In the Navy, we take exams for rank advancement -- we have a phrase. When in doubt, Charlie out. In other words, if you don't know, guess C and move on
Well with a sample size of one the 2022 test had 9/24 the answers as C which is above an expected (6/24) 25%.
Definitely some of the more annoying tests I've taken were when the answers were the same over and over. Its so weird to answer a T/F as T 10 times in a row and I've definitely lost points because I changed some of my Ts to Fs because I thought there was no way it was all Ts.
> Or maybe it's just that you need to pick something, so why not C?
Mostly this. That said, one definitely picks up when teachers bias the dice.
Some want an even distribution. If you confidently answered C's and D's early on and were later forced between A and C, A was the lucky bet. Other teachers avoided the first and last choices like the plague; this was common. With them, B and C (on a four-choice test) were the lucky bets.
In my AP European History class in high school, we had a 25 question quiz the next day every time we were assigned a chapter to read in the textbook (maybe once every week or two). They were intentionally difficult to try to encourage us to read closely, often asking about specific details that were only mentioned once in the chapter, and they were heavily curved, so getting even around half of the questions right generally would get you a decent grade. The teacher had a habit of making the last choice of each question be a combination of two of the earlier choices (e.g. "both A and C"), and these often ended up being the right answer to the point where one of my classmates came up with the idea of choosing the last choice in every single question to try to get an easy passing score. Unfortunately for him, he was a bit too proud of this idea and talked about it enough beforehand that the teacher caught wind of it and purposely made none of the answers on the next quiz have the last choice be correct. Not sure how well that would go over in the Navy, but I'd certainly still be a bit worried that the exam writer might know the rule of thumb people use and have a bit of fun at my expense!
1. it's trivia nowadays to make sure 25% distribution of answers for MC; and
2. I agree with you, in that the main problem is that you only need 17% to pass, i.e. participation awards
Well, C works better for this particular test because it has unusually many C answers in the multiple choice. Relying only on points from MC questions, you would expect to pass ~12% of these tests for any given choice of answer to guess every time.
Now, having a single answer being correct often often enough to pass this mark would happen ~47% of the time, but between the blatant irresponsibility with the standard, the fact that C is the more or less the canonical answer to just guess when you don't (this might be regional or whatever), and the exactness of the cutoff, I wouldn't be surprised if this test were intentionally arranged to make specifically the "pick C" strategy a pass either.
"This would disproportionally affect ..." the rest you know. To be completely equitable, the school should send graduation certificates by mail to those students who chose to opt out of the exams.
I mean kinda. You're seeing the result of people doing their best to create equity with a system they only control a teeny tiny part of. The boil the pond equitable solution would be to make HS optional and disallow discrimination on the basis of having a diploma [1]. The boil the ocean solution is to reorganize schools around actual learning instead of a treadmill of pointless tests. Literally no kid needs these tests, they are for the benefit of school administrators and politicians to have easy to quantify metric that they're doing the bare minimum.
What would you do if you were on the state exam board and you realize that your state's education system is failing huge swaths of the kids in your state but then turning it around and saying that's the kids' problem, and there is nothing you can do to actually affect the underlying problem? They don't have the political capital to eliminate the tests entirely because "test scores" so they make them impossible to fail.
Here, straight from the article:
> Unfortunately, this solution is difficult to implement. Implementation is difficult because of all the reasons that drove NYSED off the cliffs of insanity in the first place. Look at the incentives. NYSED and local school districts have large financial incentives to graduate as many students as possible. These incentives were put in place by both federal and state legislators who meant well and were trying to encourage good public school performance (good performance == graduation, of course, in their eyes). Teachers and principals have large quality-of-life incentives to pass students. Failing students brings scrutiny, parent and student complaints, superintendent pressure, guidance counselor inquisitions, paperwork, and meeting after meeting. It’s much easier to pass everyone.
Kids aren't allowed to fail these tests. But when the school system fails, it's the students that are the one holding the bag because their individual responsibility to pass them and so this is the band-aid. And here's the thing, the sky isn't falling due to this band-aid because the test scores were worthless in the first place. Anyone with academic pursuits didn't need the test, and anyone without them also didn't need the test.
[1] I know, bold statement. But as long as students aren't actually getting an education and there's no real motivation to fix it why even bother with the song and dance? Might as well let the kids sleep in and socialize at the pool or a skate park rather than 5 minutes between classes. I have never met a single person who doesn't echo the sentiment of their HS classes being worthless, and I'm no exception. The only lasting "education" I got when I was 16 was teaching myself calc and programming.
