Supporting All Learners as They Interpret & Use Data
When students collect data, record it in a data table, and graph it for classwork, we could call this “making sense of data” because the activity gives students an opportunity to see and understand a concept in a different mode. They are working with measurements and observations, and the experience may serve as an exploration rather than an explanation.
But let’s push the idea of “making sense of data” a little further—beyond finding “the right answer.”
Students also need opportunities to consider what the evidence may or may not support, and connect patterns in the data to what they already know. We can help them do this in ways that support all learners.
In this three-part series, we have explored how instructional practices connected to Universal Design for Learning, or UDL (CAST, 2018), can support learners as they:
In the first blog, we focused on finding and organizing data with our learners. We followed with ideas about how learners can graph and analyze data in different ways. In this third and final post, we share some strategies for supporting learners as they interpret and use data.
Making sense of data takes practice
As educated adults, many of us have built some automaticity around looking at a data visualization and making sense of what it means in whatever context we find it—whether on our electric bill, as part of the weather forecast, or while reading/watching the morning news.
In this information age, we need to help our learners get there too, and sooner rather than later.
That means breaking the process down as we give them repeated opportunities to practice—an approach that is also reinforced through UDL (CAST, 2018). The goal is not to simplify the thinking itself, but to make the steps involved in analysis and interpretation more visible and accessible.
Below are several teaching strategies that can support scaffolded data analysis and interpretation while supporting learner variability.
Reveal the visualization gradually
Use strategies like Zoom In (Ritchhart, Church, & Morssion, 2011 and a science example in Hunter-Thomson, 2026) or Slow Reveal (cross subject area examples here) when presenting data visualizations to learners.
These approaches limit the number of details that require attention at one time (CAST, 2018), while the progressive release of information can reduce demands on working memory (Dahlstrom & Wallace, 2022).
For example, with the Slow Reveal strategy learners might first see only part of the data in a graph, then the scale of an axis, followed by the axis labels, the rest of the data points, legend, and title. At each stage, they can describe what they notice, make predictions, and revise their thinking as more information becomes available.
Many teachers find that these strategies also reinforce some of the practices involved in creating effective data visualizations. As learners work to make meaning from the data, they begin to recognize how features such as the scale of the axes, labels, and legends influence what a graph communicates.
Separate noticing from interpreting
When the Notice/See, Think, Wonder routine (Ritchhart, Church, & Morssion, 2011) is applied to data visualizations, opportunities open for all learners to contribute to the sensemaking.
Hold time and focus learners first on simply noticing what is present in the visualization:
- What is the context?
- What was measured or recorded?
- What do the colors represent?
- What shapes or patterns are present?
Give learners the opportunity to annotate and describe what they notice or see before anyone begins “thinking” or “wondering” about the meaning.
This distinction matters. Asking learners to observe first gives them time to become familiar with the visualization before they are expected to interpret it. It also creates multiple entry points into the discussion because learners can contribute accurate observations even if they are not yet ready to explain what those observations mean.
As you move learners into thinking, encourage the use of sentence starters such as “I think…” or qualifying phrases such as “maybe…” as they apply content understanding or describe how one attribute or variable may be related to another.
When learners are ready to wonder, encourage questions that arise from the data, the visualization, or the concepts being explored.
Ask what the data can—and cannot—tell us
You might discover that asking learners for both examples and non-examples can keep more learners involved as they analyze a visualization.
One useful strategy is: What can I say? What can’t I say?
Learners can first consider statements that go beyond what the visualization can show and then identify statements that are supported by the data, which is typically easier for students (Hunter-Thomson, 2020). This approach highlights patterns, critical features, and relationships while helping learners distinguish evidence from assumption or inference (CAST, 2018).
For example, a graph may show that two variables changed at the same time. Learners may be able to say that the variables are related within the dataset, but from the data visualization alone they are not able to say that one caused the other.
This kind of discussion helps move learners beyond simply identifying a pattern toward considering the strength and limits of the evidence.
Give learners multiple ways to develop and express an interpretation
Learners do not always need to begin by writing a complete explanation.
They might first talk through a pattern with a partner, draw or annotate what they notice, create a concept map, or construct a short story from the data. These options provide multiple ways for learners to organize and express their thinking before they are asked to produce a more formal interpretation.
Group discussion can be especially useful before learners write a claim. Asking students to identify patterns and evidence together allows ideas to enter the conversation before the demands of writing are added.
The goal is not to replace written reasoning. It is to give learners ways to develop the conceptual understanding that will eventually support stronger written explanations.
Consider beginning CER with the evidence
Adjust your Claim-Evidence-Reasoning, or CER, graphic organizer so that it leads with evidence.
Learners can first identify what they see in the data, then use sentence prompts to reason about what that evidence could mean and how it connects to what they already know about the topic. From there, they can develop a claim.
Graphic organizers are a well-known tool for building routines and emphasizing critical features (CAST, 2018). While we want learners to make claims supported by evidence-based reasoning, it is not the order of the letters in CER that is most important. What matters is that all three features are present and meaningfully connected.
Highlight and vary where learners begin on the organizer. Encourage collaboration as they develop their reasoning so that multiple ideas and interpretations can be brought into the conversation.
You might ask:
- What evidence do we notice first?
- What might that evidence mean?
- How does it connect to what we already know?
- What claim are we now prepared to make?
Leading learners in discussion around the connection between evidence and claim provides a graduated scaffold for information processing (CAST, 2018). It can also help learners build from the parts of an explanation toward the whole.
Help learners see interpretation as something that can change
Making sense of data does not always happen in one step.
As learners encounter more evidence, hear the interpretations of others, or revisit the context, their thinking may change. Routines such as I Used to Think…Now I Think… (Ritchhart, Church, & Morssion, 2011) can help make that evolution visible.
This reinforces an important part of data literacy: revising an interpretation is not a sign that the earlier thinking failed. It is often evidence that the learner is using new information to develop a stronger explanation.
Building on abilities learners already have
Our ability to see patterns and tell stories is something many of us begin developing from a very young age. Nurturing those abilities within data literacy is something we can recognize and support for all learners.
We all begin where we begin.
Experience and research into learning give us ways to improve what we do and how we do it. Strategies such as revealing information gradually, separating observation from interpretation, discussing evidence before writing, and adjusting familiar graphic organizers do not require an entirely new approach to instruction. They help us become more intentional about practices many of us already use.
UDL provides a lens for reducing unnecessary barriers. Data literacy gives those practices a purpose: helping learners examine evidence, consider what it means, recognize what it cannot tell them, and communicate an interpretation they are prepared to support.
As you think about the next dataset or visualization you share with learners, consider:
Where might I slow down the sensemaking process so that more learners can contribute without taking away the complexity of the thinking?
We have shared some ideas that we think work well, and we would love to hear your ideas and experiences too. What strategies or tools have you used to support learners as they interpret, use, and make sense of data?
