When Pearson Monitors Students, They Prove the Inferiority of Their Product

UPDATE I: Watching Hills Regional High School has released a statement about the incident below.

Full disclosure: my K-8 district "feeds" into WHRHS, but I am not employed there as it is a separate district. I do not know who the student is but it is possible he is a former student.

UPDATE II: Looks like Bob's post is back up for now, but it's loading slowly -- possibly because this is such a big story and he's getting tons of traffic.

Also: here's a report about what Pearson -- again, a foreign corporation -- expects from students regarding test security and social media. But I can't find any equivalent information at the PARCC website or from NJDOE. Were parents and students expected to seek this out themselves?



Most of you probably know by now that Bob Braun, veteran education journalist and a personal friend, published a blockbuster of a story yesterday: Pearson Education, Inc., creator of the PARCC standardized test, has been monitoring students' social media use and, in at least one case, reported what they considered to be a violation of their test security.

Even worse: Bob's site has been under a "denial of service" attack since shortly after he published the report. As of this morning, I'm still not able to access Bob's story at his blog, but not to worry: Bob published his story on Facebook, where it appears to be immune from DOS attacks. Here's an excerpt:

Pearson, the multinational testing and publishing company, is spying on the social media posts of students--including those from New Jersey--while the children are taking their PARCC, statewide tests, this site has learned exclusively. The state education department is cooperating with this spying and has asked at least one school district to discipline students who may have said something inappropriate about the tests. This website discovered the unauthorized and hidden spying thanks to educators who informed it of the practice--a practice happening throughout the state and apparently throughout the country.

The spying--or "monitoring," to use Pearson's word--was confirmed at one school district--the Watchung Hills Regional High School district in Warren by its superintendent, Elizabeth Jewett. Jewett sent out an e-mail--posted here-- to her colleagues expressing concern about the unauthorized spying on students.

She said parents are upset and added that she thought Pearson's behavior would contribute to the growing "opt out" movement. So far, thousands of parents have kept their children away from the tests--and one of the reasons is the fear that Pearson might abuse its access to student data, something it has denied it would do.

In her email, Jewett said the district's testing coordinator received a late night call from the state education department saying that Pearson had "initiated a Priority 1 Alert for an item breach within our school."

The unnamed state education department employee contended a student took a picture of a test item and tweeted it. That was not true. It turned out the student had posted--at 3:18 pm, well after testing was over--a tweet about one of the items with no picture. Jewett does not say the student revealed a question. There is no evidence of any attempt at cheating.

Jewett continues: "The student deleted the tweet and we spoke with the parent--who was obviously highly concerned as to her child's tweets being monitored by the DOE (state education department).

"The DOE informed us that Pearson is monitoring all social media during the PARCC testing."

Jewett continued: "I have to say that I find that a bit disturbing--and if our parents were concerned before about a conspiracy with all of the student data, I am sure I will be receiving more letters of refusal once this gets out."

The school superintendent also expressed concern about "the fact that the DOE wanted us to also issue discipline to the student." Clearly, if Pearson insists on claiming test security as a justification for its spying on young people, that reasoning is vitiated by its cooperation with the state education department in trying to punish students who are merely expressing their First Amendment right to comment on the tests. [emphasis mine]

Here's a copy of the email Bob published:

I normally wouldn't put this on my site, or take such a long excerpt of Bob's work. But given the fact his blog is under a DOS attack, and given the gravity of this story, I believe it's important to get this information out by as many channels as possible.

Obviously, we have no idea who is launching the attack on Bob's site. But we all need to demand that law enforcement conduct an investigation immediately and prosecute the perpetrator to the fullest extent of the law. This is a clear attempt to silence a veteran journalist who is reporting on a very important issue.

Think about what has happened here: at least one student exercised his right to free expression about important social issues -- education and testing -- after the administration of his test. But a private, foreign corporation decided their property rights trump his First Amendment rights, and they have used their relationship with a governmental agency to demand he be punished.

By all appearances, there was no attempt by the NJDOE to conduct an investigation as to what exactly the student tweeted, because, according to the WHRHS superintendent, the report the student tweeted a picture of the test was incorrect. NJDOE apparently just assumed Pearson's report of the child's tweet was accurate; I guess we all know now who's holding the leash down in Trenton...

I can hear the objections now: "You're not allowed to talk about any tests until everyone takes them! This is no different than if a kid discussed an AP test question and had his score canceled, or if he talked about a local chemistry test he took during period 2 with a kid who was going to take the same test in period 5!"

Actually, it is different. Quite different.

Regarding AP and SAT and ACT and other exams: these are all voluntary. You don't have to take an AP exam; in fact, students can take AP courses without taking the exam, or take the exam without taking a course. No student is forced to take the SAT; if she chooses to do so, she then enters into a contract and agrees to the terms and conditions of the test.

It is unreasonable to force a child to take a test and then demand she remain silent about its contents in perpetuity. Pearson and the NJDOE, however, appear to be demanding exactly that. By insisting that all students must take the PARCC, NJDOE is, in effect, forcing students to give up their rights to free expression with no provision to opt out of the test and retain those rights.

I've been looking around the PARCC website this morning for a clear set of guidelines as to what the PARCC consortium expects from students regarding the public discussion of test items; so far, I can't find it. I certainly have never seen any indication parents have to agree to the security conditions imposed by Pearson on the test. So what did Pearson, PARCC, and NJDOE do to inform students they can't discuss test times after the administration of the exam? Where is any guidance for students, or their parents, as to what they were getting themselves into when they sat down to take the PARCC?

The comparison to local exams is, to my mind, a more critical question. Yes, every school and every district maintains and enforces a code of academic conduct, and that code would obviously preclude a student from revealing test items until every student had taken a test.

But local exams are structured so that the constraints on students discussing tests are reasonable. A teacher doesn't give two different sections the same exam two days apart; if she did, she'd be hauled into the principal's office and dressed down for not doing her job. A good teacher and a good school do not entrap their students into cheating by setting unreasonable expectations for assessment security.

But even more than that: a good teacher gives assessments that are largely cheat-proof. So if the PARCC people really think their exam can be gamed by students over social media, they are admitting they have created an inferior product.

Stephen Danley of Rutgers-Camden, another great New Jersey blogger, puts it very well (on Twitter, of all places):

.@BobBraunsLedger story has me thinking abt what is good assessment. I'm proud when my students talk abt my tests. Means my material matters

That is exactly right. Students should be talking about their tests. They should be talking about what they got right and what they got wrong. They should be talking about whether the test was "fair," or what we in edu-nerd world call "valid and reliable."

All learning is socially constructed, and these days children live a large part of their lives on social media. If a test means anything to a student, he will likely discuss it on Twitter or Facebook (do kids use that anymore?) or Reddit or through texts. It's simply unreasonable to think a kid will not tweet out his thoughts about a test, especially when they have such high-stakes attached to them.

Further: if the assessment is any good, and is really measuring higher-order thinking, it likely can't be gamed. It's easy to cheat on a multiple choice exam; it's much harder to cheat on a chemistry lab. And it's nearly impossible to cheat on a choir concert, or a personal response to a novel, or number line manipulative.

I know the PARCC cheerleaders have told us over and over that their test is "better." They seem to think that writing short answers explaining how you solved 26 divided by 5 that are then graded by low-wage non-educators is an acceptable substitute for a well-trained teacher with enough time and resources to properly assess her students. It isn't, and this incident shows us why.

Again: I am all for the appropriate use of standardized tests, employing smart sampling strategies, as accountability measures. There is a good case to be made that we will never get students -- particularly students in economic disadvantage -- the adequate resources necessary for their educations without these measures.

But if the PARCC is so vulnerable that a tweet by a student after the test compromises the entire exam, it must be useless -- particularly as a measure of student learning.*

Pearson's product is inferior to the best assessment system we have: well-trained teachers with the freedom to do their jobs who are held accountable. We don't worry about tweets after exams, because we don't have to.

A far superior method of assessment than anything made by Pearson.

* Edited for clarity. We have no idea what exactly this student tweeted out, and I don't want to in any way imply it was improper. We just don't know right now if it was.