This Technicality episode is brought to you by Skillshare.
The first 500 people who sign up can learn with Skillshare for 2 months for just 99 cents.
>> ALEX: Wait what?
Hey guys, I'm here, let's get technical.
Standardized testing isn't great.
If because I said so works for you, then you can just keep watching video, but if you want
to see how we got to that conclusion, check out this video I made to get all caught up.
I was wasn't the only one arguing that standardized tests are bad, though.
A quick Google search or a look in the comments section reveals that a lot of people feel
the same way.
Now, that's great and all, but I think what's missing in our discourse around standardized
tests is, well, what's better.
We can complain about standardized tests all we want, but without any actual solutions,
we're not solving any problems, we're just pointing out that one exists over and
over again.
So, today, let's do that!
Let's fix standardized testing.
First, we need to set some ground rules.
We need to find solutions that capture the advantages of standardized tests, which are
giving students a sense of how their doing, and allowing the government and colleges to
assess the current state of schools and students respectively, but doesn't have the downsides
of standardized tests, which are not educating students (the original goal of increased standardized
tests), inaccurately measuring overall student ability, being very stressful, being biased
towards rich students, and teaching to the test.
That being said, after doing a lot of research, I will present the three best solutions I've
found as what I think is the best way to fix standardized tests.
One!
Multiple-measures systems.
Lemme tell you a story.
This is the New York Performance Standards Consortium; it's an alliance of 28 public
New York high schools, and, while the state of New York usually requires students take
5 tests in a given year, students who go to one of these high schools only take 1; the
ELA.
Why don't they take all of them?
Well, instead of taking the other 4 tests, they would do something called Performance-Based
Assessment Tasks, or PBATs.
As the Washington Post reports, "These include an analytic essay, a social studies research
paper, a science experiment, and an applied mathematics problem.
They incorporate both written and oral components."
So how'd that work of for these high schools?
Replacing standardized tests with project-based learning?
What a preposterous notion!
Another quote from that Washington Post article: "In 2011, 86% of African American and 90%
of Latino male graduates of Consortium schools were accepted to college.
National averages are only 37% and 43%, respectively...
Consortium schools also have far lower rates of student suspension, but far higher rates
of teacher retention."
Those are some pretty spicy statistics, but can this be seen elsewhere?
Yea, these are called multiple-measures systems, and, as FairTest, the national center for
fair and open testing, writes, New York Performance Standards Consortium isn't the only example
of these systems.
The Work Sampling System, designed for younger students ages 3-8 is where students basically
create a portfolio of their work.
Wyoming's "Body of Evidence" system is much like the Consortium's approach,
but with a peer-reviewing element.
It's where they, ok so they go on a pier...
A matter of fact, a lot of countries that aren't 'Murica! do this method really
well.
Scotland, Finland, Sweden, Australia, Singapore, and the UK all have some variation of this
multiple measures system.
This solution solves many of the problems standardized testing presents.
They give students a sense of how their doing at school because they assess each part of
the curriculum.
They educate students because students research and explore their own topic in their own way
(as opposed to just filling out bubbles).
They more accurately measure student ability because they give more outlets for students
to show stuff like creativity, critical thinking, and responsibility.
They're a bit less stressful because students have more time to do the assignment well.
It supports underrepresented students, as the stats I previously presented show, and
they don't force teaching to the test because there isn't really a test to teach to.
The only real downside here is it's hard to generalize.
Each student submitting a portfolio of written work doesn't really translate to big data
for politicians and lawmakers to use when evaluating the current state of our school
system.
Two!
Make tests matter less.
You may have heard standardized testing also called "High-Stakes Testing" and that's
because they matter a lot.
In this solution, students would still take standardized tests, but the results wouldn't
determine big things, like tests wouldn't be an accountability system where teacher
and school funding is dependant on them, and test scores wouldn't be sent to colleges.
Lemme elaborate on those.
First, the fact that we use standardized tests as an accountability system is... odd.
Not only are tests not a great measure of the full student, as we previously discussed,
schools have many other ways of evaluating if a school is doing well.
This is ESSA, or the Every Student Succeeds Act.
It was signed into law by President Obama in December 2015, and, among other things,
it gives states more power over how they want to run their school accountability systems.
What have states done with this power?
The Education Commission of the States finds that some states have adopted or plan to adopt
various accountability metrics, like college readiness, career readiness, growth (as opposed
to actual achievement), attendance, graduation rate, school culture, and even a measure on
how well-rounded the education is.
Not to mention, the current accountability system we use to determine teacher pay, known
as value-added analysis, or VAA, is kinda messed up.
The short version is, in execution, VAA hurts teachers, the long version, John Oliver did
a great piece on this, so watch that.
And second, it's not like colleges are in dire need of information on applicants.
They already evaluate your grades, letters of recommendation, various personal essays,
interviews, extracurricular activities, general interest in the school, and even more, so
taking standardized tests scores out of the mix isn't a huge deal, a matter of fact,
over 1000 colleges already have.
Benefits of tests not mattering include kids, teachers, and parents getting a sense of how
a student's doing in a class WITHOUT stressing out students because that one test doesn't
determine their future.
It also wouldn't force teaching to the test, because school and teacher funding isn't
dependant on how students do, and it allows the government to collect big data and analyze
overall trends in our school system.
