What Counts as Evidence? The Question Every Observation System Has to Answer

Walk into a classroom during an observation, and there is no shortage of things to notice. A teacher asks a question. Several students raise their hands. Two students begin discussing an idea. Another student is writing independently. The teacher moves across the room, checks one student's work, redirects another, and asks the class to explain its thinking.

All of it happened. But does it all count as evidence?

That's a more important question than it might seem. Because before an observer can connect what happened in a classroom to a rubric, assign a score, or offer meaningful feedback, someone has to decide what evidence of the practice actually looks like.

Seeing Something Isn't the Same as Knowing What It Means

Consider something as familiar as student engagement.

An observer might record that nearly every student raised a hand. Another might note that students worked quietly throughout the lesson. Someone else might focus on the kinds of questions students asked, the choices they made, or whether they were doing the intellectual work rather than simply following directions.

Each of those observations tells us something about the lesson. But they don't necessarily tell us the same thing about engagement.

That's where an observation system does important work. Research on classroom observation has shown that observation instruments don't simply organize what observers see; they help define which aspects of teaching receive attention and how those practices are understood. Klette and Blikstad-Balas (2018) describe observation manuals as “lenses” on classroom practice and note that the definitions within those tools influence what counts as evidence of particular instructional practices. 

In other words, a rubric can tell observers that engagement matters. But observers still need a shared understanding of what evidence of engagement looks and sounds like when a lesson is actually happening.

Evidence Has to Be More Than a List of What Happened

Observation notes can easily become a running account of a lesson:

Teacher asked students to turn and talk. Students discussed the question with partners. Teacher circulated around the room. Several students shared responses.

Those are observations. But their usefulness depends on what happens next.

Which professional practice do those observations help illuminate? What evidence would distinguish students who are simply completing an assigned activity from students who are genuinely thinking, questioning, collaborating, or making meaningful decisions about their learning?

This distinction matters because observation systems ultimately ask people to make professional judgments. Hill et al. (2012) argue that reliable classroom observation requires more than an observation instrument; the larger system must also address how observations are scored and how raters are trained and certified. 

If the path between what happened and what it means isn't clear, two observers can collect perfectly accurate notes and still reach different conclusions.

Start With a Better Question

When organizations work to improve teacher evaluation, the conversation often begins with the rubric: Are the indicators clear enough? Are the performance levels specific enough? Do observers need more training?

Those questions matter. But there is another question that should come first:

What would we actually need to see or hear in a classroom to know that this practice is happening?

That question shifts the conversation from abstract language to observable practice. It also gives observers something concrete to discuss when their interpretations differ.

This is central to Evidence-First™. The goal isn't to replace an organization's existing rubric or eliminate professional judgment. Evidence-First strengthens the connection between professional expectations and the evidence observers use to understand classroom practice.

For schools, districts, and teacher preparation programs, that creates an opportunity to look at an existing observation system differently. Instead of asking whether the rubric needs to change, the better starting point may be examining how clearly the system defines the evidence behind it.

References

Hill, H. C., Charalambous, C. Y., & Kraft, M. A. (2012). When rater reliability is not enough: Teacher observation systems and a case for the generalizability study. Educational Researcher, 41(2), 56–64. https://doi.org/10.3102/0013189X12437203

Klette, K., & Blikstad-Balas, M. (2018). Observation manuals as lenses to classroom teaching: Pitfalls and possibilities. European Educational Research Journal, 17(1), 129–146. https://doi.org/10.1177/1474904117703228