## Monday, June 06, 2011

### SBG Analysis

As I mentioned in earlier posts, I am going to revise my SBG for next year. (Post #1 and Post #2)

As a part of my revisions, I looked at the data from my Re-Assessment notebook for the year. Any time a student wanted to re-assess, I wrote down their name, which learning targets they wanted to re-assess, and when they were coming to do it. For the most part, it is accurate (although I am certain I missed a couple).  I should mention before getting to the data that we had eight snow days and the 3rd 9 weeks was the stretch of the year where we had a long streak of weeks that were not 5 days (either planned or unplanned). Here's the data by class with comments following:

Math 1 (lower level freshmen, 40 students at the end of the year, 71 LTs assessed)

1st 9 Weeks:
27 LTs reassessed in 7 attempts = 3.86 LTs/attempt
4 unique students reassessed = 6.75 LTs per student

2nd 9 Weeks:
16 LTs reassessed in 4 attempts = 4 LTs/attempt
3 unique students reassessed = 5.33 LTs per student
(2 new students)

3rd 9 Weeks:
12 LTs reassessed in 3 attempts = 4 LTs/attempt
3 unique students reassessed = 4 LTs per student
(1 new student)

4th 9 Weeks:
14 LTs reassessed in 4 attempts = 3.5 LTs/attempt
3 unique students reassessed = 4.7 per student
(1 new student)

Analysis - Not many students in this class even opted to reassess. Several students reassessed once and that was it. I had one student who reassessed several times over the course of the year (he also happens to be on an IEP) and you could tell he put the time into learning the LT before reassessing. For that matter, the students who did reassess more than once truly wanted to have better understanding and took time to learn the material before reassessing. Those kids seemed to "get" that this was about learning the material, not just bringing up their grades better than any of my other classes.

Algebra 2 (primarily sophomores, 41 students at the end of the year, 52 LTs assessed)

1st 9 Weeks:
41 LTs reassessed in 14 attempts = 2.92 LTs/attempt
12 unique students reassessed = 3.42 LTs per student

2nd 9 Weeks:
45 LTs reassessed in 19 attempts = 2.37 LTs/attempt
8 unique students reassessed = 5.63 LTs per student
(3 new students)

3rd 9 Weeks:
25 LTs reassessed in 11 attempts = 2.27 LTs/attempt
10 unique students reassessed = 2.5 LTs per student
(4 new students)

4th 9 Weeks:
63 LTs reassessed in 21 attempts = 3 LTs/attempt
14 unique students reassessed = 4.5 per student
(5 new students)

First off, I should say that these students are average to lower ability students. In Ohio, the current freshmen will have to graduate with 4 credits of math, one of which must be Algebra 2 or its equivalent. We do not go anywhere near into depth or number as I think most of you do in Algebra 2. I teach in a primarily blue collar, lower socioeconomic community (we have about 50% on free and reduced lunch) and eduation is not highly valued in many homes.

As I talked to parents throughout the year and advised them how I do my grading, I saw more new students choosing to reassess. Some would get help from me, others from other students. I'm pretty sure most students in these classess were chasing grades more than knowledge. There were a couple who wanted to show me they really knew the material, but most of them wanted to improve their grade.

Advanced Algebra 2 (primarily advanced freshmen, 23 students at the end of the year, 71 LTs assessed)

1st 9 Weeks:
69 LTs reassessed in 23 attempts = 3 LTs/attempt
12 unique students reassessed = 5.75 LTs per student

2nd 9 Weeks:
45 LTs reassessed in 16 attempts = 2.81 LTs/attempt
10 unique students reassessed = 4.5 LTs per student
(1 new student)

3rd 9 Weeks:
64 LTs reassessed in 28 attempts = 2.29 LTs/attempt
13 unique students reassessed = 4.92 LTs per student
(2 new students)

4th 9 Weeks:
141 LTs reassessed in 50 attempts = 2.82 LTs/attempt
17 unique students reassessed = 8.29 per student
(3 new students)

Some of these kids are chasing perfection - some for the sake of having 100%, some for having mastered the material. Some of these students wanted to show me they really knew the material and said so - they couldn't believe they made such "dumb" mistakes. The 3rd 9 weeks really screwed us up here - we only covered 12 LTs and we covered 25 LTs in the 4th 9 weeks, which accounts somewhat for the huge disparity in reassessments.

