Academic Growth Over Time (aka Value Added)

Kathy is introducing the topic and it is obviously a sensitive one, so there are many disclaimers and caveats including the presentation being done by district personnel. The LA Times is gearing up to release data on elementary schools and then middle school math and English. The incoming superintendent and the board is very interested in value-added so we need to better understand the enemy.  On a side note, rebranding value added as Academic Growth Over Time reminds of creationism being rebranded as intelligent design.

This is meant to be an overview of the way it is calculated and not an in depth presentation.  Students will be evaluated in terms of absolute achievement and growth compared to prior performance and predicted growth. We are being presented with some rather simplistic models of oak trees and gardeners to compare achievement with simple growth to prediction and growth over time which is a weighted growth model and supposedly more fair to the poor beleaguered gardeners. Right away there are worried looks, sighs and verbalized concerns. The AGT model takes into consideration prior CST scores, income status, grade level, ethnicity, and such.  In a future model, attendance and team-teaching will be taken into consideration (students with absences beyond a threshold will not be part of the consideration). For this year, students will be accredited to the teacher in whose class they were on norm day, even if they haven't been with that teacher since that day.  Only students with at least two years of prior testing will be included. Students will only be compared with students of the same background/ status.  I have particular concerns about this because if I do a good job and a kid redesignates to Resource than all the good teaching that I have done doesn't get credited to me. Behavior is also not a consideration - D.O. is concerned that when a difficult student is moved into the class mid-year, the scores of all the students in the class will be negatively impacted and there is nothing you can do.  The district says that they didn't include this because the placement of behavior problems is something in control of the teacher and the school and they are only going to weight things that are out of the control of the school.
There is no benefit to you as a teacher  if you improve the attendance or behavior of a student, even if those are equally valid measures of the "goodness" of a teacher.  Currently value added doesn't affect your performance review.
Supposedly, it compares the growth of a student to students all over LA Unified that have similar attributes. Doing this for ELA grades 3-9 and Math grades 3-7. Separation of algebra and general math students in grade 8. Doesn't know how they handle it when teachers move from grade level to grade level - doesn't think it matters what you teach. Unsure of the error analysis on this process. Minimum number of students you must have in order to be included is 11 - obviously a bigger sample size will lead to a smaller confidence interval on the results.  Oooh, in case numbers are too hard for you to understand there are helpfully color-coded balloons to represent your worth as a teacher.  Blue and green are good, gray is neutral and yellow and red are bad.  You are rated on a scale from 1-5 with 3 being average for the district.

There is both a measure for the previous year and a three-year average at the school and teacher level. Instead of paying money to do these calculations, couldn't we spend the money on lowering class size which has PROVEN benefits? This is fishy and it smells like union-busting? Is there any data that shows that knowing this stuff benefits kids at all? And if any of this starts effecting our pay, why wouldn't this turn into an arms race of teachers cheating to improve their value?

Apparently there is some accounting for a ceiling effect and regression of the mean.  So if you teach very high achieving students, it is assumed that the only place to go is down, and it considered high value-added if your kids don't go down as much as other high achieving kids. Data is available for whole school and then broken down by sub-group (EL, SPED, SED, ethnicity).

Now we get to see our data! It's a sneak peak of what the paper will be publishing on April 12th. Overall, in ELA we are gray, slightly below the district average and in the three year, we are below average for our Basic and special education students. By grade level, our seventh graders are well above grade level, while sixth and eighth grade students are slightly below.  In math, we are significantly below district average across grade level at the one and three year evaluations. Students with the same name may accidentally get swapped. If you want to be a teacher who has value-added as part of your STULL next year, you can sign up.  The school would get money, but not the teacher.  Apparently we are doing this because there is a perception that LAUSD isn't getting the job done, but I must ask myself, wouldn't hiring more teachers and lowering class size draw students back while improving educational outcomes? With RESEARCH to back it up. Isn't raising class size contrary to our stated goals?