Last Week Tonight with John Oliver: Standardized Testing (HBO)

Standardized tests.

The fastest way to terrify any child with
five letters

outside of just whispering the word,
clown.

It's currently testing season all over
the country, and with that comes the

usual flood of anxiety and school-produced
videos designed to get kids in the mood.

♪ Can't read my, can't read my, no you
can't read my test-taker face ♪

♪ Let's show them we can do it! ♪

♪ Hey, we can do it! And, this is
crazy! ♪

♪ But, I'm determined. So test me,
maybe? ♪

♪ What does the test say? ♪

♪ Get your number two pencils out, your
number two pencils out! ♪

♪ Get your number two pencils out, your
number two pencils out! ♪

♪ What the test say? ♪

Get your number two pencils out!
Get your number two pencils out!

Look, standardized tests look like amazing
fun.

I wish I could take one right now!

Bring me a pencil, a number two please!

But, it gets better because one elementary
school in Texas

even held a test-themed pep rally featuring
a monkey mascot.

♪ Ooh ahh ahh! Time to get funky!
Ooh ahh ahh! Here comes the monkey! ♪

♪ Ooh ahh ahh! Time to get funky!
Oonky, honkey, funky monkey! ♪

Look, let's all agree there is no scenario
in which the words

'here comes the monkey' can fail to pump
you up.

Just imagine right now, I was your surgeon
and I said I'm about to put you under.

There's about a 20% chance of survival, and
I have four important words for you:

Here comes the monkey!

You're gonna be looking forward to that
operation!

It's gonna be a fun time!

You see? You love it!

The point is -- you've proved my point! --
those videos and monkey mascots

would have you think that testing is
amazing, which is why this spate

of recent new stories has been so
suprising.

The Lower Hudson Valley. Many districts reported that more than 25% of their students opted out.

More than 1,700 elementary, middle and high
school students

opted out of taking the PARCC test.

- There were five kids that I was with.
- They took the test?

Yeah.

Like almost the entire auditorium was filled
with kids that didn't take it.

Not a single junior showed up to take the
Common Core Smarter Balanced test this week.

Wow, the entire class boycotted the test.

The only other thing an entire class of juniors
has ever managed to agree on

is that the Scarlet Letter could be told much
simpler with emojis.

Yeah, we get it. We get it!

Red lady. Finger finger. Devil. Baby.
We've all read the book. It's a good story.

But, look, is it any wonder that American
students are sick of tests?

Between benchmark, diagnostic, pre- and
mock tests, they take a lot of them.

Students are taking between 10 and 20
standardized tests depending on the grade.

A total average of 113 different ones by
graduation.

113 is a lot of tests. It's approaching the
amount that you'd ask your doctor for

the morning after you woke up from a one-
night stand with Colin Farrell.

Just give me all of them twice.

And, this amount of testing can take a
toll.

Teachers have reported kids throwing up,
kids crying, especially the younger ones.

And, it's the pressure.

That's true. In fact, this happens so much
that official instructions for test administrators

specify what to do if a student 'vomits
on his or her test booklet'.

Something is wrong with our system when we
just assume a certain number of kids will vomit.

Tests are supposed to assessment of skills, not
a rap battle on 8 Mile Road.

Oh, Eminiem, why did your mom make
you spaghetti?

She knew tonight was rap battle night.

So, how did we get here?

Well, the explosion of testing can be traced
back to the '90s when you probably

remember stories like these about the
state of public education.

When 40 nations recently took the International
Math and Science Test,

American students scored near the bottom.

And, that must have hurt, especially because you
knew that the French children weren't even trying.

[French accent] Go on. Play with your
silly numbers.

They tell you nothing of the true nature of
the soul. I weep for you.

In response to statistics like that, President
George W. Bush on just his third day in office

announced his No Child Left Behind program.

It passed Congress with bipartisan support,
because of course it did.

Voting against No Child Left Behind is like
voting against No Puppy Left Unsnuggled.

What monster would do that? His name is
Patches and he needs love!

The program was designed to be data-driven,
and involved testing children

every single year in order to identify and
fix failing schools.

An accountability system must have a
consequence,

otherwise it's not much of an accountability
system.

It's hard to argue with any of that.

Unfortunately, accountability is one of
those concepts that everybody's in favor of,

but nobody knows how to make work, like
synnergy or maxi dresses.

No matter who wears them, they look like
a poncho fucked a waterfall.

You look like the ghost of Gwyneth Paltrow
future.

"I only haunt brunch. Goop."

No Child Left Behind increased the number
of federally mandated tests from 6 to 17,

and the fixation on testing was something
which our current President

seemed to be against as he ran for office.

Don't tell us that the only way to teach
a child is to spend too much

of a year preparing him to fill out a
few bubbles in a standardized test.

We know that's not true!

Wow, that man knew how to pander to
teachers.

