Maybe be not as big as the US but there are forums dedicated to university admissions and most university have their own reddit forums. The Ontario and Canada University subs are pretty big and active. They are the Canadian equivalent of CC.
1 Like
I do think it depends on where in Canada. I have no doubt that there are concentrations of students who are āredditā junkies. And yes, it is a horrible source of information.
I have been standing in the front of college classrooms in the USA since 1996. The first students for whom I was a TA were Gen-Xers, and the most recent students that I thought were Gen-Zs. There is this absolute and total delusion here that, somehow, WE were different, and that Kids These Days, etc.
The issues we had were the exact opposite. Every student thought that they knew everything and didnāt need any guidance. So they messed up, trashed labs, destroyed other peopleās research, etc. So these Gen-Xers, grew up, had kids, and taught them not to trust themselves to do anything by themselves, so that they donāt mess up.
As for
Thatās because their professors cannot be bothered to teach them how to separate good information from nonsense. There is an overwhelming amount of bad information out there, and they simply can no longer trust Google. Those of use who were around before Google and have an idea of how Google works, generally can figure out which is the best answers.
They are digital natives, and, since I have been surrounded by immigrants my entire life, I can say that non-native speakers of English who consciously took the time to learn the language well are better at speaking English than the vast majority of native English speakers, including college-educated English speakers. That is why younger Boomers, Gen-Xers and some older Millennials tend to be a lot better at navigating the digital world that younger Millennials and Gen-Zs who grew up without ever having to figure out how any of the digital world around them works.
Finally:
[Rant]
If there are any issues with Gen-Zs, it is because their parents, younger Boomers, Gen-Xers, and older Millennials failed as parents.
People here act as though the way that Gen-Zs act is somehow the fault of the Gen-Zs. It drives me crazy. People on CC will complain about how kids cannot do anything by themselves, three days after boasting how great they were as parents because they made sure that their kid didnāt have to struggle. People overload their kids with expectations, and them wonder why their kids and their kidsā friends are all afraid to take chances and are anxious about failing.
I have never seen a generation which wants to do good as much as the Gen-Zs want to. I am tired of parent seeing this as a failure, and I am tired of parents blaming these kids for issues that are 100% the result of overinvolved and overbearing parenting.
[/Rant]
Thatās what I wrote way above. I believe that we need to rethink both the SATs and how standardized testing is done, but it is very helpful overall to know what students are retaining when they graduate high school. We just spent 12 years educating them, we really need to know what went on and what they retained.
I just think that the SATs and ACTs arenāt doing that.
I do think that, to begin with, they should start from scratch. Taking something that was inherently flawed, and to keep on trying to āfixā it so it will no longer be flawed that way, has been a mistake. Rather than take the existing SATs, and then try and tweak it so that it says something, they should decide what they want their test to tell them, and then to start working backwards from there.
At least 50 percent are the result of the pandemic which pulled the kids out of classroom, gyms, playgrounds, sports fields, bars and cafes, other peopleās homes, had all class and EC trips cancelled and pushed them into an online world they can barely find their way out of, and parents are helpless.
I canāt send my kids outside to socialise if thereās no one there, because everyoneās online, completely isolated without their smartphones and instagram accounts.
I canāt teach my kids to be independent of me, because by definition , as long as Iām there teaching, theyāre not. O have to send them outside to be independent, but in a protected environment, a class trip or in a peer group, not on their own. Beyond middle school, I canāt create the peer group for them anymore.
Just once, I wish my oldest would go out to get drunk. While he can still come home to sober up and be hungover safely. (Just the once, mind.)
Itās not 100 percent parents. I wish!
1 Like
My three kids are all Genz> I concur that they are a much maligned āage groupā and 99% of the issues relate to the āphobiasā of their parents. I see it here on the boards as well. How so immersed parents are in each decision a kid has to make. I was fortunate in that as Genxer, my dad (a product of the 40s) was the first of our family to go to university and it was McGill. His parents didnāt speak english too well. But, what he appreciated most was freedom to make mistakes knowing your parents will have your back if you try. So, when I was in my late teens, his message to me was ādo your best, do well in schoolā and āI wonāt be on your caseā.
Do you think the SAT Subject Tests were a better option for testing? When I applied to Ivy schools in the late 60s you took the verbal and math tests in the morning and up to 3 achievement tests in the afternoon. That seems to have been a fairly full spectrum
1 Like
I just think that the entire idea of how people are tested needs to be rethought. What the questions look like, what the format looks like, etc. Somehow stop thinking of trying to modify what has been used for a long time. It should be able to test for what students have retained and can use, and studying specifically for the test shouldnāt help - only reviewing all the material that is covered.
