Jump to content

Menu

Recommended Posts

Posted (edited)

In a rabbit trail last month on the college board, some of us got to talking about time management on standardized college entrance exams. Here are excerpts from three of those posts.

On 12/7/2019 at 2:24 AM, lewelma said:

I really think the SAT is just a horrible test.  Having just spent a year working through it with a kid, I have come to believe that it is a test of accuracy, detail focus, and speed. It in no way reflects a person who is college ready.  It does not test for what I think college is about - creativity, synthesis, analysis, interpretation, and evaluation.  

And then, 

On 12/7/2019 at 11:41 PM, lewelma said:

To get a top score, we got ds a stop watch and removed the beeper physically so it could not make noise.  He took it 2.5 years ago, but from what I remember, he needed to manage his time *very* tightly to get a 1580, and thus needed the stop watch.  Not sure it is even allowed in America, but in NZ they didn't bat an eye because it was silent. 

I will say it again, at top scores it is a test of mental quickness. My older boy is wicked fast, so it worked to his advantage. But I think it is *completely* unfair to have a college readiness test be so focused on speed. Speed does NOT equal capability. No way. 

My eldest took both the SAT & the ACT within the last couple of years, so I commented:

On 12/8/2019 at 6:14 AM, RootAnn said:

Compared to the ACT, the SAT is leisurely. For kids who are not fast computers or readers, I recommend the SAT. My dd#1 took both, scored comparably, & always ended up with more time on the SAT, especially on the math sections. The Science & Math sections of the ACT require rigid time management, judgment of when to guess & skip, and an ability to skim effectively. Slow readers (& those which fatigue easily) are at a huge disadvantage on the ACT due to all the reading in the English & Reading tests.

I thought of this discussion when I was skimming through Prep Scholar's The Best Way to Review Your Mistakes for the ACT/SAT. (It is time for me to prep my #2 kid & she's different than my first kid and the young adult I've been prepping the last year.)

On the SAT, the breakdown of total time allowed for each question (including double-checking!) is as follows:

Section

Total Questions

Total Time (minutes)

Approximate Time per Question

Reading

52

65

75 seconds

Writing and Language

44

35

47 seconds

Math (No Calculator)

20

25

75 seconds

Math (Calculator)

38

55

86 seconds

 

And here's the breakdown of time per question for the ACT:

Section

Total Questions

Total Time (minutes)

Approximate Time per Question

English

75

45

36 seconds

Math

60

60

60 seconds

Reading

40

35

52 seconds

Science

40

55

52 seconds

This is in the section addressed to scorers at the 500 (per section) level on the SAT & 21 on the ACT, not the higher scorers! The time crunch is real for most test takers.

More & more colleges are going test optional. Should the US try an untimed exam? Separately, is there a way to more universally test skills needed in college (lewelma suggests, "creativity, synthesis, analysis, interpretation, and evaluation.")? What changes would you make or suggest in the college entrance exam landscape? Or should we just have sonething like Subject Exams instead?

Edited by RootAnn
Typos
  • Like 3
Posted

Oh. That's an interesting thought. I haven't seen that discussed as a possible alternative. Like, instead of changing the tests or worrying about who needs more time, just ditch the timing.

The test would have to change. With more time, it's obviously going to be hard to distinguish scores. More students would cluster at the top. So then you need a test that helps get that spread. That would potentially lead to writing... but then... you run into the same issues that the writing test has now. And it's basically defunct for a reason.

  • Like 1
Posted (edited)

They would have to make those tests harder. SAT math is so ridiculously easy, I can’t believe it is a test of college readiness.  And it’s already a slow test. 
 

Or just ditch it completely and have students do subject tests. I think those are better tests. 

 

Edited by Roadrunner
  • Like 2
Posted

I like the un-timed option even if the scores cluster higher. If it is a test for college readiness, why do we need to stretch out the top marks?  Either you are ready or not with a bit of a sliding scale. We don't need a bell curve. A bell curve is an appropriate model for a test of mental quickness or IQ because that is a trait I would expect to be modeled with that type of distribution.  But college readiness?  People are ready in different ways. I'm not sure a bell curve is appropriate. 

  • Like 3
Posted (edited)
36 minutes ago, Roadrunner said:

SAT math is so ridiculously easy, I can’t believe it is a test of college readiness.

