PARCC Comes Off the Rails

The PARCC cheerleaders' arguments for their beloved test are crumbling to dust:

Washington, DC (May 21, 2015) — The PARCC Governing Board, made up of the state education commissioners and superintendents, voted Wednesday to consolidate the two testing windows into one and to reduce total test time by about 90 minutes beginning in the 2015-16 school year.  The vote came in response to school district and teacher feedback during the first year of testing and a careful review of the test design.

The changes will improve and simplify test administration for schools, teachers and students, without diminishing the goal of the assessment—to ensure every student in every school is being taught what they need to know order to be successful in the next school year and, ultimately, in college or career. 

[Skipping Hanna Skandera's useless propaganda... -- JJ]

This year’s PARCC testing was done in two parts—the performance based testing conducted in early spring and the end-of-year testing conducted in late spring, closer to end of the school year. Five million students in 11 states and the District of Columbia completed the PARCC assessments this year.  

On May 20, 2015 the PARCC governing board voted to:

• Reduce the testing time for students by about 90 minutes overall (60 minutes in mathematics; 30 minutes in English language arts) and create more uniformity of test unit times.
   
• Consolidate the two testing windows in mathematics and English language arts/literacy (which includes reading and writing) into one.
  • ​The single testing window will simplify administration of the test for states and schools that experienced challenges with scheduling two testing windows.
  • The testing window will be up to 30 days and will extend from roughly the 75% mark to the 90% mark of the school year. Most schools will complete testing in one to two weeks during that window.
• ​Reduce the number of test units by two or three for all students.

Yes, the PARCC is so much better than those awful, old, state-level standardized tests -- so much better that the PARCC Board is changing it after only one year!

It seems like only yesterday -- no wait, it really was only yesterday! -- that the PARCC folks were telling us that "PARCC assessments are designed to provide parents and teachers with a far greater level of informative and useful data to help improve student instruction."

NJDOE's head PARCC cheerleader, Assistant Education Commissioner Bari Erlichson, told us this last year:

Along with the more detailed reports for parents, teachers for the first time will have access to a database showing the specific skill students were tested on in each question and how many of their students answered the question correctly, Erlichson said. [emphasis mine]

Well, that database is going to be a lot smaller now, isn't it? Are there enough questions left to break down each "specific skill"? Or are some of these skills being dropped? Who knows? All that matters, apparently, is that hopefully the "hysterical" moms who dislike the PARCC and what it is doing to their children's schools will stop complaining and Pearson won't lose another contract...

The entire point of a standardized test is that it is standardized. Every time you change it, you not only introduce error when making a comparison from year-to-year; you also tamper with the comparative validity of the test. Can we truly say that a test that is considerably shorter and only given in one window of time is measuring the same things as the earlier, longer version of the PARCC?

I can't believe that anyone would accept the argument that two different versions of this test should be used concurrently for high-stakes accountability measures, like school interventions or teacher evaluations. If any state agency plows ahead with accountability systems based on two versions of PARCC (and maybe the old state test PARCC replaced), they are going to face a lawsuit the first time a teacher is fired based on test results. And that teacher will win.

The PARCC madness of the past couple of years reminds me of the push for charter schools in suburban areas. Here in New Jersey, our DOE pulled back from its plan to bring "choice" to the 'burbs when parents -- who were a large part of Chris Christie's political base -- decided they didn't much care for boutique charters coming into their school districts and draining funds.

But their push still allowed for charter expansion in the cities, no matter the damage that caused. Districts like Newark and Camden are seeing their systems decimated; Hoboken is facing an existential threat. Charter cheerleaders lost the 'burbs, but they've made it impossible to seriously consider a pull-back on the sector as a whole (at least in the short term).

The PARCC has followed a parallel course: overreaching, the original plan was to have three (and possibly four) administrations of the PARCC throughout the year, and even expand testing to other curricular areas. Now the PARCC cheerleaders are pulling back...

Except they've managed to introduce the test into the high schools, a radical change. Even though it was never required by No Child Left Behind, we're now testing kids in math and language arts from Grade 3 to Grade 11. The PARCC people can afford to beat a strategic retreat, because they've managed to introduce themselves into schools where mandatory statewide testing was previously limited to Grade 9 biology and a high school exit exam. If you're playing the long game, pulling back a few minutes of the current tests is an acceptable, temporary loss.

As I've said before, limited testing has its place if used appropriately. But the high school exams have struck me as the worst part of all this testing madness. No one has ever shown that standardized end-of-course testing improves instruction in high school, and states that have it, like Tennessee, trail far behind states like New Jersey that had, until recently, eschewed the practice.

So while it's good to see the tests pulled back to once a year, we still have a way to go to rein in testing overkill. There's no need, for either accountability or instructional purposes, to test every kid in every year in multiple subjects. And it's indefensible to continue using the tests to compel school or teacher interventions (as opposed to informing those actions).

The PARCC pullback lays bare an uncomfortable truth for the reformy: these tests are not objective measures of student learning, let alone measures of teacher or school effectiveness. Standardized tests are political instruments, subject to the varying fancies of popular opinion. Using them for high-stakes purposes denies this now undeniable reality.

No, this is the test that came after we changed the last test that came after we changed the test before that...

ADDING: By the way, teens: during that 30-day window, you'd better not discuss the test! Especially on social media, the primary way you communicate!

Because what's more reasonable than asking a high school student not to talk about what's happening at school...