Hey, it is easier to say “you are actually fine, here’s your passing score!” to the disproportionately affected, instead of, maybe, solving the actual problem!
> We keep lowering the bars on tests to give a trophy to every kid
If you read the example you'll see it has nothing to do with giving trophy passing grades to the kids.
Instead it has everything to do with money. There is a bunch of federal and state funding that is keyed off of passing rates, number of kids who graduate, etc. Then for teachers anything other than kids passing means they have to deal with parents complaining, their superintendents applying more pressure, etc. For parents who can't afford to pay for their kid to go to tertiary education of some form, their kid is dependent on "passing" these tests. Finally for the regents: they get negative public pressure if large amounts of kids fail, in addition multi choice tests are _vastly_ easier and more importantly cheaper to grade, so the types of question are limited to those that can have multi choice answers.
So note that in no part of the above does "trophy for every kid" come up, nor is even relevant. Kids feelings aren't considered nor relevant. It is just easier, cheaper, and results in them receiving more money, if the graduation numbers are increased due to passing grades being lower.
the low rating requirements aside, am I understanding it correctly that this regents diploma is for people who finished secondary education? This is what they ask 18 year olds?
You can start taking the Regents test as soon as you've taken the class (it's not just an exit exam). You have to have passed one math class to graduate. So, yes, this could be the step that stands between an 18-year-old and their diploma.
However, if you're on a college track, you have to take more math classes and tests than just the lowest level one. So you probably take it in 9th grade. And there are a lot of levels above "pass" to compete for. So to pass you needed to get 17/86 points on the test. But the cutoff for people who care at all about their academics is 67/86. And those who want to get into the public universities and have a good courseload there are aiming for 76/86 on all their tests.
And again those people are taking the test at 15, so they can take harder tests in subsequent years.
I'm genuinely confused at the criticism of the content of the questions and especially the criticism that they're "not hard". I mean, I guess they aren't algebra in the sense of "use the quadratic formula," but I'm not sure that's a useful sense. The question about percentages and the one about reading the graph are both eminently practical questions, ones that very directly respond to the algebra student's "When are we going to use this in real life" plea. I can see people needing to know this in any practical field of work - and I can realistically see people not knowing it, too. And plugging numbers into the quadratic formula isn't hard, either. It's a matter of knowing the definition, which is precisely a thing the author criticizes, and it's much more direct and mindless application of the definition than that question about whether a certain number is rational.
(And is it supposed to be hard, or a good thing if it's hard? The goal should be that everyone passes; nobody needs a weed-out ninth grade math class. The questions should be meaningful, yes, and they should show that you actually understand the subject. But this is a subject we would, ideally, want everyone to understand well.)
The actual June 2022 Algebra I Regents exam is available online and you can see what the questions look like: https://www.nysedregents.org/algebraone/622/algone62022-exam... Most of them look like conventional algebra to me. And there is indeed one use-the-quadratic-formula question!
(The criticism that a zero-bits-of-information response passes is certainly valid, and could be easily solved by subtracting fractional points for incorrect answers, a common approach on other standardized tests.)
Even that's not punishing enough, since it becomes straightforwardly strategic to guess if you expect your "average" score to fail. You really want wild guesses to be negative expectation, AMC-style. That doesn't exactly solve the problem, but at least becomes less exploitable—if you're competent to optimize the number of guesses you make dynamically during a test, you probably less likely to need to.
I think it's okay for wild guesses to be zero expectation; if you look at it from an information-theory perspective, leaving the question blank and wildly guessing both communicate the same thing, "I don't know." What's the benefit of incentivizing test takers to leave questions blank instead of wildly guessing? I get that it feels better to discourage guessing, but how does it improve the accuracy of the test as a measure of understanding?
And if you can rule out at least one answer, that communicates that you know something about the material under test - provided, of course, the test has no joke or otherwise implausible answers, which appears to be true of the Regents test that I posted above. Take question 6, for example, which asks whether the days of the month are best described as "(1) integers (2) whole numbers (3) rational numbers (4) irrational numbers". Suppose you forgot the distinction between integers and whole numbers, but you knew that the other two answers were incorrect, and so you randomly picked one of the first two options. Doesn't that convey understanding of part of the material under test, such that a score of a positive fractional point in expectation is reasonable?
(A more logistically complicated system would be to permit students to mark any answers they think might be correct and therefore remove the "in expectation" part. If you don't select the correct answer, you get no points, but if you select n answers including the correct answer, you get 1/n points, and leaving a question blank is considered equal to selecting all the answers. But the cognitive overhead of this system probably makes it not worthwhile.)
I had a look at the test and it's indeed reasonable, I don't get why this guy is outraged. I don't even see why this test has to be a pass/fail test, you could just get a score. The questions he featured are clearly cherry-picked. But as you said I don't think this test is even supposed to be hard, it's a test for kids just out of junior high, it's not like a test to be the head of the Federal reserve or something.
The trend is to either eliminate these exams or make them so easy that everyone passes. Why? Because we are unable to deal with the fact that certain ethnic groups consistently fail at dramatically higher rates than others. See the movement to eliminate the SAT and LSAT, to eliminate magnet programs, etc.
why do those ethnic groups perform poorly? are we talking ethnic groups versus whites who have gone to the same 3 elementary/middle/high public schools in the same cities/counties?
> the fact that certain ethnic groups consistently fail at dramatically higher rates than others
I'm having a lot of trouble with this statement. It leaves the reader to guess at which are the groups in question, and for no good reason. If it is indeed a fact why not link a credible source?
I also asked them for a source on a follow up comment the user made, all I received was a downvote and no response.
Pretty safe to say they are just spewing nonsense and dont actually have any evidence.
And of course you refuse to talk about the reason why those particular ethnic groups fail at higher rates. Your also comment suspiciously reminds me of eugenics.
It's only a part of it that is multiple choice, the "outrage" exists because answering all the multiple choice questions randomly can get you a passing grade.
The reason being the passing grade was already low, close to 30% of the entire test and was lowered even more in 2022.
There is also some ranting about some cherry-picked questions that are really reading a graph. I had a look at the test linked by another comment, and I have to say the guy is over reacting, the test is age appropriate, the passing score is super low, but so what, I don't even get why the test has a passing score, you could just get a score.
Also this is a test for children finishing junior high, there is no reason to have a high bar, it probably does not matter.
Tests are easier to grade in a multiple choice format, since the grading problem is reduced to a set of circumstances that are friendly to being automated. Nowadays I'm sure it's possible to have a computer OCR the writing, check the algebra, and grade a problem... but it's a lot cheaper and easier to have the test taker choose the answer from a labeled subset and then check their choice against the correct label.
50 comments
[ 4.2 ms ] story [ 53.7 ms ] threadI've heard there's a tendency to put the "right" answer in the middle, especially if the answers to a question are numerical.
Or maybe it's just that you need to pick something, so why not C?
Definitely some of the more annoying tests I've taken were when the answers were the same over and over. Its so weird to answer a T/F as T 10 times in a row and I've definitely lost points because I changed some of my Ts to Fs because I thought there was no way it was all Ts.
Mostly this. That said, one definitely picks up when teachers bias the dice.
Some want an even distribution. If you confidently answered C's and D's early on and were later forced between A and C, A was the lucky bet. Other teachers avoided the first and last choices like the plague; this was common. With them, B and C (on a four-choice test) were the lucky bets.
We keep lowering the bars on tests to give a trophy to every kid, and we “pickachu face” when we fall behind china on STEM
This article works equally well for guessing A, B or D on every question. The problem isn't randomization.
1. it's trivia nowadays to make sure 25% distribution of answers for MC; and 2. I agree with you, in that the main problem is that you only need 17% to pass, i.e. participation awards
Now, having a single answer being correct often often enough to pass this mark would happen ~47% of the time, but between the blatant irresponsibility with the standard, the fact that C is the more or less the canonical answer to just guess when you don't (this might be regional or whatever), and the exactness of the cutoff, I wouldn't be surprised if this test were intentionally arranged to make specifically the "pick C" strategy a pass either.
What would you do if you were on the state exam board and you realize that your state's education system is failing huge swaths of the kids in your state but then turning it around and saying that's the kids' problem, and there is nothing you can do to actually affect the underlying problem? They don't have the political capital to eliminate the tests entirely because "test scores" so they make them impossible to fail.
Here, straight from the article:
> Unfortunately, this solution is difficult to implement. Implementation is difficult because of all the reasons that drove NYSED off the cliffs of insanity in the first place. Look at the incentives. NYSED and local school districts have large financial incentives to graduate as many students as possible. These incentives were put in place by both federal and state legislators who meant well and were trying to encourage good public school performance (good performance == graduation, of course, in their eyes). Teachers and principals have large quality-of-life incentives to pass students. Failing students brings scrutiny, parent and student complaints, superintendent pressure, guidance counselor inquisitions, paperwork, and meeting after meeting. It’s much easier to pass everyone.