I also think that ABET should develop an entry test for anybody who wants to study engineering. They need to take it if they want to be accepted directly to engineering or if they want to transfer to engineering. The SATs are neither sufficient, nor do they actually test for the sort of approach to problem solving that engineering requires. At very least that will keep students from applying to engineering and dropping out, when they discover that engineering is very math heavy. Colleges should have āpre-engineeringā for kids who did not do well enough on that test, if they simply donāt have the math background.
There are already some interesting programs out there for students interested in engineering. For example, Purdue has started four polytechnic high schools in IN; they have engineering summer programs for students starting in middle school up through rising seniors; and summer start for accepted students who may need some shoring up/catching up.
Cornell has had a summer start option to help incoming students catch up for decades now.
2 Likes
I think this is a terrific ideaā¦butā¦I can see a lot of people saying you are just coming up with a barrier that will prevent students from poorly resourced (under-performing) schools from becoming Engineers.
2 Likes
Right. And any new test will be second guessed just like the current ones.
This thread has become sound and fury signifying nothing. (Who said that?)
Not if there is a pre-engineering major for kids who havenāt yet taken the required courses, or who need a repeat to be ready for an engineering major.
Iām all for relieving US high school students from what Iād consider an unnecessarily heavy research burden in this process, butā¦if a student is blindsided by the fact that engineering is math heavy, they are not college ready at all.
And Iām pretty sure there is data on the correlation between maths scores in the SAT or ACt and engineering first year GPA and retention and that the correlation is strong enough to be useful already. Happy to be disproven, but maybe this is not a wheel that needs to be reinvented?
I think someone said UT Austin has something like this. This would be great for public/state universitiesā¦likely not an ideal expectation for highly selective private universities.
The fact that a statistically significant correlation in isolation exists does not mean that there is no reason to change/improve the system. For example, the UC study found math SAT score explained 7% of variation in engineering class grades. That 7% of variance explained was statistically significant, but there is also a lot of room for improvement with the other 93% not explained by math SAT.
First year GPA has been well discussed, but there has been little discussion about first year retention. The study at https://peer.asee.org/prediction-of-sophomore-retention.pdf reviews sophomore retention of engineering students at Notre Dame. The author notes that students who persisted in engineering had a mean SAT of 720 math / 680 verbal, while those who did not persist had a lower mean SAT of 700 math / 670 verbal. 720 vs 700 SAT score is a statistically significant difference, yet SAT score did not add a statistically significant degree of improvement to prediction of engineering retention in any of the evaluated models. When they added in a control for first semester grades, the math SAT score regression factor went negative, meaning that a higher math SAT score was negatively correlated with persisting in engineering among kids with similar first semester grades. However, in many of the models number of HS AP credits, having relevant HS ECs, and personal motivations (may be inferred through other parts of application) were predictive of engineering persistance.
Looking at this study, I donāt think the conclusion would be there is a correlation between SAT score and first year retention in isolation, so no reason to change anything. Instead, I think it suggests looking deeper in to why some students succeed and others do not, and addressing relevant issues.
For example, there is certainly a connection between having good HS preparation and succeeding in engineering, but # of AP credits seems to be a better predictor of this HS preparation than does SAT score. Perhaps there is a more effective way to evaluate HS preparation than either # of APs or SAT score in isolation. It might be more effective to manually evaluate course rigor, strength of schedule, highest math level taken in HS, etc. If you want to use a standardized test, a standardized test that emphasizes the types of material that is most important to be successful in engineering may be more effective, probably more similar to the placement tests that many colleges discussed in this thread give to all freshmen.
However, even if you can improve prediction of engineering retention by a combination of such measures, that doesnāt mean the college should only admit kids who do best on such metrics. A college may also want to admit students who had relatively weaker HS preparation through no fault of their own (took highest level courses available at their relatively weaker HS) and give those kids a chance to catch up This could include things like a having different math starting points for kids from weaker HS backgrounds before advancing to the standard sophomore year engineering curriculum. Youād of course need some way of evaluating whether such programs were successful, what measures predict which of those weaker HS background kids are likely to catch up and ultimately persist in engineering, and what happens to the kids who do not persist (switching to a social sciences major and graduating with 3.high GPA is a much better outcome than dropping out).
But thatās the problem with all the studies performed at institutions that required the SAT for admissions at the time.
The SAT scores of students at all selective universities that required the SATs will have clustered in a narrow high band because they were selected for that.
In theory, the difference between a 720 and a 700 SAT may be significant. In practice, everyone knows itās not.
MIT knows itās not, which is why they always used to set a threshold and didnāt look at SATs beyond that. It was only when the highly selectives stopped considering it that they noticed it does make difference. One would need to look at the student body of a large flagship in states like Arizona or Wyoming where SAT scores actually vary to find out how much difference it makes. itās not about whether it makes sense to admit (or have them self select) the 720 student over the 700 student for engineering. Itās whether it makes sense to set a 700 or maybe 600 threshold for a test on material every single student has had a class in.
