Standardized Testing

[MacAttack]

Link to video.

I have always hated standardized testing, and now that my kids are going through it, even through they are tearing through them, I still loathe the tests. Just saw this not too long ago on John Oliver's show.

I know Galileo once said "Measure what is measurable, and make measurable what is not so."

Is standardized testing really the way to go about measuring education on a state/national level?

 

[Cheetah]

If the standardized tests have no direct effects on the test subjects or anyone administering the tests, then they can give useful data. What kind of data depends on the tests.

As they are used today however, they're a complete disgrace. Their only actual purpose seems to be to rob money for the public and give it to test-making companies.

As far as I know the national tests in Norway are at least designed and owned by the government, so there's no direct profit motive involved (unless by people who are interested in keeping their jobs, but I believe they have enough other tasks that this effect is avoided).

 

[Wrymouth3]

I understand that both the US and UK are having severe problems with this issue as we speak. As far as I'm concerned, the test is discriminatory and aims to keep profit going for such companies as College Board and Kaplan. The secondary problem with such tests as the SAT and ACT is that they don't actually measure scholastic aptitude and are only based on a few subjects. Even though both the Advanced Placement and IB tests have their own big problems, they at least provide an outlet for a student to make a substantial analysis and provide their knowledge through a writing assessment.

 

[civvver]

You need some objective measure to see if students are learning or not, but once that measure is known most teachers just teach to pass the test and not actually learn the material.

Ultimately I believe every problem with the education system can be linked to homelife and parents/guardians. If parents set expectations high for their children and hold them to those they will succeed. If they don't and expect the schools to teach them everything, including social behavior, then typically don't succeed or succeed at a much lower rate. I just do not believe government can solve these issues as they start and end primarily in the home.

 

[Terxpahseyton]

Governments can perhaps not "solve" it. But they certainly can be more or less helpful.
/stating the obvious

I think good teacher education is a hella more important than putting pressure on teachers with tests. But America seems to have gone with the latter to solve its problems. Which after more than a decade seems to have not worked at all, as John Oliver demonstrated.

 

[nc-1701]

I saw that John Oliver episode, I was unimpressed. A good friend of mine summed up his opinion on John Oliver after watching this as follows:

You cannot imagine how little it impresses me that Oliver thinks something's stupid because it's too mathy for him to understand.

That roughly correlates to my own take on this, value added calculations absolutely make sense in teacher evaluations and it is a disgrace that they aren't more widely used. The anti-math bias in the Common Core and testing critics is disgusting.

All that said there are two things very wrong wth the way we do this testing:
1) We let private companies do it. Pearson, The College Board, Kaplan, etc. Are all leeches taking our money while providing a crappy product. I would ban schools from using commercial tests/curriculums, everything should be done in house by the Department of Education.
2) We give these tests but they don't affect the kids grades, if we are judging the teachers based on how the kids perform on something the kids need to care.


Oh and making the SAT about more subjects misses the point utterly. It absolutely should just be math and reading, nothing else represents potential at all.

 

[Wrymouth3]

I saw that John Oliver episode, I was unimpressed. A good friend of mine summed up his opinion on John Oliver after watching this as follows:



That roughly correlates to my own take on this, value added calculations absolutely make sense in teacher evaluations and it is a disgrace that they aren't more widely used. The anti-math bias in the Common Core and testing critics is disgusting.

All that said there are two things very wrong wth the way we do this testing:
1) We let private companies do it. Pearson, The College Board, Kaplan, etc. Are all leeches taking our money while providing a crappy product. I would ban schools from using commercial tests/curriculums, everything should be done in house by the Department of Education.
2) We give these tests but they don't affect the kids grades, if we are judging the teachers based on how the kids perform on something the kids need to care.


Oh and making the SAT about more subjects misses the point utterly. It absolutely should just be math and reading, nothing else represents potential at all.

There is the matter of the human face in all of this though. Mathematics is essential if not the most essential subject in 2015, however the gamut of other subjects should not be sacrificed in order for that to happen. I would imagine "potential" counts for literally anything someone hopes to understand greater for testing or for knowledge's sake.