Agreed--with the exception of a very few problems (and sometimes none), it is a test of 8th grade math and below (if you assume that the average college bound kid takes Algebra 1 in 8th grade).  Then again, the GRE quantitative section is even more of a joke.

I've been taking practice ACTs and SATs (math section only) this week.  It is striking the difference between the tests.  It used to be that the SAT was (potentially) harder than the ACT because the problems were more puzzle-like.  But the new SAT is much easier.  Not only can I finish in 2/3 the time (vs struggling to finish the ACT), the ACT has a lot of geometry problems, Algebra 2/precalculus problems, and trig beyond right triangle stuff.

That said, unless you're going into a quantitative major, you aren't going to encounter much math in college so "college readiness" when it comes to math is a joke.  Not as much with reading and writing though--everyone needs to be able to do that, regardless of major.  And I assume that this is why the GRE reading section is actually at an early college level whereas the math is essentially all arithmetic (with a smidge of probability and statistics), since most college graduates in non-quantitative majors won't remember much math from high school.

Edited by EKS
  • Like 3
Posted
2 minutes ago, lewelma said:

If it is a test for college readiness, why do we need to stretch out the top marks? 

There needs to be some means to separate mediocre from good from outstanding.  Perhaps there should be two tests, one pass/fail college readiness test and another that can distinguish better at the high end. 

  • Like 2
Posted

With regard to timed vs untimed, I think that the ACT/SAT would do well to move to an adaptive, untimed test that has a good sized pool of really difficult items to draw from.  That way the people who can't do the harder problems won't be demoralized by seeing item after item that is beyond them, and the ones who can do them get a chance to shine.

  • Like 3
Posted
1 hour ago, Roadrunner said:

They would have to make those tests harder. SAT math is so ridiculously easy, I can’t believe it is a test of college readiness.  And it’s already a slow test. 
 

Or just ditch it completely and have students do subject tests. I think those are better tests. 

 

It’s not a slow test for many kids. Mine improve significantly when they take it untimed.

  • Like 3
Posted
1 hour ago, lewelma said:

I like the un-timed option even if the scores cluster higher. If it is a test for college readiness, why do we need to stretch out the top marks?  Either you are ready or not with a bit of a sliding scale. We don't need a bell curve. A bell curve is an appropriate model for a test of mental quickness or IQ because that is a trait I would expect to be modeled with that type of distribution.  But college readiness?  People are ready in different ways. I'm not sure a bell curve is appropriate. 

Yeah, I do agree with this. If a kid scores a 5 on an AP exam, they score a 5. End of story. The SAT is nothing like that. It’s one of my complaints.

  • Like 1
Posted
47 minutes ago, regentrude said:

As much as I disliked the SAT/ACT, an untimed test is completely impractical. Who is going to proctor that for the kid who wants to sit there for ten hours? 

I agree, but there are other achievement tests that are untimed--the Accuplacer, for example, the MAP, and various state tests.

  • Like 2
Posted (edited)
13 minutes ago, Farrar said:

It’s not a slow test for many kids. Mine improve significantly when they take it untimed.


amybody would improve when it’s untimed.

 

the entire idea behind those tests is to differentiate. Unless all colleges open doors for everybody, there has got to be some competitive score to differentiate them.

I get that people think SAT favors rich because rich can prep the kids,  but everything favors rich. You should see the tutoring local private school kids get for subjects in addition to $45k tuition. Of course their AP scores and GPAs will be higher when they have a tutor for every subject. 
 

But for a kid with no learning disabilities and very good commend of math, SAT math at least is not complicated at all. 

Edited by Roadrunner
  • Like 1
Posted
1 hour ago, EKS said:

With regard to timed vs untimed, I think that the ACT/SAT would do well to move to an adaptive, untimed test that has a good sized pool of really difficult items to draw from.  That way the people who can't do the harder problems won't be demoralized by seeing item after item that is beyond them, and the ones who can do them get a chance to shine.

I like this idea. Since the ACT is moving to computer-based administration, this would be a way to show college ready/not ready & still continue to differentiate for the high scorers. Plus, if it is adaptive, it might not take a zillion hours even if it was untimed. 

1 hour ago, EKS said:

There needs to be some means to separate mediocre from good from outstanding.  Perhaps there should be two tests, one pass/fail college readiness test and another that can distinguish better at the high end. 

An adaptive test could do both, right?

The only thing with adaptive is to make sure the system didn't kick you completely out for a typo . Two questions about the same difficulty level before dropping down & still give the chance for building back up to a tougher level through the adaptive system.

I still don't know what to do about writing, though. Now, many schools place kids in or opt kids out of English (writing) classes based on ACT English/Reading or SAT EBRW scores. Now, I only have two data points on this so far, but super high scores on those do not always equal a great writer & decent (but not stellar) scores can sometimes be from a great writer.

Posted
9 minutes ago, EKS said:

I agree, but there are other achievement tests that are untimed--the Accuplacer, for example, the MAP, and various state tests.

our state tests are timed. They happen at school during school time. How does an untimed test work in practice? The testing center people want to go home eventually

Posted

One could also move to standardized high school exit exams that certify college readiness. They exist in many countries. We have those in Germany, the Abitur. Students take examinations in several core subjects over the course of a couple weeks,  actual essays and worked out problems, plus oral examinations. 

  • Like 1
Posted (edited)
9 minutes ago, Roadrunner said:

anybody would improve when it’s untimed.

Erp. Gotta disagree on this one. If you don't know the content, more time doesn't help. I have given practice tests out where the kids finish early because they have absolutely no idea so they just guess. Some don't even have enough knowledge to narrow down the choices (thus using up time) before guessing. Some people's scores will not improve with more time. In my limited experience, this is in the ACT of 18 or so range +/- 3. (SAT 950?)

Edited by RootAnn
  • Like 2
Posted
3 minutes ago, RootAnn said:

Erp. Gotta disagree on this one. If you don't know the content, more time doesn't help. I have given practice tests out where the kids finish early because they have absolutely no idea so they just guess. Some don't even have enough knowledge to narrow down the choices (thus using up time) before guessing. Some people's scores will not improve with more time. In my limited experience, this is in the ACT of 18 or so range +/- 3. (SAT 950?)


improve,  not get a perfect score. 
 

My kid can solve AMC8 problems almost all given 1.5 hours. On an actual test that’s 40 minutes, he can’t even make it to half until the time runs out. 
 

there are certainly questions on SAT that given plenty of time kids can figure out. Maybe not all, but a lot of kids could improve. I mean the material on this test is so basic. There is no calculus or precalculus in it. All schools offer algebra 1 and 2 as far as i know. If you paid attention and studied, you won’t encounter any material that you simply haven’t learned. 
 

So if the material on the test was made harder, people would start screaming that poor children don’t get that sort of education. I mean there is no winning here. 

Posted
14 minutes ago, regentrude said:

How does an untimed test work in practice? The testing center people want to go home eventually

I think that there are two things going on.  First, the vast majority of people are going to give up before they sit in the exam room for 10 hours.  Second, all of the tests I know about that are untimed aren't as high stakes as the SAT/ACT.  So I could see a perfectionistic high achiever sitting there forever.  Perhaps having a really generous time limit, like double what is given now.  The exams could also be taken in pieces rather than all at once as is done now.

  • Like 1
Posted
24 minutes ago, Roadrunner said:


amybody would improve when it’s untimed.

 

the entire idea behind those tests is to differentiate. Unless all colleges open doors for everybody, there has got to be some competitive score to differentiate them.

I get that people think SAT favors rich because rich can prep the kids,  but everything favors rich. You should see the tutoring local private school kids get for subjects in addition to $45k tuition. Of course their AP scores and GPAs will be higher when they have a tutor for every subject. 
 

But for a kid with no learning disabilities and very good commend of math, SAT math at least is not complicated at all. 

My point is that the primary differentiation is between fast and slow, not capable and less capable, or smart and less smart. It’s unbelievably narrow. And it’s one thing for a score to boost a little, and another for it to go from no options to lots of options, which is the range I’m talking about. Like, it’s great that you and your kids think it’s such a leisurely test. So... they get options and my kids... don’t. Again, not because they find it hard, but because they move slower. Could we get accommodations? With the testing I have, we qualify, but I doubt we’ll get them.

  • Like 2
Posted
5 minutes ago, Roadrunner said:

improve,  not get a perfect score. 
 

there are certainly questions on SAT that given plenty of time kids can figure out. Maybe not all, but a lot of kids could improve. I mean the material on this test is so basic. There is no calculus or precalculus in it. All schools offer algebra 1 and 2 as far as i know. If you paid attention and studied, you won’t encounter any material that you simply haven’t learned. 

