Start – Stop – Assess – Repeat

David Bulley

This ten-minute survey will change how you teach forever.

Every day for one week I handed my 10th grade history class a form, marked in ten-minute increments. It asked two questions: Are you on task or off right now? And: What did you learn in the last ten minutes?

Time Are you on task or off What did you learn in the last ten minutes
9:40 On The French revolution was extremely bloody! There were even reports of cannibalism, but mostly it was like the whole city of Paris was a mob!
9:50 Off Something about the French revolution

I personally kept a journal taking careful note every ten minutes of what I had been doing. Was I lecturing? If so, what and how? Was I helping a student through individual work? Was I with one group, or sitting at my desk taking attendance, or grading?

At the end of the week I asked for help from some upper-class student interns and a few teachers. I asked them to evaluate each student learning block, marking between one and five. One for ‘little to no learning’, and five for the ‘most learning possible in ten minutes’. We then compiled the results on Microsoft Excel. Each day looked like this:

Time # On task # Off task Rank learning 1-5
1 20 2 5
2 20 2 5
3 19 3 4
4 17 5 3
5 15 7 3
6 17 5 3
7 19 3 4
8 19 3 5

For each day, we produced a chart like this:

I then matched my data against my lesson plan and journal to see when students were learning and when they were not. The results drastically changed how I teach and were sometimes counter-intuitive.

First and most dramatically, students were on task and exhibited most learning when I was lecturing. If the lecture was in the form of a narrative that contained a protagonist, there was rarely a single off task student. Not only could these students recount with accuracy what I had said in the last ten minutes but they were able to make connections to previous topics and draw inferences.

Students exhibited high learning and on-task reports when they were doing independent reading, but only under certain circumstances. When the assignment was to read a chapter and answer the questions at the end, students were frequently off task, and even when they reported they were on task, they very often had low learning outcomes. On the other hand, when the assignment was to read each paragraph, write the main idea of that paragraph, then answer the questions at the end, the learning shot up from a 2 to a 4! And they tended to finish the assignment much more quickly. Observationally, I believe this is because when asked to only answer questions, they read the questions first and then searched for the phrase that might answer instead of digesting the material, so even when they found the answer they reported less learning because it didn’t stick, it was just words.

Students did the worst when transitioning from one task to another. This took about twice as long as I had allotted for the transition and depending on the next task, could have very long implications, sometimes reducing learning for the rest of the class.

Students reported almost no learning for worksheets, vocabulary lists, matching worksheets, tracing maps and other individuated, non-reading tasks. This one is tricky and counter-intuitive because when the worksheet is the assessment, it looks like they are learning a great deal. But when we move the assessment outside of the assignment we see very different results.

On group projects, some worked well and others did not. I got good outcomes when the project was “staked” so that students felt they had something to gain from completion. It also had to be engaging and very interesting. It had to be structured in a way that involved everyone and solicited their involvement in observable and accountable ways. If not, then one or two leaders would complete the task while other students just quietly talked, trusting the “smart” kid to pull them through. As a way to combat this phenomenon I began handing out accountability sheets where the student not only had to hand in the completed task or make the presentation, but they also had to negotiate together what percentage of the task was their personal contribution. Everyone had to agree. If four people worked on a project, Bobby definitely wanted to make sure his contribution was 25%!

Another teacher did this experiment as well, and while some things matched, others did not. The other teacher did not get the kind of results I did for lecturing and so he adjusted his teaching style accordingly, but for myself:

  1. I lecture more.
  2. I transition less, meaning longer stretches of the same thing.
  3. I choose projects extremely carefully, looking for the high involvement and then I increase the stakes.
  4. I threw out all worksheets. Instead of a vocabulary worksheet, we all write all over the whiteboard in many colours, defining, illustrating, and illuminating each word.
  5. When I assign reading, I structure the assignment to maximize engagement, and do everything possible to make sure hunt and peck will not work.

The trend in education in recent years is to use “data driven” methods to improve your practice. But whose data? Which students? Teachers understand that just as they teach students, there is more than one way to solve a problem, there is certainly more than one way to teach it. Yes, use data.

The above narrative is a simple tool to use your data and your students, instead of generic data and average students compiled from global trends. What kind of learners are your students? What kind of a teacher are you? Find out. Then use that information.

The author is a history teacher and high school administrator in Massachusetts, U.S.A. He can be reached at

Leave a Reply