 

[Cutlass]

I'm opposed to using much standardized testing. I know where the idea comes from. There is no real objective way in which to judge the comparative quality of the work, the ability/effort of the teachers, the ability and level of the students. We have 50 states, each with it's own education system, perhaps 100s of individual school systems, and many 1000s of schools in each state. With no universal yardstick, how do you determine which are succeeding and which are failing?

But, that said, standardized testing fails to accomplish this as well.

 

[Terxpahseyton]

value added calculations absolutely make sense in teacher evaluations

I don't know much about those calculations, but it seems extremely fishy to me that those equations supposedly can predict how the children of a class should develop or whatever else those equations exactly even do. And this impression may be shared by others holding a "disgusting anti-math bias".
We can be pretty good in mathematically accounting for the average or cumulative behavior of large populations because a lot of variation will cancel each other out. And that may be good enough for a life-stock farm. But as anyone should know, no class is as the other. They are not large enough to cancel out the variation. They will vary. A lot. And then you force this into one equation based on the behavior of life-stock?
Yep, sounds extremely fishy. It isn't automatically useful because it involves math magic.

 

[nc-1701]

There is the matter of the human face in all of this though. Mathematics is essential if not the most essential subject in 2015, however the gamut of other subjects should not be sacrificed in order for that to happen. I would imagine "potential" counts for literally anything someone hopes to understand greater for testing or for knowledge's sake.

Math and writing are skill based "predictor" skills, especially the type of math on the SAT. On the other hand a test on say History is just just testing knowledge. I argue that a knowledge test of a high school student is essentially useless in determining how well they might do in college. Whereas a reading/math test will determine if they have the core skills to succeed in college.

i.e. knowing a bit of history in high school won't help as much taking history in college as having strong reading comprehension skills will.

I don't know much about those calculations, but it seems extremely fishy to me that those equations supposedly can predict how the children of a class should develop or whatever else those equations exactly even do. And this impression may be shared by others holding a "disgusting anti-math bias".
We can be pretty good in mathematically accounting for the average or cumulative behavior of large populations because a lot of variation will cancel each other out. And that may be good enough for a life-stock farm. But as anyone should know, no class is as the other. They are not large enough to cancel out the variation. They will vary. A lot. And then you force this into one equation based on the behavior of life-stock?
Yep, sounds extremely fishy. It isn't automatically useful because it involves math magic.

Math magic sounds horrible and dangerous... But these equations don't claim to predict anything crazy.

I'm sure the actual equations used are a bit more complicated than this, but a rough example of how this might work is as follows:

Spoiler :

A certain school has 1000 students in it. Each student takes a certain test and gets a score, we then rank the students 1-1000 based on the score.
Each student goes through a whole school year and takes the test again next year and we rank the students again.
We calculate a rank change for each student based on how many ranks they went up or down.
Now each teacher has 5 classes with 30 students each for the year, so we have 150 students to compare for this teacher. We total up all of their rank changes and see if his students moved up or down over all.
Then we rank the teachers based on how many cumulative ranks their students gained or lost.

If most of a teachers students fell in the rankings compared to their peers we conclude that teacher is less effective than the others and vice versa. Each teacher has many students so while some students might plummet because their mom got addicted to crack or their parents divorced each teacher has enough students that these cases will fall out in the average.

Actual Value Added models will use additional statistical methods to help smooth out some of the problems in my example and remove outliers, but my example stands. Teachers teach several classes and the variation absolutely does get cancelled out.

The livestock comment of John's was pure ignorant scaremongering. Of course extremely basic math concepts get reused in different fields. It's like saying Tesla's can't be good, they use some of the same technology as battery operated toys for kids

 

[downtown]

NC-1701 is exactly right.

I used to be a teacher. I administered these tests, taught to these tests, had my pay set in part by these tests, etc...and I find most arguments against standardized testing 100% unconvincing. That isn't to say that the tests are great, or that these arguments do not exist (they do), but that most of the common arguments, especially here, suck.

