Monday, February 1, 2016

CCSS Flunks Complexity Test

The Winter 2016 issue of the AASA Journal of Scholarship and Practice includes an important piece of research by Dario Sforza, Eunyoung Kim, and Christopher Tienken, showing that when it comes to demanding complex thinking, the Common Core Standards are neither all that nor the bag of chips.

You may recognize Tienken's name-- the Seton Hall professor previously produced research showing that demographic data was sufficient to predict results on the Big Standardized Test. He's also featured in this video from 2014 that does a pretty good job of debunking the whole magical testing biz.

The researchers in this set out to test the oft-repeated claim that The Core replaces old lower order flat-brained standards with new requirements for lots of higher-order thinking. They did this by doing a content analysis of the standards themselves and doing the same analysis of New Jersey's pre-Core standards. They focused on 9-12 standards because they're more closely associated with the end result of education; I reckon it also allowed them to sidestep questions about developmental appropriateness.

The researchers used Webb's Depth of Knowledge framework to analyze standards, and to be honest and open here, I've met the Depth of Knowledge thing (twice, actually) and remain relatively unimpressed. But the DOK measures are widely loved and accepted by Common Coresters (I had my first DOK training from a Marzano-experienced pro from the Common Core Institute), so using DOK makes more sense than using some other measure that would allow Core fans to come back with, "Well, you just didn't use the right thing to measure stuff."

DOK divides everything up into four levels of complexity, and while there's a temptation to equate complexity and difficulty, they don't necessarily go together. ("Compare and contrast the Cat in the Hat and the Grinch" is complex but not difficult, while "Find all the references to sex in Joyce's Ulysses" is difficult but not complex.) The DOK levels, as I learned them, are

Level 1: Recall
Level 2: Use a skill
Level 3: Build an argument. Strategic thinking. Give evidence.
Level 4: Connect multiple dots to create a bigger picture.

Frankly, my experience is that the harder you look at DOK, the fuzzier it gets. But generally 3 and 4 are your higher order thinking levels.

The article is for a scholarly research journal, so there is a section about How We Got Here (mainstream started clamoring for students graduating with higher order smarterness skills so that we would not be conquered by Estonia). There's also a highly detailed explanation of methodology; all I'm going to say about that is that it looks solid to me. If you don't want to take my word for it, here's the link again-- go knock yourself out.

But the bottom line?

In the ELA standards, the complexity level is low. 72% of the ELA standards were rated as Level 1 or 2. That would include such classic low-level standards like "By the end of Grade 9, read and comprehend literature, including stories, dramas, and poems, in the grades 9–10 text complexity band proficiently, with scaffolding as needed at the high end of the range." Which is pretty clearly a call for straight-up comprehension and nothing else.

Level 3 was 26% of the standards. Level 4 was  a whopping 2%, and examples of that include CCSS's notoriously vague research project standard:

Conduct short as well as more sustained research projects to answer a question (including a self-generated question) or solve a problem; narrow or broaden the inquiry when appropriate; synthesize multiple sources on the subject, demonstrating understanding of the subject under investigation

Also known as "one of those standards describing practices already followed by every competent English teacher in the country."

Math was even worse, with Level 1 and 2 accounting for a whopping 90% of the standards.

So if you want to argue that the standards are chock full of higher order thinkiness, you appear to have no legs upon which to perform your standardized happy dance.

But, hey. Maybe the pre-Core NJ standards were even worse, and CCSS, no matter how lame, are still a step up.

Sorry, no. Still no legs for you.

NJ ELA standards worked out as 66% Level 1 and 2, Level 3 with a 33%, and Level 4 a big 5%.

NJ math standards? Level 1 and 2 are 62% (and only 8% of that was Level 1). Level 3 was 28%, and Level 4 was 10%.

The researchers have arranged their data into a variety of charts and graphs, but no matter how you slice it, the Common Core pie has less high order filling than NJ's old standards. The bottom line here is that when Core fans talk about all the higher order thinking the Core has ushered into the classroom, they are wrong.

1 comment: