71–80 of 451 posts
They told me that if your kid has a 4.0 GPA (a perfect GPA), do not take the SAT/ACT ... because, without those standardized test scores - your kid is "perfect".
But unless your kid has a perfect 4.0 GPA, taking the SAT/ACT typically helps your child get into a better school than their GPA might indicate.
YMMV
Here's the money quote: >> Importantly, these test scores better position Admissions to identify high-achieving less-advantaged applicants and high-achieving applicants who attend high schools for which Dartmouth has less information to interpret the transcripts. Precisely. SATs aren't about hurting the poor or disadvantaged. It's about giving them a chance.
IMO the thing that tests do not sample is persistance. Missing that degree of freedom misses important mind-sets to achieving goals. One of the most interesting studies, not specifically focused on "success" / GPA, but on "life-satisfaction" comes from a longitudinal study "Project Talent"[1]. The key correlation to "life-satisfaction"/happiness is how many questions a person _attempts_ to answer on an otherwise un-f…
Here's the money quote: >> Importantly, these test scores better position Admissions to identify high-achieving less-advantaged applicants and high-achieving applicants who attend high schools for which Dartmouth has less information to interpret the transcripts. Precisely. SATs aren't about hurting the poor or disadvantaged. It's about giving them a chance.
Then they should be publicly run and free to take and send to universities.
If this was a paper submitted by one of my students, they would fail. Why would these faculty members sign their names to this drivel? If you read carefully, you see how useless SAT and ACT scores really are. Here are just a few insane things. > Figure 1 shows a linear relationship between SAT scores and cumulative first-year GPA for Dartmouth students. SAT scores are grouped into 16 equal-sized bins, and we plot mea…
Earlier quoted context omitted.
The fact that SAT scores are directly proportional to how much time, effort and money you have spent in preparation should tell you that it is very different from a general IQ test.
Do we know that general IQ tests are not also something that a person can improve on with practice?
Here's the money quote: >> Importantly, these test scores better position Admissions to identify high-achieving less-advantaged applicants and high-achieving applicants who attend high schools for which Dartmouth has less information to interpret the transcripts. Precisely. SATs aren't about hurting the poor or disadvantaged. It's about giving them a chance.
Obviously this is a proxy for socioeconomic status, and people higher on that ladder will have better support structures to do better in school.
This is begging the question as socioeconomic status is the primary indicator of educational attainment in our country today.
They aren't measuring potential, but the persons ability and advantages. Just as buying more TVs won't improve your students performance.
Here's the money quote: >> Importantly, these test scores better position Admissions to identify high-achieving less-advantaged applicants and high-achieving applicants who attend high schools for which Dartmouth has less information to interpret the transcripts. Precisely. SATs aren't about hurting the poor or disadvantaged. It's about giving them a chance.
IMO the thing that tests do not sample is persistance. Missing that degree of freedom misses important mind-sets to achieving goals. One of the most interesting studies, not specifically focused on "success" / GPA, but on "life-satisfaction" comes from a longitudinal study "Project Talent"[1]. The key correlation to "life-satisfaction"/happiness is how many questions a person _attempts_ to answer on an otherwise un-f…
Earlier quoted context omitted.
There is a lot of support for students who can’t afford to take the test. What isn’t really included is test prep. China has the gaokao, which everyone can take, but the prep factor is real even if they tried to outlaw it.
But your local library probably has test prep books. I know things have gotten more competitive since back in the day but you can probably get 90% of the way there without expensive test prep courses. (Yes, it takes time and commitment but it's mostly not the case that paying for test prep courses automatically translates into higher scores.)
Earlier quoted context omitted.
If the universities were sincere about that, they would figure out how to increase enrollment. Why is every parameter of admissions open to change despite tradition, except that one?
Dartmouth in 1950 was an all mens college that enrolled less than 2,500. Women weren't allowed until the 1970s. Dartmouth today has 6,700 students of all races, genders, religions. Enrollment numbers can and do change, but it is much harder and slower to change something that is physically and resource constrained than it is to change policy.
If this was a paper submitted by one of my students, they would fail. Why would these faculty members sign their names to this drivel? If you read carefully, you see how useless SAT and ACT scores really are. Here are just a few insane things. > Figure 1 shows a linear relationship between SAT scores and cumulative first-year GPA for Dartmouth students. SAT scores are grouped into 16 equal-sized bins, and we plot mea…
You are putting words in their mouth from that quote. They didn't use that Figure to claim the test scores are predictive of academic achievement. They specifically say it is evidence of a linear relationship. They argue in the next paragraph that they are and use regression analysis to make their case.
> So we're explaining less than 1/5th of the variance. This makes SAT scores basically worthless, this account for almost none of the real world performance of people, never mind how redundant they are with GPAs.
Compare that to high school GPA, which explains ~9%. The test optional movement implicitly puts more weight on high school grades by putting less weight on test scores.
I'm not saying their analysis is perfect, but your arguments aren't that strong...
At the end of the day, the analysis focuses on a subset of people who are going to Dartmouth. This is a highly filtered subsample of the total population. If you took a random sampling of 18 year olds, gave them the SAT, and then enrolled them at Dartmouth, then I hypothesize you would see a significantly stronger relationship between test scores and college grades than what is observed in their data.