Brain dump on Alaska’s AK STAR results, EED, ED (educational dysfunction), and Moneyball

The Alaska EED just released the 2023 AK Star test results. This is the big, annual “high stakes” test that Alaska uses these days. It was first used in 2022, and we’re just now finishing the testing here in April, 2024. I can’t understand why on earth it took a YEAR to release the results from last April’s testing. It’s an online test for 99% of the kids, so the scores should be available like within hours. I think us teachers first gained online access to the results in the summer of 2023, but only for our own students.

Summarizing the article, with LKSD’s results added:
• 32% of Alaska students tested were proficient in language arts. (3.7% in LKSD, 4.1% in 2022)
• 32% of Alaska students tested were proficient in math. (4.2% in LKSD, 2% in 2022)

As the article noted:

In January, the Alaska Board of Education approved lowering the standards of what’s considered proficient, citing the fact that Alaska had set the bar unusually high compared to other states.

This means that the apparent improvement in LKSD’s math proficiency might just be reflective of a lower standard. : – ( Or a combination of improvement and a lower standard. And it makes our slight dip in reading proficiency somehow even more horrifying. This all falls in line with what feels to some of us in the trenches as a tremendous lack of urgency in LKSD, from top to bottom. Everything is on fire, and it kind of feels like business as usual. Even on a micro-level at our village, things have been very rough in our school in terms of weak leadership that keeps turning over, and a total dearth of supervision & support from the top. And it is very frustrating when it feels like we care a LOT more about this than anyone at the district office. This extends to the state as a whole in a less urgent sense, with so little support for education reform and funding. Some conservatives want drastic reforms (aka “accountability”) without funding, and some in the education establishment just want funding without reform. Both positions are indefensible, but defended in the shrillest terms nonetheless.

_______________________________________________________

The one other thing I want to remember to mention in regard to last year’s STAR test results is a little wonky. OK, a lot wonky. The other big test we do in the fall and the winter (Aug & Jan) is the MAP test, in math, reading, and language usage. The MAP is made by the NWEA, and I believe the AK STAR is, too. But the AK Star has TWO tests (Reading, Math), and sometimes science (5th, 8th, 10th grades only). The test experience feels pretty similar, but the rules are more strict for the April AK STAR test because it is the one “high stakes” test that is going to be used as the official yardstick to measure progress. Anyway, the NWEA gives school districts access to MAP growth reports that we can use to check on the progress/performance of individual students, classrooms, grades at a school, entire school, entire district, entire state, etc. And the NWEA (or someone) translates the spring AK STAR test results into the familiar “RIT Scores” that kids get on the MAP tests. That means they look at their AK STAR results on 2 tests, and translate that somehow to an equivalent score on the MAPS 3 tests. The thing is, that translation, from the 2023 AK STAR test window, was crazy. It showed tons of our kids with gigantic leaps in their scores, in all 3 areas on the MAP growth reports. Like unprecedented, impossible leaps. Kids who have scored at the first percentile all of their lives, suddenly at the 16th percentile or something. And then in the fall when we did the usual MAP testing, all of their scores shot back down to normal. And I don’t mean a few of our kids, I mean like half or more of them. So in short, those translations were bassura!

For me, as a bit of a data nerd, this is very basic information that districts should be using for a hundred things, such as noting the rate at which students meet or exceed their projected growth on one of the 3 tests for a given teacher, and if that is a pattern, using that data to identify teachers who are having a big impact, for better or worse. If LKSD is doing that, I haven’t heard anything about it. But why not?? I did my own homemade version for myself and my own students, and my goal was always to beat the projected growth for my class as a whole, something that LKSD teachers obviously generally fail to do given our overall test scores. I’m proud that I accomplished that. And I’ve seen other teachers who consistently bring their kids way up, and these are kids who have had precious little growth for years, so it’s a big deal. And of course I’ve seen teachers who are not so great. I asked our district test coordinator one time for the average lexile (reading level) for LKSD’s kids, by grade. The NWEA MAP growth reports give an estimate of every kid’s lexile, based on their reading test results. So I have to think that the admins should be able to get an average for any classroom, school, or the whole district, by grade. But he said there was no such report, that he knew of. So either there is a report and he, the district test coordinator, is unaware of it, or he’s right, and the NWEA is failing to provide a super basic way of evaluating our students and a way to monitor progress/growth on a macro level. Either of those 2 scenarios is hard for me to believe, but I don’t really see a more likely third alternative.

SO, I guess the last 2 paragraphs were trying to express 2 concerns. The data we’re being given from the AK STAR testing seems suspect, at least when we try to tie it in to our tri-annual testing regime. Which prevents us from getting a fuller picture of how a single student or teacher is doing. And secondly, I have a sense of dread that school district leadership, and probably even the EED itself, seems to be doing very little with the mountains of data that are available to them. I’m a baseball fanatic, and sabrmetricians use baseball data to continually refine how the game is played and evaluated, and it has generally led to dramatically better teams. That was led from the top down (data guys convinced GMs and ownership), and it faced significant resistance from the old guard (scouts, old-school GMs who were not easily convinced, not into data, etc). Where is the equivalent data-driven revolution / reform (“moneyball”) for the educational establishment?