First, re: the SAT/ACT. These are not meant to be content knowledge tests, but, as NC said, they're supposed to help predict and measure potential. Do you have the necessary foundational skills needed to be successful in college? History content won't help you, especially if you aren't studying history. Those foundational skills are reading comprehension (and assorted skilled), basic Algebra and Geometry (which really help measure logical thinking as much as they do the actual math stuff), and more recently, writing.

Is it perfect? No. The current tests are culturally biased and end up giving another advantage to the wealthy in the US. But given the immense variance between High Schools in this US (both in terms of quality and actual curriculum), having some sort of standardized exam play SOME role is about as fair as we're going to get without showing the stomach to really rip the whole system down.

The description of Value Added scores is mostly right, although it's tied to core subject matter mastery, rather than their ranks against each other. You test students at the start of the year in a fairly involved process (over a week, with multiple proctors, and with multiple styles), to get an idea for what grade level their knowledge base is (typically for reading and writing). You do this again 4 times a year, and chart their progress.

A kid might fail their final 5th grade exam, but if they came in as a 2nd grader and left as a 4th grader, that teacher still did a really good job, even if they still "failed". Value-added data is critical in helping us better understand not just classroom quality, but student achievement.

I actually think the tests given then are WAY better than typically end of year assessments/SAT type stuff, because they hit multiple levels of Blooms Taxonomy, whereas scantron tests typically don't.

I'll have more later.

 

[Narz]

Standardized tests are a good way to make the people who work in schools job's easier, no need for creativity, just make sure your students do well on the questions & maybe the school will get more money.

Also it helps in real life because almost every job involves receiving 4 answers & choosing which one fits the best.

If you have rich parents you can just get tons of coaching to do well on these tests. Unfortunately there is likely zero correlation between doing well on them & success if life or even in college (not counting for income & whatnot). The only possible benefit is the discipline to study for them, which of course could be put to more useful ends like learning a 2nd language or useful skill.

 

[nc-1701]

Also it helps in real life because almost every job involves receiving 4 answers & choosing which one fits the best.

I have a rather well paying job and doing exactly this was literally the most significant part of my interview.

 

[Timsup2nothin]

I have a rather well paying job and doing exactly this was literally the most significant part of my interview.

I got my first job in sales based strictly on a Wonderlic test and nothing else.

 

[Wrymouth3]

Until the standardized testing lessens the what does this vocabulary word mean or insert two different words in the blanks to come up with the same definition, it will never effectively test what it is hoping to achieve. There should be a heavy writing portion that has way more emphasis than it currently does.

 

[Narz]

I have a rather well paying job and doing exactly this was literally the most significant part of my interview.

Cool for you.

I'm assuming this is not a job dealing with people.

 

[cav scout]

I think most of the hostility towards standardized testing comes from people who aren't smart enough to do well on tests.

 

[Narz]

I scored within the 95th percentile or higher in pretty much every standardized test I've ever taken but its a pretty shallow measure of understanding & I wouldn't want my child to grow up in a world where her worth is determined by such a crude measure.

The issues facing us in the 21st century aren't going to be solved by little minions who can memorize mathematical formulas & vocabulary words & guess at a correct answer out of two choices (from what I remember from standardized tests usually two of the four answers are blatantly false).

 

[madviking]

oliver's show has slowly turned into a bit of a circus recently. instead of tackling medium sized issues with precision, they've turned towards a shotgun based approach. yes, there are bad things about testing. i know it first hand. yes, it's sometimes a bad system. but the (completely justified) complaints of a few does not a systemic problem make.

 

[Valka D'Ur]

Link to video.

I have always hated standardized testing, and now that my kids are going through it, even through they are tearing through them, I still loathe the tests. Just saw this not too long ago on John Oliver's show.

I know Galileo once said "Measure what is measurable, and make measurable what is not so."

Is standardized testing really the way to go about measuring education on a state/national level?

"This video does not exist."

BTW, I'm reasonably sure Galileo wasn't talking about test scores in school, even though part of his life was spent as a teacher.