Comments on: Test scores inch up in Oakland, Alameda County

By: Steven Weinberg

Steven Weinberg — Fri, 20 Aug 2010 05:05:40 +0000

Oakteach, I appreciate push back too, and I took another look at the statewide data and I have to revise my statements. The big increase in state-wide scores in ELA has left middle school ELA scores fairly consistent across the grade levels, but there is a big and consistent gap between 3rd, 4th, and 5th grade ELA scores, with scores going down each year in 3rd, up in 4th, and back down in 5th, but not a low as 3rd. It appears to me the scoring of the third grade test is harder and the scoring of the fourth grade test is easier. I think the same thing is happening with the sixth grade math test, but there is less evidence for that from the statewide data, since the scores don’t recover in 7th grade. (When I last checked Oakland data, I believe they did, but that was several years ago.)
Your theory about the effect of middle school transition on test scores may be correct. I would like to see figures comparing K-8 schools and K-6 schools with 6-8 schools for changes in matched 5th 6th scores.

By: oakteach

oakteach — Fri, 20 Aug 2010 01:08:15 +0000

Hot R – agreed. Bottom line: Oakland under performs. But the success stories prove that it’s possible. And the quicker we stop wasting our time with useless data points and start looking at the stuff that can really determine competency, the better.

Steven,
Appreciate the push back. But no one’s been able to get data (that I know of) that separates the social factors for the 5/6 drop (middle school transition). From a purely scientific standpoint, the difficulty between the two tests doesn’t match the drops seen. Granted, I have a skewed perspective: both of the schools I worked at, the 6th grade teachers got the HIGHEST matched scores of anyone.
As for the 7/8 drop, that is transparently due to the switch to Algebra (and the former general math test, which penalized each kid by a level).

Hence my second suggestion (and the only true way). Pre and post test for the course. Compare that data. The first suggestion was grounded only in peoples’ insatiable desire to use those tests as growth models. I know the cohort matched scores are flawed, but if people are going to continue to use CST’s to measure “school growth,” they might as well do it as close to the truth as possible.

By: Steven Weinberg

Steven Weinberg — Thu, 19 Aug 2010 22:26:53 +0000

Oakteach, while I agree with many of your points I want to point out that the dip in matched scores between fifth and sixth grades are largely a result of the difference in the difficulties of the tests at the two grade levels. The California Department of Education testing site specifically says that test scores cannot be used to compare students from grade level to grade level. I understand the urge to do so because it would be a better measure of school and teacher effort, but you have to measure each school or teachers’ results against the average for the district (I would prefer the state average, but it is not available, as far as I know.) As long as the CSTs have been given the district average for sixth graders has been lower than fifth, seventh higher than sixth, and eighth lower than seventh. If you don’t compensate for the differences in the tests, most seventh grade teachers end up looking good, and most sixth and eighth end up looking bad.

By: Hot R

Hot R — Thu, 19 Aug 2010 04:27:03 +0000

Look, I care about all those kids too. My heart breaks when I read these stories of low achievement. The schools seem in a death spiral, and even the responsible adults seem unable to stop them from spinning out of control. It is hard to make a teacher responsible for a student who is absent 2 days a week, or is new to the class as Maestra describes. Clearly, Oakteach knows what she is talking about, but I fear that no matter what method was used to measure achievement at the high school level, there would be none due to a variety of socio-economic issues which are exacerbated in the classroom and condemn all but a handful of kids to failure. This makes it even more remarkable that some kids can rise above it all and succeed.

However, I do not think the low test scores will have anything but a negative effect on the Oakland parcel tax vote.

By: J.R.

J.R. — Thu, 19 Aug 2010 03:09:07 +0000

Maestro,
That is a very valid point about transient children, it is becoming more common every day.

By: maestra

maestra — Thu, 19 Aug 2010 01:21:58 +0000

As a teacher in the flatlands for 8 years, a couple of thoughts:

-yes, there are bad teachers teaching and teachers who don’t understand basic math.
-yes, if those teachers are tenured, it is almost impossible to get rid of them
-no, the older kids don’t take the tests seriously. some of the younger kids take it so seriously that they have anxiety attacks and throw up.

there are many, many problems. One that I have not seen mentioned here (and this is not an excuse, just a fact that we should think about) is the transience at some schools. The 20 kids I started with were never ever the 20 I ended up with. There are so many homeless kids and kids with custody battles going on that they move in and out of schools. This is tragic for them for many reasons, as they get lost in the cracks, but as a purely practical test concern, it means that “my” test scores don’t accurately reflect who I have taught.