"You know what else? There should be pool
tables in teachers' lounge!

And, every year you should be able to
slap one parent!

Vote for me! I'm outta here!
I'm outta here!"

But, once this President took office, he
didn't get rid of tests,

instead he added his own education
initatives, like Race to the Top,

which encouraged states to adopt the
common core,

which featured a logo of snails sixty-nining.

And, again, the intentions here were good
because we do have underperforming schools,

and there are major economic and racial disparities
in the quality of education children receive.

And, anything that can help us narrow those
gaps is obviously a good thing.

The problem has been the implementation.

For instance, many states now tie teacher
pay to performance using one particular approach:

It's called value-added analysis, rating
teachers based on student test scores.

For instance, if a student who ranked in the
60% percentile test higher at the end of year,

the teacher gets a better rating. If the student
falls, the teacher's rating falls.

Okay, well that explains why many teachers'
classroom decorations

that used to read 'Believe in Yourself', now
say 'Don't Fuck Me On This'.

And, while the idea of tying teacher pay to
student improvement sounds great in theory,

here's how it can work in practice.

I have four students whose predicted scores
were literally impossible.

One of my sixth grade students had a predicted
score of 286.34.

However, the highest a sixth grade student can
earn is 283.

The student did earn a 283, incidentally.

Despite the fact she earned a perfect score,
she counted negatively towards my evaluation

because she was three points below her
predicted score.

That is ridiculous. The only way she could
have hit her predicted score

was if she answered everything right, wrote
a few extra questions of her own,

got those right, and then stapled them
to the test.

That teacher lives in Florida, which uses this
formula to assess teachers.

A formula which looks like the kind of thing that
aliens carve into an anti-semite cornfield.

And, many of these formulas on which teacher's
careers depend were partly inspired

by research - and this is true - that modeled
the reproductive trends of livestock.

Basically, we judged the nuance of what happens
in the complicated world of a child's mind,

the same way that we judge this.

Look, I don't know what we did wrong, but
your child is either going ot pass algebra

or birth a healthy calf, I don't know.
Flip a coin.

With the stakes this high, the tests had better
be good.

But, there is ample reason to suspect that is
not the case.

Just look at the Florida Comprehensive
Assessment Test or FCAT.

A Florida school board member was concerned
and a little suspicious when he learned that

only 39% of his state's tenth graders had performed
at or above grade level in reading.

So, he had an idea:

I asked the district at that point to give me
the closest thing they could legally

to the FCAT reading and math test, and
I took it.

That test labelled me as a poor reader, and
I have a couple of masters' degrees,

and I've been re-elected four times and
I teach 39 graduate courses

at six universities in this country.

Okay, okay, we get it. The test sucks!

Anything else you want to brag about
there?

"Oh, I also know how to play 'Mary Had
a Little Lamb' on the recorder.

And, guess who can do sixteen non-consecutive
pushups? This fucking guy!

But, look. He does have a point.

If a test fails to reflect ability, there
are human consequences;

because, one shy Florida eighth grader who
had a near perfect score in her

advanced Language Arts class was asked to
leave it last year due to her

inexplicably low scores on the FCAT.

And, last fall she told a school board meeting
exactly how that felt.

Every year I do good in school, but I get low
test scores on the FCAT,

and it feels like a punch in the stomach.

It's just unfair and I don't want to lose my
opportunity to take my advanced classes

or get a better education just because of
one test.

That is just awful. I take back everything
I said about wanting to take a standardized test.

In fact, you know what? Bring out the monkey!

Turn off his music! Do not applause it!

What the fuck is wrong with you?

What is wrong with you? You made
that little girl cry!

Think about what you've done!

No, no, no, don't you dance it off!

You go and think about what you've
done!

Shame on you! Shame on you!

Look, at this point, you have to ask yourself
if standardized test

are bad for teachers and bad for kids,
who exactly are they good for?

Well, it turns out they're operated by
companies like all these.

And, let's just focus on the largest one:
Pearson.

As of 2012, they had nearly 40% of the testing
market, almost triple their nearest competitor.

And, if you've never heard of them, then
congratulations.

But, just mention their name to any parent
or teacher in the state they operate in,

and you see what happens, because Pearson
are education equivalent of Time Warner Cable.

Either you've never had an interaction with
them and don't care,

or they have ruined your fucking life!

Pearson have a shocking amount of influence
over America's schools.

So much so, that at this point a hypothetical
girl could take Pearson tests

from kindergarten to at least eighth grade -
a test, by the way, that she studied for

using Pearson curriculum and text books
taught to her by teachers who are certified

by their own Pearson test.

If at some point she was tested for a
learning disability like ADHD,

that's also a Pearson test!

And, if she eventually got sick of Pearson and
dropped out, well she'd have to take the GED

which is now, guess what, also
a Pearson test.

In fact, the only test they have no hand
in is the HPV test she might take in college,

and, I can only assume they'll get on that
as soon as they see this fucking show.

