Sunday, October 24, 2010

Followup Post: Common Formative Assessments

So we did it. Was it worth it?

Recap: Students were given a benchmark. The benchmark was broken into 4 topics—Matter, States of Matter, Atoms, and Chemical and Physical Changes. The following week they spent Monday through Thursday in leveled classes. Monday was Matter, Tuesday was States of Matter, etc. They retested on Friday.

For all the classes, 294 kids, 26% were proficient in the first go and 43% in the second. 35% of the kids were in the lowest range (scoring 35% or less) in the first round and that went down to 18% in the second.

On the downside, only 2/3 of the kids made any growth which means that 1/3 of them were flat or went backwards.

I made 60% proficient and 75% advanced for this benchmark. It was somewhat arbitrary but I figured all students should, at minimum, know the first three questions for each topic.

All students who scored less than 60% have another level of intervention. For students scoring less than 50% on their benchmark they signed up for an after school or lunch tutorial this week based on their lowest topic score. They also signed up for an additional tutorial for any topic that they got 0 or 1 correct.

I'm signing them up on a week by week basis. They can exit out of tutorial sessions with a half-sheet quiz taken sometime before the next session. It's just two or three questions that are representative of the middle level questions. They'll need to get 100% to exit.

Students who scored 50% or 55% were given the option to instead do the relevant section from a workbook. They showed me they had it completed on Friday and took the exit quiz. Maybe 70% of those students took that option.

This week I had Lego League meetings twice and a staff meeting so I met with them twice after school and every day at lunch. It's been fast. 15 minutes at lunch and maybe 20 minutes after school (larger crowd). I'd say there's a 90% attendance rate with a few extra kids that just come with their friends. Talking with the absentees I'm fairly sure they're just honestly forgetting (especially the kids who sign up for lunch but I don't see them normally until after lunch) so I need to start sending notes out 4th period and before school gets out.

I have no idea why they're coming. I've made it pretty clear that the benchmark doesn't directly affect their grade, although if they're behind right now it will get more and more difficult to keep up. That's not to say they're excited about coming, but I think we often underestimate how much our students really do want to learn.

Other observations:
  1. The test was noisier than I would have liked but I'm not too worried about it. I don't want to fall into the "must have more data" rabbit hole. It's super low stakes and the worst that happens is a student is placed in the wrong level for a single period or that they need to come for extra tutoring. I can live with that.

  2. Clearly the skill-based lower level stuff had more bang for the buck. The earlier questions, which mainly were vocab and simple skills, made huge growth. The latter questions which required more conceptual understanding barely budged. It makes sense given the time constraints but I'm not quite sure what the best approach is here. Do we acknowledge the limitations and just target high growth skill-based stuff? Or do we take another stab at higher-level concepts and hoping it clicks for whatever reason? My natural inclinations says to always go higher-level but then again, perhaps needing a stronger base is what's holding back the higher-level stuff. I don't know yet.

  3. The periods where we were able to have 3 teachers for 2 classes worth of students made a HUGE difference. Probably obvious, but the difference was big at every level of student for every topic. I'm putting out feelers to non-science teachers to come the next time we do this. I'll help in your class during my prep sometime in exchange for your help that week. Nobody has committed yet but a few have sounded interested.

  4. Greg suggested using 1, 2, 3, 4 for the MC answers instead of A, B, C, D. This was super clutch. Data entry was blazing and it took me less than 30 minutes to enter in the scores for all the kids. Another teacher has this 10-key USB keyboard he uses with his laptop and says he was able to get it all done in less than 20 minutes.

  5. I did pre/post class graphs and they're on the bulletin board. I don't know how I feel about it.

  6. To sort classes, we wrote all our students names on index cards with topic scores down the right side. We just sorted them into piles each day.

  7. Switch up what teacher takes the "high" kids. Students started thinking they were in the dumb class when they saw me every day. There was a fair amount of movement between levels though. I probably saw about 80% of the 8th graders at one point or another. I loved meeting new students. For me, that was one of the highlights.

  8. Our secretaries were not fans because our attendance kept getting messed up. By the end we had a system but the first two days I marked a few kids absent who just wandered off into the wrong group.

  9. We didn't get to collaborate beforehand at all. I was home with a sick kid on Thursday and Friday everyone is outtie as soon as the bell rings. I ended up doing the first day rosters myself over the weekend and emailing them. I'm pretty sure one of the teachers didn't think I was serious about this whole plan so it caught him off guard Monday morning. The first day we had some overlap in what had already been done in other classes. We sorted it out but next time I'm going to have us really talk it out.

  10. Letting them sign up for their own tutoring session has been helpful. They don't mind coming with friends and they've been good about reminding each other.

  11. I was the only one who looked at everyone's data. I'm not sure how to introduce the next stage of having us compare and figure out how to teach certain parts better. Seems touchy. Also, both teachers are probationary so I don't know how much me being their evaluator would be in the back of their heads.

