If you’ve spent more than five minutes in a staff room, you’ve likely heard the term “formative assessment” tossed around like a pedagogical hot potato.
But few people articulate its power as clearly as Dylan Wiliam. In his book, Embedded Formative Assessment, Wiliam argues that if we want to improve student achievement, we shouldn’t be looking at new curricula or fancy tech—we should be looking at the quality of the teacher in the room.

While the whole book is a masterclass in evidence-based practice, Chapter 2: The Case for Formative Assessment is arguably the most vital. It is the “why” behind everything else. In this chapter, Wiliam doesn’t just build a foundation; he systematically dismantles the distractions that often plague our professional development.

Slaying the Educational “Dragons”

Before Wiliam introduces the solution, he spends time clearing the deck of popular but ineffective strategies. This section is a breath of fresh air for anyone tired of “brain-based” gimmicks.

• The Myth of Learning Styles: Despite 90% of teachers in some countries believing that matching instruction to a student’s “style” (visual, auditory, kinesthetic) helps, Wiliam is blunt: there is no evidence for it. In fact, he argues that trying to make learning “easy” by matching a style can actually reduce learning. True retention comes from “desirable difficulties”—making the brain work to make sense of information.
• Neuromyths: The chapter takes a sledgehammer to the idea that we only use 10% of our brains or that “left-brain vs. right-brain” dominance determines how we learn. Wiliam argues that while neuroscience is fascinating, it’s cognitive psychology that provides the actual “heavy lifting” for classroom practice.
• The Subject Knowledge Trap: This is perhaps the most counterintuitive part. Research suggests that while a “profound understanding” of the school-level curriculum is vital, simply increasing a teacher’s advanced subject knowledge (e.g., a math teacher getting a PhD in advanced calculus) has almost zero impact on student achievement. You can know the subject perfectly but still be unable to “elicit student thinking” or understand why a child is struggling with a basic ratio.

It’s a Process, Not a Product

One of the most vital takeaways is Wiliam’s insistence that formative assessment is a process, not a tool. He traces the origins back to Michael Scriven (1967) and Benjamin Bloom (1969), clarifying the vital distinction:

1. Summative Evaluation: Used by administrators to decide if a finished curriculum or student grade is sufficient. It’s a post-mortem.
2. Formative Evaluation: Used during the teaching-learning process to provide feedback and “correctives.” It’s the check-up that prevents the need for a post-mortem.

Wiliam explains that describing an assessment as “formative” is actually a category error. An assessment is just an instrument. What makes it formative or summative is what we do with the data. As he famously writes, calling a test formative is like “describing a rock as happy”—it’s a property it simply cannot have.

The Evidence: Double-Speed Learning

Wiliam cites a massive body of research, including his own 1998 review with Paul Black. After looking at over 250 studies, they found that the effective use of formative assessment can increase the rate of student learning by 50 to 70 percent.

In a world where we are constantly told we need more “instructional time,” formative assessment effectively gives us more time by making every minute twice as productive. The secret, however, is systematic action. Research by Fuchs and Fuchs (1986) found that the benefit of assessment doubled when teachers used systematic rules to decide what to do next, rather than just relying on their own judgment.

From “Data-Driven” to “Decision-Driven”

We’ve all been told to be “data-driven.” Wiliam flips this on its head. He argues that most school data is a “data-push”—reports that arrive weeks after a test, long after we’ve moved on.

Instead, he advocates for “decision-pull.” Think about a history teacher using exit passes. If she collects 30 cards to see if she should move on tomorrow, she doesn’t need to grade them. She is collecting the minimal amount of information needed to make the next instructional decision. She didn’t even have the kids put their names on the cards because individual feedback wasn’t the goal—identifying the “starting point” for tomorrow was.

The Teacher as an “Engineer”

Wiliam concludes the chapter by redefining the teacher’s job. You are not a “transmitter of knowledge,” nor are you a “facilitator” (a word Wiliam calls “the F-word” because it implies the teacher just hangs around hoping learning happens).

Instead, the teacher is an engineer of effective learning environments. To help us navigate this, he provides the Five Key Strategies:

1. Clarifying, sharing, and understanding learning intentions.
2. Eliciting evidence of learning.
3. Providing feedback that moves learning forward.
4. Activating students as instructional resources for one another.
5. Activating students as owners of their own learning.

Conclusion: The Dog Whistle

The chapter ends with a humbling joke.

Amy says she taught her dog to whistle. Betty asks to hear it, but the dog is silent. “I thought you said you taught him?” Betty asks. Amy replies, “I did. He just didn’t learn it.”

We often mix up teaching and learning. But because learning is “mysterious” and “unpredictable,” teaching must be a contingent activity. What we do next must depend on what the students have actually learned.

Formative assessment is the bridge between the two. It acknowledges that teaching is “so difficult, so complex, that one lifetime is not enough to master it,” but provides a roadmap for every teacher to get better, starting tomorrow.