon wasn’t just a label, it had depth, purpose, and context. That’s the kind of discernment we need from school leaders when they work with data.

Too often, data conversations in schools are confined to surface-level observations, test scores, attendance rates, survey results, each treated as a stand-alone piece of information. But much like Maisie saw the nuance in a crayon color, educational leaders must learn to look beneath the surface of each data point and ask, “What is this really telling us?” They need to develop a lens that seeks patterns, contradictions, and connections across seemingly unrelated data sets.

Just as mahogany reveals its richness through the red it holds within, data become truly informative when we examine the information through an integrated lens. The intersection of Bernhardt’s four types of data—Demographic, Perception, Student Learning, and School Processes—is where leaders and their teams have to spend time so they can become more informed.

When leaders bring these perspectives together, they move from simply naming problems to understanding root causes. They begin to ask better questions, design more impactful interventions, and engage their teams in joint work that is grounded in evidence and not just gut feelings or assumptions. It’s in those intersections that we can finally address school improvement.