If cognition defines what we want students to think, and observation determines how we make that thinking visible, then interpretation is where we answer the most important question:

What does this student work actually reveal about their thinking?

This is the third corner of the assessment triangle—and, in many ways, the most complex. Our best-laid plans, or in this case, assessments, often go awry because, well, we are dealing with humans. Making sense of student responses—which can sometimes travel down unexpected interpretative paths—can still be disorienting.

When a student evaluates  a House of Commons speech given by John A. Macdonald in 1866 and concludes, “This source is biased because Macdonald supported Confederation,” how should we interpret that?

• Do they understand how perspective shapes evidence? 
• Are they using “bias” as a vague label? 
• Are they connecting author → perspective → limitation? 

A more sophisticated student might write: “This speech reflects Macdonald’s support for Confederation, as he is clearly emphasizing benefits and downplaying risks to persuade those on the fence in Parliament.”

Here, the analysis is clearer. Vague notions of bias are replaced with a more precise explanation of how authorship and context shape what is included and omitted. This is the thinking we are interested in. 

While the example above is relatively straightforward, other factors—such as literacy demands—can sometimes obscure students’ historical thinking. A student may have a strong grasp of sourcing but struggle to express it clearly, while another may write in polished, academic language that sounds convincing without demonstrating genuine understanding. Without a clear model of what proficient sourcing looks like, these responses can easily be misinterpreted.

Consider the responses below:

• Student A: Strong Thinking, Weak Expression

“Macdonald is for Confederation so in this speech he mostly talks about the good things and not really the bad stuff because he wants other government people to agree with him and vote for him so he can continue to do the things he wants as Prime Minister.”

Although the student’s phrasing is awkward and informal (“good things,” “bad stuff,” “government people”), they identify Macdonald’s position on Confederation, explain how that position shapes the content of the speech, and recognize the purpose of the speech—to persuade members of Parliament to support Confederation. Despite the weak writing, the historical thinking is strong.

• Student B: Weak Thinking, Strong Expression

“This speech reveals Macdonald’s clear bias. His perspective influences what he includes and how it is stated, which is shaped by his audience and may limit the source’s reliability.”

This response contains much of the academic language teachers often look for (“perspective,” “audience,” “reliability”) and is written in fluent, controlled prose. However, it never explains how Macdonald’s perspective shapes the content of the speech. There is no reference to Confederation, Parliament, or the persuasive purpose of the address. Instead, “bias” functions as a label rather than an explanation. In fact, this response could be used as a generic template for almost any historical source.

This is where our earlier work—breaking down sourcing into specific cognitive moves—becomes essential. Interpretation depends on having clear criteria for what counts as evidence of thinking. Instead of asking, “Is this a good answer?” we ask:

• Did the student identify relevant sourcing information? 
• Did they connect that information to perspective or reliability? 
• Did they use that connection to qualify the usefulness of the source? 

In other words, we interpret student work by mapping it back onto the cognitive moves we value.

One of the most powerful outcomes of this approach is that it allows us to see partial understanding with precision. And then interpreting student work becomes less of an endpoint, and more like a bridge to instruction. Once we can see student thinking clearly, we can:

• Provide specific feedback 
• Adjust instruction 
• Design next steps that target actual needs 

I set out to design assessment tools for Canadian History teachers. But over time, I’ve come to see them differently. These are not just assessment tools—they are teaching tools. Yes, they can be used for evaluation. But the real value lies in the conversations they generate and the feedback they make possible. Those moments—when thinking becomes visible and can be responded to—are where the most meaningful learning happens.

We can have a clear, well-articulated model of historical thinking. We can name the moves: sourcing, contextualization, corroboration, inference. We can even break down each guidepost move into precise, components. But none of that matters if the task we design does not actually elicit those moves from students.

This is the central challenge of observation: how do we design a task that makes invisible thinking visible?

Observation should begin with the specific cognitive moves we want to see. If our model tells us that sourcing includes identifying author, purpose, and audience, then the task must require students to do those things in a visible way.

Below is a series of questions a teacher might use to elicit historical thinking in relation to the provided source. 

• Question A: Analyze the Source. 
• Question B: Why might this source be useful as evidence of what happened during the battle at Ticonderoga/Lake Champlain? What about this source might make it less useful as evidence of what happened during the battle at Ticonderoga/Lake Champlain?
• Question C: “How does the author’s position and the time in which the source was created affect its reliability?
• Question D: To what extent is this source useful in understanding what happened during the battle of Ticonderoga. Your answer must compare the sources using at least two of the following:
  • Author/creator 
  • Time of creation 
  • Purpose/audience
• To what extent can the painting “Jacques Cartier Meeting the Indians at Stadacona, 1535” be used as evidence of what actually happened during the meeting?
  • In your response, you must:
    • Identify when the painting was created relative to the event 
    • Explain the purpose of historical paintings like this 
    • Explain what the painting can show historians 
    • Explain what it cannot reliably show

Which question do we ask? My honest answer is, it depends. 

Broadly speaking, I’d suggest that constraints help reveal thinking. Particularly for novice historians. The best tasks must require students to attend to author, time, and purpose—not simply allow for it. This often means embedding those elements directly into the structure of the question or response space. Rather than asking, “Is this source reliable?”, we might ask, “How does the author’s position and the time in which the source was created affect its reliability?”