Pearson has enjoyed spectacular growth and
profit, and yet, their track record is

littered with complaints concerning technical
glithches,

slow grading, and even the contents of their
tests.

Take what happened in New York just
a few years ago.

Almost thirty different test questions have
now been declared invalid

because they're confusing or have outright
errors.

They already pulled six questions from
an English exam

related to a bizarre passage about a
talking pineapple.

A talking pineapple? Well, at the risk
of sounding like a Dreamworks Executive

talking to a CGI animator, tell me
more about this talking pineapple?

Students had to answer questions about
the story which they say goes like this:

"A pineapple challenges a hare to a race.
Other animals figured the fruit has

a trick up its sleeve, but the hare wins
and the animals eat the pineapple."

It ends with the moral: pineapples don't
have sleeves.

I was really confused because I expected a
lot more from them.

That article about the pineapple and
the hare was stupid and absurd.

Yeah, she's not wrong about that because
we looked up that test section

and we couldn't work out all the answers.

That pineapple item doesn't remotely
work as a test question.

It barely works as a Doors lyric.

It's not just Pearson's questions that are
a problem, it's how they check the answers.

The company posted this ad to Craigslist.
It's to find people to grade the exams.

Craigslist! They looked for scorers on
Craigslist.

Pearson chooses test graders the same
way you'd look for a mattress

full of bedbugs or a no-strings-attached
hand job.

And, to be clear here, this is not just
a Pearson problem.

Across the whole testing industry, you can
find former graders

who will tell you horror stories.

You looked at an essay every two minutes. A short
answer every five seconds, every ten seconds.

We don't understand your kids! We don't
understand anyone's kids.

That is not an acceptable answer from a person
who may be responsible for the future of your child.

It's barely acceptable from the manufacturers of
American Girl dolls.

"Oh, we make dolls for a hundred bucks that
kids can't play with in case they get them dirty."

"We don't understand your kids. We don't
understand anyone's kids."

As a scorer points out, sometimes grades
are given out not based on merit but on quota.

I was told when I was beginning a project, that
last year there was a certain amount of 2's,

a certain amount of 3's, a certain
amount of 4's.

We expect that to be similar this year.

If that's not similar, they will tell you
we're scoring too many 3's or,

we're scoring too many 4's.

They'll say, "You have to learn to see more
papers as a 3."

"You have to learn to see more papers as
a 4."

But, that makes no sense if the content of
what you're looking at is not changed.

That's like telling a baseball umpire, "Hey
we've got a problem with batting averages.

You need to see more home runs as strikeouts
and more strikeouts as doubles. Do it now!"

And, I would love to show you more questions
from these tests,

but unfotunately, that's not only difficult,
it's often illegal.

Because, both states and companies have fought
to keep test questions secret

by having teachers and students sign
statements like:

"I will not use or discuss the content
of secure test materials."

And, while they'll say this is to protect
against teaching, it does seem odd

that even if you see something wrong on the
test, you can't tell anyone.

Standardized tests basically enforce the rule
that all subway riders insitinctively obey:

If you see something, keep it the fuck
to yourself.

We've all seen someone vomit in a purse
before!

Leave it! Focus ahead and leave it!
Bury it!

Look, we've had more than a decade of
standardized testing now.

And, maybe it's time to put the test to
the test.

The original goal was to narrow the
achievement gap and boost our scores

relative to the rest of the world.

Well, a 2013 study found no support
for the idea that No Child Left Behind

narrowed the achievement gap, and our scores
on the international tests have not only

failed to rise, they're slightly down.

And, I do not want to hear what that French
kid thinks of those results.

[French accent] "Oh, all this time and all
this money, and your rise to the top

has been, how you say, a meandering
jog on a treadmill."

All of this calls for a little of what both
Presidents asked for when selling their reforms.

And, the accountability system must have
a consecquence,

otherwise, it's not much of an
accountability system.

Right, so let's look at that because as
far as I can see,

this is a system which has enriched multiple
companies and that pays and fires teachers

with a cattle birthing formula, confuses
children with talking pineapples,

and have the same kind of rules regarding
transparency that Brad Pitt had for 'Fight Club'.

So, for Pearson, the other companies, and
all the lawmakers who have supported this system,

the true test is going to be either convincing
everyone it works or

accepting it doesn't work and fixing it.

Because, at the risk of sounding like a
standardized test scorer,

you're numbers are not good.

If it seems unfair to have your fates
riding on a complicated metric

that fails to take institutional factors into
account and might not even tell the whole story,

well, you're not wrong about that, but you
do not get to complain about it.

And, if all this pressure to increase your numbers
is making you feel nauseous,

like you might vomit at any second,
then don't worry I've got four words for you

that will make you feel better:
here comes the monkey!

Higher standards are the right goal.
Accountability is the right goal.