Let Students Write the Quiz
Explore how having students generate their own quiz questions can deepen retrieval practice, boost comprehension, and reveal whether they truly understand the material. The conversation also breaks down research on question generation, the gap between rereading and active recall, and simple stems to scaffold higher-order thinking.
Chapter 1
The quiz you don't write
Renata Salas
So I was- I was reading this piece by Youki Terada in Edutopia from, oh, back in September of 2020, about how to use student-generated questions to promote deeper thinking. And then, just- just recently, July of 2025, Daniel Leonard revisits it in this list of fifteen quick retrieval practices. And it hit me, Colin, we are- we are completely doing retrieval wrong when we do all the work.
Colin Whitfield
Right, the classic teacher-as-quizmaster trap. You spend your prep time crafting these beautiful, elegant questions, and the kids just... consume them. Or try to avoid them. So, what is Terada actually suggesting here? Is it- is it literally student-generated questions? Like, they write the test?
Renata Salas
Yes! Exactly that. Question generation. The mechanism is so simple it almost feels like a cheat code. Your students write the quiz questions about yesterday's lesson. But here is the twist: the act of writing the questions *is* the retrieval practice. It is not about them answering them, and it is definitely not about you grading them. It is- it is the cognitive lift of constructing the prompt itself.
Colin Whitfield
Right, so it- it looks like you're just being a bit lazy, letting the kids do your job, but cognitively, you are actually turning the tables. They can't just passively sit there and let the information wash over them. They have to- to actively pull that knowledge out of their long-term memory, organize it, and shape it into something another human can understand.
Renata Salas
Yes! And I- I actually tried this. Third period ELA, we're doing characterization in *The House on Mango Street*. Normally, I'd start with a quick slide, you know, "What does Esperanza's house symbolize?" Instead, I gave them ten minutes. I said, "You are writing three questions for a quiz. One recall, one 'why' question, and one 'how' or explain question." And- and the room was dead silent. They were- they were sweating over it.
Colin Whitfield
Oh, I bet. Because they can't just look at the page and highlight a random sentence. To ask a "why" question, you have to actually understand the "why" yourself first. I remember the first time I did something similar in a chemistry lab—asking them to write the troubleshooting guide for an experiment instead of just doing it. They suddenly realized *why* the temperature had to be exactly seventy degrees, because they had to explain what went wrong if it wasn't. They knew it so much better when they had to ask the question.
Renata Salas
Exactly! If you can't formulate the question, you don't actually know the material. You're just familiar with it. And those are- those are two very different things.
Chapter 2
Why writing the question beats rereading the page
Colin Whitfield
Well, and this is where the research gets incredibly robust. Mirjam Ebersbach and her team at the University of Kassel did this brilliant study in 2014. They wanted to compare three things: students who just reread the material, students who did traditional self-testing, and students who did question generation.
Renata Salas
Oh, interesting. So rereading versus actually making the questions. What did they find?
Colin Whitfield
Well, the numbers are- they're pretty stark. On the final assessment, the reread group scored forty-two percent. But the question-generation group? They scored fifty-six percent. That is a fourteen-point jump, Renata, just from changing *how* they engaged with the study time.
Renata Salas
Fourteen points? That is- I mean, in classroom terms, that's a whole letter grade and then some. Just for writing questions instead of staring at the textbook?
Colin Whitfield
Precisely. And Ebersbach explained why. She wrote, "question generation promotes a deeper elaboration of the learning content. One has to reflect on what one has learned and how an appropriate knowledge question can be inferred from that knowledge." It's- it's that word, *inferred*. You are working backward from the conclusion to the premise.
Renata Salas
That makes so much sense. Because when you reread, your brain is just saying, "Yep, seen that, know that, familiar, move on." But to infer a question, you have to decide what actually matters on the page. You have to- to weigh the concepts. Like, is this a major theme or just a minor detail?
Colin Whitfield
Exactly. And it gets even better. There was a 2018 study looking at this through the lens of Bloom's taxonomy. They found that students who were coached to write *higher-order* questions—not just "what is the definition of X," but "how does X apply to Y"—those students scored seven points higher on their actual finals than the kids writing basic recall questions. So the- the depth of the question directly mirrors the depth of the learning.
Renata Salas
Okay, but Colin, let's- let's be real for a second. If I just tell my eighth graders to "write a higher-order question," half of them are going to write, "Why did Esperanza do that?" which is- it's vague, it's lazy, it doesn't actually show they've processed anything. Aren't we just going to get a lot of really bad, surface-level questions if we don't hold their hands?
Colin Whitfield
Well, yes, if you just leave them to their own devices, absolutely. But that is where the scaffolding comes in. You don't just say "write questions." You use question stems. You give them the sentence starters—things like, "What is the relationship between..." or "How would you show..." or "What would happen if..." That structure is what nudges them away from simple copying and forces them into transfer and reasoning.
Renata Salas
Right, right. The stem does the heavy lifting of the structure so their brains can focus on the content. It's like training wheels for cognitive processing. I- I can see that working beautifully, actually. It's not about them being perfect test writers; it's about the struggle of trying to fit their knowledge into that mold.
Colin Whitfield
Exactly. It's the productive struggle. Well, that is our quick take on question generation. Give it a go in your next lesson—let the kids write the quiz. See you next time.
Renata Salas
Bye everyone!