  12. Certain students are kept in different classes on purpose (usually rival gang reasons) so we were worried about mixing them up. We mainly lucked out. We cheated and put one kid in the wrong level. He ended up being suspended for the week anyway but it's a bridge we'll have to cross next time.   

A story I have to share:
One of the frequent flyers to the office was 1 of 2 students in the school to score 95%. He first claimed he copied. When it was pointed out he couldn't copy and do better than the person he copied from he claimed to have just guessed. We told him it was nearly impossible for him to guess and get 19 out of 20 right. Finally, he told us to be quiet about it because he didn't want anyone to know he was smart. He got sent to the office later in the day and mentioned his high score to the vice-principal. Then he explained what we were learning about. Then he told her not to tell anyone. I've seen him twice since then and both times he's managed to mention his score.

I'm planning to keep this going for the year. The part-time teacher was really enthusiastic about her results (from 25% to 55% and the boy mentioned above) and so she's on board. The other teacher has been out sick this week so I didn't get a chance to talk with him about it. I'm not sure about his buy-in and, for better or worse, he's not the type to openly fight it.

There are some details we're going to change for next time but it's mainly procedural. We like the general format. The attendance issue was a big problem and we're going to try to keep better track of how we teach certain topics.

I'm also not sure what to do about the pre/post tests. I just gave the same one twice but gave zero feedback on their scores the first time. The dummy/smart kid class thing is something I've really tried to avoid so I was hoping by not letting them know their results they wouldn't be so focused on which class they were in. The SBG monster inside me is yelling to give them immediate feedback so they know what they need to actually focus on during that week. We framed each period as, "You're here because you probably need to work on...." or "You probably missed questions about...." but I'm not sure if that's enough. This is something I'll need to experiment with.

Overall I'd say it was worth the effort. Generally the results were positive and at least one teacher is enthusiastic about it. I tried some feedback forms from the kids but they were less than helpful. Their satisfaction basically tracked their improvement from pre/post. I should have seen that one coming and I should have given the feedback form to them before they got their results back.

If you've made it this far and have any advice or questions - comment, twitter, email.


  1. Wow! That was quite a week. It seems like it was a lot of setup. Is the idea that this might eventually be done every week? Think the amount of work will go down?
    Also you appear to be the driver of this bus? Is that sustainable? Will the other teachers drive too?

    Nice work. It interesting to read this because I can't see this happening for me anytime remotely soon.

  2. Not every week. It'd be impossible to create any sort of flow if we constantly rotated the classes. But the idea is every 6-8 weeks or so for specifically targeted help.

    It wasn't as much work as it seems. We worked through our preps but you'd end up with 20 kids instead of 30 so each period was less effort.

    Plus I like to think of it as saving work. Something like 30% of the kids didn't know what a proton was. Now we're on periodic table and you've got to know what a proton is. By taking the time to go back with those 30%, we're saving time right now and (hopefully) guaranteeing a certain baseline of knowledge.

    I'm the driver but only really because everyone else is new. I have no idea what's going to happen next year. Although I'm not a huge test score fan, truthfully, that's going to dictate things. If the department has big growth on our state test, we'll have some momentum. Otherwise, it'll be hard for me to justify.

  3. Hey Jason,

    This is impressive work and I think you and your colleagues are working in an incredibly responsible way. The intervention practices that you're piecing together are targeted, systematic and leveled....characteristics of good interventions.

    My concern for you is the same concern that I've had in my own building: No where in your interventions were professionals beyond the classroom involved in your efforts.

    The challenge there is the work you are doing is pretty darn complex and overwhelming. It's too much to expect a team of teachers to manage on their own. Systems of intervention that are left in the hands of classroom teachers alone are destined to fail simply because teachers have too much on their plates already.

    Now, if you're anything like me, getting outsiders involved in intervention systems is hard because:

    1. There's no time for shared planning with people beyond the classroom.

    2. People beyond the classroom have other responsibilities during the time periods that you are intervening.

    3. People beyond the classroom don't see themselves as responsible for being a part of classroom level interventions.

    4. There's just no one beyond the classroom to get involved in interventions.

    All of those barriers have gotten in the way of our systematic intervention efforts in the past few years and it's frustrating times ten.

    We've got to get to the point in our schools where intervening is the job of the entire faculty, not just the classroom teacher. Otherwise, interventions will never succeed.

    Any of this make sense?

  4. Thanks Bill for the feedback. A lot to digest. What you describe echoes my fears. The gap between what we have now and what you describe is huge. Do you have any suggestions for first/intermediate steps and some quick wins?

  5. As a student myself, I would have to say that I do want to learn, and that this desire is reflected in my classmates. However, your idea about underestimating students' will to learn is intriguing. Where do you think that this misconception comes from?