Wednesday, June 28, 2017

Estimation 180 Bins

I'd love some feedback on this tool I'm tinkering around with right now.

I was inspired by Graham Fletcher, seeing him place student estimates inside ranges, or bins if you will. My goal was to make a tool that replicated this experience, is efficient for teachers to use, and added some visual enhancements. The tool could be used with Estimation 180 challenges and was made in Desmos.

What successes might you see happening in your classroom when using this tool?
What challenges might you see happening in your classroom when using this tool?
What advice do you have for improving it?

Thanks in advance for your responses.



  1. I like this Andrew. Question I have for both you and Graham: how do you set the ranges? I admit I've only asked for estimates this way a handful of times (I always forget), but when I do I'm never really sure how to set them. I always think I've made them either too small or too large.

    1. I'll admit that I have not tried this exact tool with students yet. I have tried the bins idea with students and wrote their ranges on a large whiteboard. I made an executive decision at the time to make ranges (in terms of seconds), 05, 6-10, 11-20, 21-30, 31-40, 41-50, 51-60, 60+.
      Additionally, I paused the video reveal at some point and allowed students to change bins if they wanted to now that they had more information. There's definitely a process I left out of the two-minute video above.
      I think this post deserves a follow up post to go deeper into some of the nuances.
      Regarding your question, this is why I'm tinkering around with the tool. I like the idea of listening to student submissions of what might be "too low" and "too high" while still including the range of numbers from zero to the "too low" and everything greater than the "too high".
      Although it might not include the congruent bin sizes of a traditional histogram, that's one thing I pleasantly find unique to this tool. Based on some of the feedback, I do like the idea of looking to see where the first round of estimates fall and then ask the students if they want to change the too low and too high so it better reflects the class estimates.

    2. Absolutely I agree that the video pause and opportunity to change bins is really important. The bins idea gets at the issue I was exploring in the recent estimation post I published.
      Asking which bin you fall in, not what your estimate was emphasizes the sense-making, not the answer-getting and is a good solution to that problem.
      In your example above you made a decision to go first in increments of 5, then increments of 10 up to 60+. What thought went into that decision? And then what happens if there's a much bigger difference between the too low and the too high? How many bins do we want? These would be decisions that teachers could think about beforehand but would then have to make on the fly based on what their class's too low and too high was. I guess I'm looking for a rule of thumb. Maybe it's a stats thing?

  2. Very cool Andrew! I get a similar visual display by using the number slide with Pear Deck which produces a box & whisker plot / number line display of the student estimates. You can see an example in this post:
    What I love about yours is that the kids are watching as you build it and the upper & lower limit come from the too high, too low. Great work!

    1. I love Pear Deck!
      It's a great tool and I appreciate you sharing.
      Thanks for your feedback. Let me know if you try it with students.

  3. I love the possibilities for class discussion based on visualizing with the histograms, well before students even tinker with and then eventually find out the actual amount. It might also be interesting to see if students want to change their estimates based on what they see in the histogram. (Kind of like the "old" raisin activity where students changed their estimates of how many raisins in the box once they saw what was in the top layer.) Could there be a way to have one histogram lay on top of another?

  4. I don't have any advice per se, but we do Estimation 180 daily and use TI Navigator with our estimates at least once per week. Students submit their too low, high, estimate and confidence (they record their reasoning on a tracker sheet but only submit numerical data). When we do it without technology, I just sample--usually two kids for each of the data points above. Sometimes that happens to be two kids with particularly low or high answers. When we do it this way, its 100% their own thinking and not influenced by what others think. (As soon as I hear a number way higher/lower than mine, I second guess myself too.) We usually just do dotplots, cycling through low, high, estimate and confidence, with discussion about trends too. It also gives us a chance to talk about reasonableness without having to call anyone out. So the kid that said 1000 was too high doesn't have to tell the class but we can still have a discussion in which her peers explain why that doesn't make sense. I haven't yet, but my colleague also has them do histograms and boxplots and they discuss different facets of the distribution which I am totally trying this year. We LOVE E180. Every Algebra 1 class does it daily here. Thank you!

    1. Hi Mary,

      Thanks for sharing the routines and learning around your uses of Estimation 180. These are exciting ways to infuse estimation into class routines and I'll have to remember these.

  5. Here's my wondering - isn't the purple bin superfluous here, if the blue bin maxes out at the number that is too high? Similar with the red bin having a max of the number that's too low.