You used the word "anybody" and my point is that is just not true. Sure, time helps some. Many, probably. That was the point. You had said it was a slow test. @Farrar said it isn't slow for many.

I am in a completely different part of the US than you, education-wise. We see completely different calibers of kids on a regular basis. The state that I live in right now requires 3 yrs of math but does not specify what those 3 yrs have to include (as in, it doesn't say you have to at least cover Alg 2). It has only been in the last 5 yrs that they've required 3 yrs. It used to be 2 yrs. Just because they might gave taken & passed an Algebra 2 class doesn't mean they understand enough to be able to correctly answer even algebra 1 content. There is a reason the average score is what it is on these exams. And not all of the reason is time. For many, more time would help them get more correct - absolutely! That's part of my point in this thread.

But not all. Not "anybody." My world contains more 35-60% scorers than 90%+ scorers. The local 4 yr college has a 25% ACT Composite of 16 & a 75% of 23. On the 2017-18 CDS, 36% of their enrolled freshmen class has an ACT Composite of 12-17, 43% score in the 18-23 range, and 20% in the 24-29 range. This is my world (and why my eldest had to leave the area to find academic peers). So, your "basic" does not equal these kids' basic.

But yes, more time would help many kids score better. Would eliminative the timing show more college readiness? Are the kids with these low scores "college ready"? Some might be. Some might not be. I'm not convinced the current ACT/SAT actually shows college readiness. Thus part of the reasoning behind this thread.

  • Like 2
Posted (edited)

The goal of this test is not the equal outcome from kids. I mean it’s the opposite. Capable is a big word. 
AMC makes their tests super fast and I know kids who ace them are much, much smarter than my kid. Mine needs double the time for The same outcome. It is what it is. 
 

College readiness is also a big word. Most schools take kids with very average SATs. So this debate is only really about a tiny select group of colleges where you really need to tip the top, right? I mean of you are going to major in math at UCLA and can’t ace SAT math.... 

 

i am all for subject tests instead. A future history major doesn’t need 800 on a math section. They can pick and shine in relevant subjects.

Edited by Roadrunner
  • Like 1
Posted
22 minutes ago, RootAnn said:

You used the word "anybody" and my point is that is just not true. Sure, time helps some. Many, probably. That was the point. You had said it was a slow test. @Farrar said it isn't slow for many.

I am in a completely different part of the US than you, education-wise. We see completely different calibers of kids on a regular basis. The state that I live in right now requires 3 yrs of math but does not specify what those 3 yrs have to include (as in, it doesn't say you have to at least cover Alg 2). It has only been in the last 5 yrs that they've required 3 yrs. It used to be 2 yrs. Just because they might gave taken & passed an Algebra 2 class doesn't mean they understand enough to be able to correctly answer even algebra 1 content. There is a reason the average score is what it is on these exams. And not all of the reason is time. For many, more time would help them get more correct - absolutely! That's part of my point in this thread.

But not all. Not "anybody." My world contains more 35-60% scorers than 90%+ scorers. The local 4 yr college has a 25% ACT Composite of 16 & a 75% of 23. On the 2017-18 CDS, 36% of their enrolled freshmen class has an ACT Composite of 12-17, 43% score in the 18-23 range, and 20% in the 24-29 range. This is my world (and why my eldest had to leave the area to find academic peers). So, your "basic" does not equal these kids' basic.

But yes, more time would help many kids score better. Would eliminative the timing show more college readiness? Are the kids with these low scores "college ready"? Some might be. Some might not be. I'm not convinced the current ACT/SAT actually shows college readiness. Thus part of the reasoning behind this thread.


So we really need to work on bringing up the quality of education and state mandate certain level. Every high school should offer through precalculus. I am appalled that people don’t care enough to fight for this sort of thing. 
this isn’t the problem of a test. It’s a problem of a state.

