"I am not a good test taker" as an Excuse for Low SAT/ACT Scores

Parent Cafe
101

I tend to disagree a lower SAT kid will, as some rule, present less well in class. In an environment where candidates are judged (for admits) on more than their tests, there are other qualities to be gleaned that can serve as well or better. The complexity is that the kid is asked to make that fine presentation in the app- and many, at 17, can’t (and no matter the stats.) That’s not enough reason to shift more emphasis to std test scores.

The old saying is that the doctor last in his class is still titled doctor, once he passes muster. Your accountant (or nurse) may not, in fact, be the highest scorer.

But, we will continue to disagree on hairs about the tests.

102

Not too sure that such is applicable here, LF, since anyone who has a strong enough MCAT to get into med school is, by definition, a strong test taker. Even the lowest MCAT scorers likely did very well on the SAT/ACT – in general. (Yes, there will be all of those anecdotes of 1500/25 kids who go to a community college, transfer and score a 30, but I’m betting big time internet dollars that such kids are a handful.

103

You don’t have to be a rocket scientist/Einstein to complete med school. But you’ve passed organic chemistry and a bunch of other “test heavy” courses, so I’m going to bet on the “above the mean” being a pretty common denominator for physicians.

Your analogy is a poor one. Test taking is a core competence of a newly minted doctor. Whether that is predictive of becoming a world class physician is another story- but studying and doing well on tests- de minimus requirement.

104

I think you do need an answer, because you don’t seem to understand what “success” is. Making money is success, the more the better. Building wealth is success. Getting married and staying married is success. Staying out of jail, staying off of illegal drugs, not having a child out of wedlock, finishing high school, finishing college, finishing grad school, etc., is success. Any positive achievement and/or avoidance of negative outcomes is success in life, and usually it can be measured quantitatively, and these measures of success tend to rise with IQ in an ALMOST linear fashion. The ACT and SAT correlate so strongly with various IQ tests that they serve as pretty good IQ tests themselves.

You are correct, but the tests do not need to be similar. Test scores predict how likely one is to finish college (and how well one will do while there), how likely one is to stay out of jail, how likely one is to have an above-average income, etc. These are all “tests” of life which test scores can predict.

It seems to me that many people just can’t grasp the notions of “on average” or “more likely to.” These words mean exactly what they say. A person with a 32 on the ACT is more likely to finish college than someone with a 27. They are more likely to and on average will have higher incomes. But there is absolutely no way to compare these two people and say with any degree of certainty that the person with the 32 ACT will be more successful than the one with a 27. Factors such as hard work, moral choices, and dedication simply play too large a role. But give me 1,000 random people with a 32 ACT versus 1,000 with a 27, and I will gladly bet the farm that the 32 group will be more successful than the 27 group by almost any measure imaginable.

The greater difference is IQ or test scores, the more likely that the higher IQ person will be more successful. In the overwhelming majority of cases, a person with a 32 on the ACT is going to be more successful by almost every measure than a person with a 12. This ought to be painfully obvious, and I don’t think many people would be willing to challenge this assumption. But if we know and accept that on average people with a 32 on the ACT are substantially more liley (almost certain) to have better life outcomes than those who make a 12, then why is it so hard to grasp that on average a person with a 32 on the ACT is going to have a slightly better chance at success than a person with a 30?

Ultra-liberal Slate featured an essay on IQ and the SAT entitled, “Yes, it really matters.” The link is below:

http://www.slate.com/articles/health_and_science/science/2014/04/what_do_sat_and_iq_tests_measure_general_intelligence_predicts_school_and.html

105

Hoo-boy. I am aware of holistic considerations in med school admissions. And the amount of in-school help to reach the final goal.

“Making money is success, the more the better.” Very narrow. In discussions about Yale’s search for leaders, even they define leadership (as a future goal) as more than financial success or power/influence over large groups. It’s simply not that sort of hierarchical. Not always that sort of self-serving or “me and mine.”

Since I’m involved with a college, I can see value in high scores, for a segment of high performing kids wanting admission. You want a willingness to push themselves in various ways, to value the conventional measures, the hoops. Great. “Fit and thrive” includes some conformity (go to class, contribute, grow, etc.) But there is so much more to success than gpa or dollars. Or stats.

I’m going to try to retreat from this thread, if I can. I can be very hard line about certain college matters, but am a bit idealistic about what makes for success. On one hand, face it: No, you can’t to this or that without passing relevant tests (original question here.) Otoh, more is required and there is more potential, in life

106

Aggregate level of achievement for large groups is rather different from assuming that it is necessarily the be-all end-all for an individual. Large 32 versus 30 groups will differ on average, but it becomes much less certain if you are comparing one individual with a 32 to another individual with a 30. Individual overperformance or underperformance on the ACT or other tests compared to performance elsewhere is not all that rare (particularly when test-specific preparation or lack thereof is thrown in), but such overperformance and underperformance washes out in large groups that are not otherwise selected for characteristics associated with such overperformance or underperformance.

107

Why can’t the students with the 12 ACT and 550 verbal SAT just study and improve their scores?

108

While they have a non-zero correlation, that correlation is often not large, particularly when filtering for other application criteria. For example, the DARCU study of 210,000 students at 356 colleges found SAT + HS GPA explained 3.3% more of the variation in 6-year graduation rate than HS GPA alone. The Geiser studies of 80,000 students on UC campuses found HS GPA + SES + SAT I explained 4.3% more of the variation in college GPA than HS GPA alone. Among students at Berkeley, the graduation prediction rate was best if SAT I M and SAT I V had small, negative regression coefficients. That is among students with similar GPA, SES, SAT W, …; the lower the SAT I M+V, the higher the predicted graduation rate.