The difference is subtle, but significant. In the latter, the thinking is specified—and therefore more likely to be visible in student responses. Constraints such as requiring students to justify a claim using specific sourcing factors make particular forms of thinking more probable and can help us pinpoint exactly where student reasoning breaks down. 

But they also introduce a new problem: over-scaffolding the thinking itself. 

While a constraining question may be helpful at younger grades, at a certain point, we will want to determine whether sourcing has become a habit of mind, that students are attending to the necessary metadata independently rather than simply following instructions.  

The goal is not just that students can perform sourcing when prompted, but that they begin to internalize these moves—to notice author, context, and purpose without being told.

For this reason, most tasks on To The Past aim for a middle ground, often leaning slightly toward less structure. Questions such as “What are the strengths and limitations of this source as evidence for this inquiry?” are intentionally open enough to require students to select and apply relevant sourcing moves on their own. And, depending on their student performance and assessment data, to allow teachers to provide more flexible and responsive scaffolding as needed. Some students may need explicit prompts which can be provided orally or written on a classroom whiteboard (“Consider the author, time, and purpose…”), while others may be ready for more open ended tasks. 

Through careful guidance and prompting from experienced teachers, all students can move towards these more complex open-ended prompts without need for explicit prompting. 

Designing for observation is ultimately about discipline. It requires us to stay tightly aligned to our model of cognition, to be intentional about how tasks elicit specific thinking moves, and to use structure and constraint in ways that make student thinking visible.

When this is done well—and it is difficult—the need to “read between the lines” is reduced. Instead, we are able to read student thinking more directly in their responses, because those responses have been deliberately shaped to reveal it.

That clarity is what makes meaningful feedback possible. 

Designing assessments that genuinely capture students’ historical thinking remains one of the most persistent challenges in Social Studies education. Too often, what we assess drifts toward recall or vague “critical thinking,” rather than the disciplined habits of mind that define historical inquiry – “the process of interpreting the past through sourcing, contextualizing, corroborating, and constructing evidence-based accounts.”

Because historical thinking is complex and difficult to surface, effective assessment must be built on a clear foundation. We need to articulate how students think within the discipline (cognition), design tasks that make that thinking visible (observation), and determine how to interpret what students produce (interpretation). These three elements form the “assessment triangle,” described by Pellegrino, Chudowsky, and Glaser (2001) , and provide a practical framework for building valid and reliable assessments.

Drawing on this model, Seixas et al. (2015) translate the triangle into three guiding questions for history educators:

1. What model of student thinking is being assessed?
2. What tasks will elicit evidence of that thinking?
3. How will student responses be interpreted?

Over the past year, I have taken up these questions in an effort to design more precise and usable historical thinking assessments, work that now lives on ToThePast.ca. In this first post, we’ll provide the Cole’s Notes answers to these questions, and in future posts, unpack each in greater detail.

What model of student thinking is being assessed?
Any strong assessment begins with a clear model of cognition. In history education, this means grounding our work in a disciplinary framework rather than generic skills. I chose to work within the “Big Six” historical thinking concepts, in part because they are embedded across provincial and territorial curricula, and because they offer a shared language for teachers.
Within this framework, I narrowed my focus to the benchmark of evidence. Regardless of whether students are analyzing cause and consequence, continuity and change, or historical significance, their claims must ultimately be grounded in the historical record. Evidence is the common denominator.

Building on Seixas and Morton’s work, this benchmark can be broken into key cognitive moves: questioning, sourcing, contextualization, drawing inferences, and corroboration. Being explicit about these moves matters. If we cannot name what historical reasoning looks like, we make it harder to teach, harder for students to learn, and harder to assess with consistency.

What tasks will elicit evidence of that thinking?
Once the cognitive target is clear, the next step is designing tasks that allow students to demonstrate it.

Here, the work of the Stanford-based Digital Inquiry Group (DIG) was foundational. Their Historical Assessments of Thinking (HATs) demonstrate how short, tightly designed tasks—often built around one or two primary sources—can make student thinking visible. Crucially, these are not extended essays. They rely primarily on short-answer formats, reducing literacy demands while still targeting sophisticated thinking.

Taking inspiration from DIG task models and developing further alternatives I created a bank of approximately 60 ready-to-use assessments through To The Past. Each task is built around a focused source set, aligned to a specific cognitive move, and designed to be used in regular classroom conditions.

How will student responses be interpreted?
In earlier iterations, I relied on teachers to interpret student responses independently—to anticipate what strong and weak answers might look like. However, this likely limits both reliability and instructional impact. More recent work has begun to address this through the development of rubrics and exemplar banks. These supports help clarify what distinguishes emerging, developing, and sophisticated historical thinking. They also serve a dual purpose: not only improving assessment consistency, but functioning as instructional tools that make expectations visible to students. We hope that you consider using To The Past as both an assessment resource and an instructional tool—one that makes historical thinking more visible, more precise, and more teachable. The goal is not simply to evaluate students more efficiently, but to sharpen the alignment between what we value in historical inquiry and what we actually ask students to do in our classrooms.

Stay tuned. As mentioned, we will be providing weekly posts that explore each of the three questions posed above in greater detail. Next week: What model of student thinking is being assessed? We will do a deep dive into how we can break down each of the individual competencies embedded within the benchmark of Evidence to more effectively teach, assess, and support our students in mastering these skills.

Guest Contributor: Nathan Moes