In this class, I had two students who were consistently re-assessing and would re-assess 5 or 6 LTs at a time. Both would get help and wouldn't ask to re-assess until a week or so after the initial assessment. I think one of them was taking the practice fairly seriously, the other wasn't as much, especially at the end. I suspect they were both chasing points to an extent and the one who wasn't taking the preparation seriously definitely was point chasing at the end. For the most part, students in this class wanted to learn the material to the best of their ability and this was most in evidence to me at the end of the year. The last assessment I gave them (on logarithms and exponentials, which is about the toughest material we have done all year), they did well. They were working until the end to learn the material.

Calculus (upper level seniors, 13 students at the end of the year, 66 LTs assessed)

1st 9 Weeks:
57 LTs reassessed in 13 attempts = 4.38 LTs/attempt
8 unique students reassessed = 7.13 LTs per student

2nd 9 Weeks:
60 LTs reassessed in 14 attempts = 4.29 LTs/attempt
9 unique students reassessed = 6.67 LTs per student
(2 new students)

3rd 9 Weeks:
28 LTs reassessed in 5 attempts = 5.6 LTs/attempt
5 unique students reassessed = 5.6 LTs per student

4th 9 Weeks:
42 LTs reassessed in 10 attempts = 4.2 LTs/attempt
7 unique students reassessed = 6 per student

This was definitely a point chasing class. I had a student in particular who would ask to re-assess several LTs from the previous assessment right before the next one and would do okay on the re-assessment and poorly on the regular assessment. This same student would consistently ask to re-assess 6-8 LTs at a time and would come back with only having attempted 2 or 3 of them. This was also my lowest-ability upper-level class ever.

However, there were a couple of students who really took the re-assessment process seriously. One in particular stands out in my mind because he would ask to get some extra help a week or so after the first assessment and would come in with problems worked out (whether they were from the homework or the test) and had questions ready for me every time. He tried to get as far as he could on his own before coming to me for help. This young man will go far with that preparation and attitude.

So where does this put me for next year?
Well, first and foremost, I do think I am going to limit how many LTs a student can re-assess at once. Right now, gut reaction is 3 at a time, max, although I may adjust that down to 2. Someone had suggested to have a fixed number of re-assessments for the grading period or year. I'm not sure I want to go down that road. I like limiting a student to 3 attempts to re-assess - I only had 1 or 2 students who went to 3 attempts and they both mastered on that third attempt. However, if I go to the Dan Meyer/Kate Nowak system of assessing a LT twice and a student has to earn the "4" or "5" twice to earn mastery.

Secondly, as I look back at who was re-assessing and when, I definitely want to have some proof of practice before allowing a re-assessment. I haven't totally decided what that is going to be yet. One thought I am considering is submission of error analysis and correction of a missed test or re-assessment question coupled with a brief conference with the student as I review said analysis and correction. Another thought is to see the original practice problems completed as I had stated in my original SBG explanation. A third possibility is combining both of those ideas.

I still think I am going to work on assessing 3 or 6 LTs at a time on a rolling basis as I've mentioned earlier (1-2-3, then 1-2-3-4-5-6, then 4-5-6-7-8-9, etc.) and students have to have mastered a problem on a particular LT twice to earn mastery. If you are new to this see Dan Meyer or Kate Nowak's blogs for more info.

As always - feel free to offer comments and suggestions (as well as other stuff you see in my data that maybe I didn't point out). I appreciate you taking the time to read!