The studies above only considered HS GPA, not HS course rigor. Studies that filer for both HS GPA and a measure of HS course rigor usually find SAT adds an insignificant addition to the prediction. For example, the CUNY mismatched college selectivity study found:

“The ATT indicates that there could be a small SAT effect on graduation (2.2 percentage points for a standard deviation increase in college SAT), but this does not reach statistical significance. The ATU is much smaller in magnitude and is not significantly different from zero.”

The studies among submitters and non-submitters at test optional colleges (non-submitters have lower average scores) generally find no notable difference in college outcome or life outcome, with the exception of submitters being more likely to enter fields that where high tests scores are especially important, such as med school.

109

With all due respect, LF, they are only holistic after one passes the mcat threshold. In a survey of med school deans, the #1 criteria for scoring an interview was mcat score (followed closely by GPA). And since ya gotta be interviewed prior to receiving an offer, that test becomes a mighty high threshold.

110

There’s a lady, Stacy Howe-Lott, who is somewhat famous for increasing her math score from a 480 to 700. She’s now an effective tutor with a good record of success helping other students.

http://www.stellarscores.com/about-me/

This goes to show that with enough effort (and money), any reasonably intelligent person can do well on the standardized tests.

111

@lookingforward I never said making money was the only measure of success, and agree that would be a very narrow definition. I gave half a dozen examples of “success,” and then said that any positive life outcome was success. I also said that usually, regardless of the measure, IQ and others tests show a strong positive correlation with success.

@ucbalumnus You are saying exactly what I’m saying. Individuals, through effort, can overcome the disadvantage of low IQ. But as a sheer numbers game, the higher the IQ or test scores, the better. The important thing is that the scores are not without meaning, as some have claimed.

112

I am a great test taker (NMF back in the 80s) and my oldest is too. Before D took PSAT/SAT/ACT I probably would have agreed with most of you about the validity and value of those tests. But for whatever reason, D is just bad at them. I’m thinking it’s directly related to reading speed (possibly an undiagnosed learning difference?) She tested into gifted (top 2%) back when she was seven (and you guys argue that IQ doesn’t change) and she has exhibited more measurable brilliance in her work than either S or I ever have, but her SAT/ACT doesn’t seem to budge above 85th-90th percentile. She’s doing an IB diploma and does really well on those exams and assignments. Most of the schools on her list happen to be test optional and will be excellent fits for her. I’m glad there are schools that acknowledge that there is more to a student than their test scores.

113

For those who believe test scores reflect real world outcomes, how do you compare students who test without extra time compared to those students who test with extra time? Do you believe speed is indicative of long term success? If a student scores avg with regular time, but near perfect with time and a half, does that mean he will be less successful in academia and career than the perfect scoring student with regular time? If that is the case, why is extra time not disclosed? Should it be disclosed, especially if post-college employers want to know those scores?

Bc…that really begs the question of how many students’ scores reflect time issues vs performance issues and does that matter?

114

The Magnetron measure of success:

Number of people who you love and they genuinely love you + Your job satisfaction (1-10 scale) + number of volunteer hours/(10*week) you spend in your community. My guess is it does not correlate all that well with SAT scores. If you are in the single digits, you need to reevaluate your life.

I’m wondering for kids who study a bunch and raise their perceived “IQ” by 20 points, if there is a measureable difference in outcomes between them and kids who took it cold in one sitting.

115

@EarlVanDorn Being the parents of an adult who struggles as an adult but has a high IQ, my husband and I are very aware that people are far too complex to classify based on a single measure. Focus, self-control, internal drive, energy…it takes a multi-dimensional evaluation to be a good predictor of life outcomes.

116

No, what I am saying is that a given test score may overestimate or underestimate what it is attempting to measure for an individual, even if it may show some correlation to future performance over a large group.

An IQ test attempts to measure IQ, but there are other factors that it (perhaps unintentionally and unavoidably) measures that mean that a given IQ test result may not be the be all end all for an individual.

117

@PNWedwonk

If those test-optional schools are her favorites, then that’s wonderful --but don’t drop a target college from the list merely because her tests are below median for those schools. Most of the top schools have holistic admission practices. My D got into colleges where her test scores were in lower quartile, and in the years since I have posted on line and PM’d many parents and kids who were told on CC that their test scores were too low for particular schools that I thought their chances were good… and in the end I received many reports back of success. The test scores are usually the least important piece information to the college, after consideration of the big factors – grades, course rigor, interests & accomplishments in EC’s or outside of school, etc. Class rank, where reported, is often a much more important to the college. High test scores are one one way otherwise indistinguishable applicants to set themselves apart from others, but they are only one factor. Scores that are at 85th-90th percentile that are counterbalanced by high demonstrated achievement are pretty much good enough for admission anywhere. There is and will be a cutoff – but it’s way lower than 85th percentile.

118

When I see a statement like this I really have to shake my head, especially when it comes from someone advocating for the value of tests. There is a huge statistical difference between a score of 32 and 12 – and virtually no statistical significance between a score of 32 and 30. I think that anyone who had taken an introductory stats would see the problem.

But I’ll leave it to @Data10 to explain this to anyone who does’t get it. :wink:

119

We know that, between two high school seniors with 3.5 GPAs intending to go into non-engineering majors in college, the one with a 32 ACT will be more successful at getting more scholarship money from the University of Alabama than the one with a 30 ACT. :slight_smile:

120

@calmom - You should post #116 as its own thread. It would be “live” for years.