Kids aren't allowed to fail these tests. But when the school system fails, it's the students that are the one holding the bag because their individual responsibility to pass them and so this is the band-aid. And here's the thing, the sky isn't falling due to this band-aid because the test scores were worthless in the first place. Anyone with academic pursuits didn't need the test, and anyone without them also didn't need the test.
[1] I know, bold statement. But as long as students aren't actually getting an education and there's no real motivation to fix it why even bother with the song and dance? Might as well let the kids sleep in and socialize at the pool or a skate park rather than 5 minutes between classes. I have never met a single person who doesn't echo the sentiment of their HS classes being worthless, and I'm no exception. The only lasting "education" I got when I was 16 was teaching myself calc and programming.
If you read the example you'll see it has nothing to do with giving trophy passing grades to the kids.
Instead it has everything to do with money. There is a bunch of federal and state funding that is keyed off of passing rates, number of kids who graduate, etc. Then for teachers anything other than kids passing means they have to deal with parents complaining, their superintendents applying more pressure, etc. For parents who can't afford to pay for their kid to go to tertiary education of some form, their kid is dependent on "passing" these tests. Finally for the regents: they get negative public pressure if large amounts of kids fail, in addition multi choice tests are _vastly_ easier and more importantly cheaper to grade, so the types of question are limited to those that can have multi choice answers.
So note that in no part of the above does "trophy for every kid" come up, nor is even relevant. Kids feelings aren't considered nor relevant. It is just easier, cheaper, and results in them receiving more money, if the graduation numbers are increased due to passing grades being lower.
https://miro.medium.com/max/1400/1*g2H5nPUjOqRALs-AssFEng.pn...
However, if you're on a college track, you have to take more math classes and tests than just the lowest level one. So you probably take it in 9th grade. And there are a lot of levels above "pass" to compete for. So to pass you needed to get 17/86 points on the test. But the cutoff for people who care at all about their academics is 67/86. And those who want to get into the public universities and have a good courseload there are aiming for 76/86 on all their tests.
And again those people are taking the test at 15, so they can take harder tests in subsequent years.
That said it's still inexcusably bad.
(And is it supposed to be hard, or a good thing if it's hard? The goal should be that everyone passes; nobody needs a weed-out ninth grade math class. The questions should be meaningful, yes, and they should show that you actually understand the subject. But this is a subject we would, ideally, want everyone to understand well.)
The actual June 2022 Algebra I Regents exam is available online and you can see what the questions look like: https://www.nysedregents.org/algebraone/622/algone62022-exam... Most of them look like conventional algebra to me. And there is indeed one use-the-quadratic-formula question!
(The criticism that a zero-bits-of-information response passes is certainly valid, and could be easily solved by subtracting fractional points for incorrect answers, a common approach on other standardized tests.)
And if you can rule out at least one answer, that communicates that you know something about the material under test - provided, of course, the test has no joke or otherwise implausible answers, which appears to be true of the Regents test that I posted above. Take question 6, for example, which asks whether the days of the month are best described as "(1) integers (2) whole numbers (3) rational numbers (4) irrational numbers". Suppose you forgot the distinction between integers and whole numbers, but you knew that the other two answers were incorrect, and so you randomly picked one of the first two options. Doesn't that convey understanding of part of the material under test, such that a score of a positive fractional point in expectation is reasonable?
(A more logistically complicated system would be to permit students to mark any answers they think might be correct and therefore remove the "in expectation" part. If you don't select the correct answer, you get no points, but if you select n answers including the correct answer, you get 1/n points, and leaving a question blank is considered equal to selecting all the answers. But the cognitive overhead of this system probably makes it not worthwhile.)
Seems a reasonable thing to do.
Please post a source for your claim.
I'm having a lot of trouble with this statement. It leaves the reader to guess at which are the groups in question, and for no good reason. If it is indeed a fact why not link a credible source?
The reason being the passing grade was already low, close to 30% of the entire test and was lowered even more in 2022.
There is also some ranting about some cherry-picked questions that are really reading a graph. I had a look at the test linked by another comment, and I have to say the guy is over reacting, the test is age appropriate, the passing score is super low, but so what, I don't even get why the test has a passing score, you could just get a score.
Also this is a test for children finishing junior high, there is no reason to have a high bar, it probably does not matter.