  • Like 1
Posted

Add me to the camp of the ACT/SAT being tests of fast-processing speed vs college readiness or even academic ability.  Fast-processing speed is what the test designers have decided to use as the defining factor for distinguishing amg students.  But, that is a poor option, imo, for a filter. Nothing about academia or real life focuses on the same type of time crunch, especially the ACT.  (definitely not college tests or careers)

  • Like 4
Posted
2 hours ago, Roadrunner said:

So we really need to work on bringing up the quality of education and state mandate certain level. Every high school should offer through precalculus. I am appalled that people don’t care enough to fight for this sort of thing. 
this isn’t the problem of a test. It’s a problem of a state.

So are you saying that the state should ensure that all students master precalculus or just that they should have the opportunity to take it?

Posted (edited)
44 minutes ago, EKS said:

So are you saying that the state should ensure that all students master precalculus or just that they should have the opportunity to take it?


I think all kids should have an opportunity to take it. 
 

obviously anybody taking SAT wants to go to college and precalculus is considered high school level math and should be offered in high school. I am fairly flexible in my thinking about what kids want to actually take. I could see a kid wanting to go to an art school not taking anything beyond algebra. But a kid who wants to go into a technical field shouldn’t be disadvantaged by not having an opportunity to take precalculus at school. 

Edited by Roadrunner
Posted
2 hours ago, Roadrunner said:

So we really need to work on bringing up the quality of education and state mandate certain level. Every high school should offer through precalculus. I am appalled that people don’t care enough to fight for this sort of thing. 
this isn’t the problem of a test. It’s a problem of a state.

I think you missed my point. (My state's average ACT score is only 2 pts lower than California's and the majority of our graduates take the ACT because it has become an "exit exam" (albeit taken in spring of junior yr).) But the average score, even for kids headed to college, isn't very high. Are these tests a good measure of ability?

In my opinion, these tests are not good measures of college readiness. How many of us have had a kid score highly on the ACT at a young age but who isn't ready for college in our opinion? They are not necessarily even a good way to identify top performers among the population (as those who have lower processing speed aren't necessarily dumber than those with higher processing speeds). 

I can point to two sisters -- one with a much higher ACT than "college ready" and one who received an average score (and thus not "college ready," at least not in all areas). Both went to college. Both got decent grades (mostly As) while they attended. Their scores did not predict their college readiness. (Both ultimately dropped out before finishing. I'm not sure any test would have been able to predict that.)

I'm sure others can point to low scoring kids who thrived in college -- some who graduated with honors from tough programs. What improvements could be proposed to better show both college readiness and identify the "top" ones who colleges should offer enrollment or merit scholarships to?

Maybe Subject Tests should be a part of that. I'm not sure it should be all, but I like the possibility if showing readiness for a certain area that way.

Posted (edited)
7 minutes ago, RootAnn said:

I think you missed my point. (My state's average ACT score is only 2 pts lower than California's and the majority of our graduates take the ACT because it has become an "exit exam" (albeit taken in spring of junior yr).) But the average score, even for kids headed to college, isn't very high. Are these tests a good measure of ability?

In my opinion, these tests are not good measures of college readiness. How many of us have had a kid score highly on the ACT at a young age but who isn't ready for college in our opinion? They are not necessarily even a good way to identify top performers among the population (as those who have lower processing speed aren't necessarily dumber than those with higher processing speeds). 

I can point to two sisters -- one with a much higher ACT than "college ready" and one who received an average score (and thus not "college ready," at least not in all areas). Both went to college. Both got decent grades (mostly As) while they attended. Their scores did not predict their college readiness. (Both ultimately dropped out before finishing. I'm not sure any test would have been able to predict that.)

I'm sure others can point to low scoring kids who thrived in college -- some who graduated with honors from tough programs. What improvements could be proposed to better show both college readiness and identify the "top" ones who colleges should offer enrollment or merit scholarships to?

Maybe Subject Tests should be a part of that. I'm not sure it should be all, but I like the possibility if showing readiness for a certain area that way.

 
I don’t think any test alone is a good measure of anything really. I am not sure one can come up with a system that will be satisfactory for everybody. I do strongly believe that we need to have something that is more “objective” (everybody measured against the same set of problems) than just relying on GPA (which is vastly inflated in rich districts especially) or essays (often written by paid adults for rich kids) or extracurriculars (also favors wealthy parents who can invest serious cash into artistic and athletic areas). I am open to experimentation of content/time/provider, but I am afraid that test optional movement will really backfire on less financially fortunate. 
 

CA has a significant percentage of kids who are English learners.

 

 

Edited by Roadrunner

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...