However, this method is not without faults, because it fails to tackle of the core, inherent
problems with standardized tests like them not being a good measure of a student in general,
or them being biased towards rich kids, or them not even educating students.
Moreover, the benefits are only benefits if students take the test seriously, and if the
test has no real impact on their future, the external incentive is pretty much taken away,
so you'd have to rely on students being internally motivated to try their best, which,
uh, which uh may have varying degrees of success knowing teenagers.
Three!
Stealth assessments.
You've probably heard of Khan Academy.
Maybe you know them just from their videos, but if you go on their website, you quickly
notice the prevalence of actual quizzes.
So much so that if you click on the topic you want to learn, the first thing Khan Academy
does is prompt you to take an intro test.
And these quizzes are SMART.
They identify which areas of a subject you need more help in and point you to other quizzes
and videos to improve your knowledge in that certain area.
Not to mention, Khan Academy has the ability to save all the data you put into it, if they
don't do that already, which means it can analyze both trends over large populations
of students, AND the progression of an individual kid, seeing how the kid learns and grows over
the course of a semester, or a year, or a lifetime.
And Khan Academy isn't the only company doing this, Scholastic, Dreambox, Brilliant.org,
and others do too.
So do Khan Academy videos improve knowledge?
Well, as Mr. Peanut Butter used to say on Hollywoo Stars and Celebrities, what do they
know, do they know things, let's find out, let's find out.
One study written of in this Huffington Post article found that when 7th grade students
who typically struggled in Math watched Khan Academy videos, students in the Proficient
and Advanced categories increased from 23% to 41%; that's almost doubling.
Khan Academy videos work, and the built in quizzes are the perfect counterpart to track
and measure mastery.
The implementation would be simple: A teacher would just assign X minutes of Khan Academy
for homework, students would do quizzes on the topics they're learning, the concepts
they don't know would be identified, they'd be directed to videos to help them, and the
process would repeat until the homework time is up.
Eddy Burback green screen.
Stealth assessments like Khan Academy present many benefits because
of the way it's fundamentally structured.
Standardized tests are built so you have to show your knowledge, while Khan Academy is
built so you learn and grow until you reach a point of mastery.
Standardized tests are built so you have 1 big opportunity to show your knowledge, while
Khan Academy is built so you have multiple opportunities to see how you improve over
the course of a semester.
Standardized tests are built so getting wrong answers is discouraged, while Khan Academy
is built so getting wrong answers is perfectly fine, as long as you keep learning and work
to improve.
This checks a lot of our boxes.
Students a sense of how their doing because they can literally see which skills they've
master and which skills to improve on, it allows government to track big trends in education
since Khan Academy could track all the answers to all the questions, it's easy to send
to colleges, it educates students, it more accurately measures student ability because
it tracks progress, it's less stressful, it's not biased towards privileged students,
and teaching to the standardized test isn't a thing because there are no standardized
tests.
Would you look at that!
A possible downside, however, is over time these quizzes could get the same stigma as
standardized tests; because they're sent to colleges, you must get every.
single.
answer.
Right.
It'll be important to remember to remind students it's okay and actually the standard
to get something wrong, because that's the whole point of these quizzes, you gets stuff
wrong, you learn from them, you do better.
Those are my 3 top picks.
Some runner up options include, sampling, where instead of giving everyone a standardized
test, you only give them to a statistically representative sample of students, eliminating
multiple choice, which studies should is not the *best* test format, and social and emotional
skills surveys, which are surveys that attempt to measure hope, engagement, and well-being.
However, I believe the three I presented are our best solutions to problems standardized
tests presents.
Honestly, the best solution here may not be to just one change to fix all of standardized
tests, but a combination of each of these solutions.
If we can do that, we won't just fix standardized tests, but make a better and more effective
education system overall.
If you're gonna reform standardized tests, you'll need to know skills like public speaking
and negotiation, and what better way to learn these skills than with SKILLshare.
See what I did ther--
Skillshare is an online learning community with thousands of classes in design, business,
technology and even more.
A Premium Membership gives you unlimited access to high quality classes on must-know topics,
so you can improve your skills, unlock new opportunities, and do the work you love.
Whether you want to learn how to edit your videos like me, or learn to take a stunning
picture of a tree, Skillshare is perfect.
Sponsorship aside, I genuinely use Skillshare.
I signed up about a month ago after hearing my favourite YouTuber talk about it.
*winks zooms in*
I'm currently learning motion graphics with Kurzgesagt and video creation with Wendover
productions, and if you want to join me, you wanna support my channel, go to skl.sh/technciality
to get 2 months of skillshare for just 99 cents.
From then on, the annual subscription is less than $10 a month, it's one of the most affordable
learn platform out there.
Trust me, it's worth it, click the link in the description.
Hey it's March now how'd that happen what?
To the people who came here over from SciShow, hello, welcome to Technicality!
I make videos about science, humanities, and everything that blows my mind, so if you wanna
come along for the journey, click that subscribe button, if you liked this, it means a lot
if you hit that like button, and watch the videos recommended on screen right now.
Thank you to all my patrons over at patreon.com/Technicality, especially these awesome people.
And if you're still watching, wow, that's awesome,
thank you this means a lot.
Thanks for watching, DFTBA, and explore on.
Không có nhận xét nào:
Đăng nhận xét