The most extreme year was when, in a class of 19 kids, my top four students moved in late April, right before the test, and I got four students who could not read, one of whom would not mark any answers on the test at all. That’s almost 25% of my class that went from grade level to far, far below. That may have been extreme, but it’s worth considering.

By: oakteach

oakteach — Wed, 18 Aug 2010 22:49:38 +0000

The “wrong test” is administered by the state. Not a site problem. Motivation is a site problem, and any school or admin worth their weight would acknowledge that.
ELA is sequential, Math is sequential for 6-7 and again Alg 1 – Alg 2. In between is Geometry and they aren’t tested for Pre-Calc and above. History and Science are NOT sequential from a growth standpoint – there are no overlapping standards from one test to the next (“sequential” in this context). So back to my original statement: CST’s can only be used to measure growth in those specific cases (and if looked at by strand, not just overall score), but are NOT a growth model in any other light. That in itself does not excuse low scores. But in the case of high schools, the only course taken in sequence for this purpose is ELA. Geometry is sandwiched between the two math courses. So if you wanted to see TRUE growth trends, you would have to look at 3 years of math data, not just 2 years to get a comparable measure to the middle school sequence.

There is nothing flippant about my comments. It’s an attempt (probably in vain) to get all the arm chair quarterbacks to focus on more accurate measures of school progress. Good test scores can hide bad schools, and vice versa. It’s not about excuses, it’s about digging below the surface to find what is truly effective (which the LA times started to do, albeit in a limited way).

The truth goes the other way as well: there are some middle schools on that list that showed “growth” in proficiency because the 6th graders are coming in higher (as a result of stellar progress at the respective elementary schools). In reality, those schools are actually losing ground: the proficiency “growth” of those schools is actually less than the increase caused by the incoming students. But those schools won’t be identified for any support because the data ignorant folk will focus on those schools below a cut mark (even if they are improving their students). So the schools that are doing nothing for kids slide by, while schools that may be growing low students get criticized. (Case study: Alliance and Elmhurst CP were both on the infamous state “list” for failing schools this past year. They had the largest growth I think of any middle schools. They are accelerating the students they get faster than most if not all, but are still identified as “failing” because people can’t read data.)

Unfortunately there are a finite number of resources and manpower in education as it stands, and pointing out the inaccuracies in data interpretations should not be seen as “flippant” if it helps identify the schools that are in TRUE need of help. I’m not arguing data means nothing. I’m arguing that the average bear does not take the time to look at what the numbers actually tell them.
If you want to know what a school is REALLY doing based on test scores, you need to do one of two things:
1) Look at the cohort matched scores for sequential courses to identify if real growth is taking place or kids are just coming in higher or lower.
2) Give the same test at the beginning of the course and the end of the course. Compare THAT growth across teachers and schools.

Education is a very esoteric field to start. There’s no checklist to success. And student performance data is one of the few concrete things we have to measure success, so it tends to be over valued and used in cases it shouldn’t. Data is a tool, and like any other tool it has to be used correctly to be effective. The way most lay people read data is equivalent to building a bookshelf with a toothbrush.

That’s my rant.

By: Chauncey

Chauncey — Wed, 18 Aug 2010 22:18:12 +0000

Let the excuses begin for the flatlands!!

By the Way, yes Arise and Aviation should be closed- (Arise is led by a fomer OUSD Principal right? A Bayces crony?

Overall-damn charters still whipped some tail in comparison. One day the board will either get it or they will dissappear. If I were a board member I would include charters under my watch.

It shows their pro union schools and no accountability to oakland”s overall community,

By: J.R.

J.R. — Wed, 18 Aug 2010 18:55:38 +0000

Lacy,
Bravo, and kudos to the MLK principal,staff, and parents. Good job!

By: Lacy Asbill

Lacy Asbill — Wed, 18 Aug 2010 18:52:06 +0000

I want to take a moment to focus on what IS working well, and offer congratulations to MLK Elementary and their 38% ELA gain! This just goes to show what a dynamic new leader and strong community partnerships can do to transform a school in a single year! Not only are their scores up, but parental involvement and support also went through the roof. Think about it–last year at this time, a new principal found her school on the possible closure list, and this year, they’re an improvement leader in the district! Congratulations to MLK!