When every student in a study had an SAT over 700 and a GPA over 3.5 because thatās what they were selected for you will find that what makes a difference in what they were actually different in, and thatāsā¦high schools and their rigour. Not surprisingly the studies found exactly that.
So now they find that AP preparation is predictive precisely because not everyone has it. If everyone were admitted to engineering on a combination of their AP Calc grades and score, youād find a great correlation if they varied in a wide range. If everyone admitted will have As and 5s, shocker, youāll find not having much difference will not make much difference.
4 Likes
700/720 were the mean SAT scores, not minimum scores. Many had scores below the mean. The 25th percentile score was ~660. That said numerous studies have been linked in this thread that have a wide score variation, which found the opposite conclusionā¦ essentially every study that compared cumulative GPA at graduation and/or graduation rate between submitters and non-submitters. For example, Ithaca what most in this forum would consider a highly selective college. At the time of the study, the 25/75 SAT range was ~550/650. The previously linked study found.
Test Optional Cumulative GPA at Ithaca ā 2.83
Test Submitter Cumulative GPA at Ithaca ā 2.86Test Optional Graduation Rate ā 63.3%
Test Submitter Graduation Rate ā 63.9%
The Bates study includes the scores of optional and submitter. It found:
Test Submitters ā SAT ~= 1280, College GPA = 3.16, Graduation Rate = 89%
Non-Submitters ā SAT ~= 1075, College GPA = 3.13, Graduation Rate = 88%
The NACAC analysis also lists scores and includes larger colleges. Taking the sum of all the ~20 reviewed colleges, it found:
Submitters: HS GPA = 3.64, SAT = 1234, 5-year Graduation Rate = 82.0%
Non-submitters: HS GPA = 3.63, SAT = 1108, 5-year Graduation Rate = 82.8%
The studies that that found a more notable score contribution generally focused on first year GPA and often looked at scores in isolation, rather than in conjunction with a measure of course rigor and/or strength of HS schedule.
Coming into testing season (PSAT and fall SAT), I am increasingly aware that time management is a major scoring factor on these exams. The questions on both sections of the exam are clearly designed so that some can be answered in about 30 seconds or less and others will require a few minutes to thoroughly parse.
The outcome of this is that test takers can not simply take a section of 45 questions, divide by 60 minutes, and assume that if they have completed 22 questions and have 35 minutes left that they are āon trackā. They might be way ahead or a bit behind depending on when the time-consuming questions occur.
Further, Iāve noticed that the most time consuming questions tend to be in the final third of the test, though not at the very end. The result is that a test taker almost has to budged disproportionately for the latter part of the test or risk not completing the final shorter questions.
The test writers are clearly evaluating time management. Given that about 1/3 to 1/2 of the students in our neighboring town have testing accomodations allowing for extra time and qualifying for accomodations correlates much more strongly with wealth than GPA, athletics or any other metric out there, is there some possibility that accomodations are significantly increasing the wealth / SAT correlation that is most troubling to those who oppose testing?
2 Likes
I find the testing accommodations thing to be a bit of a racket for rich kids. Iām not saying that there arenāt legitimate LDs or that all affluent kids are taking advantage of a loop hole, but in my own limited circle, I saw a lot of kids (and parents) game the system. To take it a (cynical) step further, I think high schools (especially privates) are motivated to have as many kids qualify for accommodations as possible since it raises the average SAT/ACT scores for their school. There is no question that students with accommodations have an advantage on the tests and in the classroom, as students who get extra time on class tests and can often use that to their advantage in terms of studying (or returning to complete an unfinished test after seeing what was on the test).
I think colleges should either be made aware of who has accommodations, or simply give everyone more time. At the very least, I think high schools should have to report the percent of students who receive accommodations as part of their School Profile to Colleges so that colleges know if they stand out as āabusers of the systemā.
As for whether they are raising the wealth/SAT correlation, I think itās quite possible. I also think that kids coming out of strong high schools are better prepared for the test overall and have the resources and time to study. That said, I think colleges set the bar higher for kids who have these advantages.
5 Likes
Letās be honest. End of the day, certain schools want a ācertain type of studentā and will make the rules of the game work so they get those students without getting sued.
It is well known that kids from wealthier families are far more likely to get time accommodations for SATs. While it is, in part, because everything related to physical and mental wellbeing is too expensive for low income families, especially mental wellbeing, so wealthy kids who need accommodations are far more likely to receive them. However, in schools serving wealthy families, up to 1/4 of the students are getting accommodations. It is HIGHLY unlikely that 25% of the kids who grew up with the best diet, healthcare, and childcare will be dealing with learning disabilities. That means that a good number are either being diagnosed as having LD based on the flimsiest of evidence, and/or that there are professionals who will provide a learning disability diagnosis on demand.
So yes, it is more than a ābitā of a racket for rich kids. It is a serious racket for rich kids.
5 Likes